a_dinosaur/tmx2gba
1
0
Fork 0
mirror of https://github.com/ScrelliCopter/tmx2gba.git synced 2025-02-21 03:29:25 +11:00
Code Issues Releases Activity

Merge branch 'refs/heads/tmxlite'

Browse Source
This commit is contained in:
a dinosaur
2024-04-07 13:05:33 +10:00
parent 1db1797a27 bc930b8ca4
commit e53f988e46
156 changed files with 96734 additions and 4045 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
.github/workflows/cmake.yml vendored
View File

@@ -3,6 +3,7 @@ name: CMake
on:
push:
paths:
- ".github/workflows/cmake.yml"
- "src/**"
- "ext/**"
- "CMakeLists.txt"
@@ -22,7 +23,7 @@ jobs:
- { name: "Windows MSVC x86", os: windows-latest, artifact: windows-x86, arch: x86 }
- { name: "Windows MSVC x64", os: windows-latest, artifact: windows-x64 }
- { name: "Windows MSVC ARM64", os: windows-latest, artifact: windows-arm64, arch: amd64_arm64 }
- { name: "Ubuntu", artifact: "linux", os: ubuntu-latest }
- { name: "Ubuntu", artifact: "linux", os: ubuntu-latest, extra: "-DUSE_BUNDLED_ZSTD:BOOL=OFF -DUSE_BUNDLED_PUGIXML:BOOL=OFF" }
runs-on: ${{matrix.config.os}}
steps:
@@ -34,6 +35,11 @@ jobs:
if: ${{startsWith(matrix.config.os, 'windows')}}
with:
arch: ${{matrix.config.arch && matrix.config.arch || 'x64'}}
- uses: awalsh128/cache-apt-pkgs-action@latest
if: ${{matrix.config.artifact == 'linux'}}
with:
packages: libzstd-dev libpugixml-dev
version: 1.0
- name: Configure CMake
run: >-

41
CMakeLists.txt
View File

@@ -1,19 +1,48 @@
cmake_minimum_required(VERSION "3.15" FATAL_ERROR)
project(tmx2gba VERSION "0.3")
project(tmx2gba
VERSION "0.7"
DESCRIPTION "Simple CLI utility for converting Tiled (.tmx) maps to GBA-friendly charmaps."
HOMEPAGE_URL "https://github.com/ScrelliCopter/tmx2gba")
# Options
option(TMX2GBA_DKP_INSTALL "Install into DEVKITPRO prefix" OFF)
option(ASAN "Enable address sanitiser" OFF)
option(USE_ZLIB "Use zlib instead of bundled miniz" "${UNIX}")
option(USE_BUNDLED_PUGIXML "Use bundled PUGIXML" ON)
option(USE_BUNDLED_ZSTD "Use bundled libzstd" ON)
option(USE_BUNDLED_TMXLITE "Use bundled tmxlite" ON)
if (ASAN)
option(TMX2GBA_DKP_INSTALL "Install into DEVKITPRO prefix" OFF)
option(ENABLE_ASAN "Enable address sanitiser" OFF)
if (ENABLE_ASAN)
add_compile_options(-fsanitize=address -fno-omit-frame-pointer)
add_link_options(-fsanitize=address -shared-libasan)
endif()
list(APPEND CMAKE_MODULE_PATH "${CMAKE_SOURCE_DIR}/cmake/modules")
# Libraries
if (USE_BUNDLED_PUGIXML)
add_subdirectory(ext/pugixml)
else()
find_package(pugixml REQUIRED CONFIG)
endif()
if (USE_ZLIB)
find_package(ZLIB REQUIRED)
else()
add_subdirectory(ext/miniz)
endif()
if (USE_BUNDLED_ZSTD)
add_subdirectory(ext/zstd)
else()
find_package(Zstd REQUIRED)
endif()
add_subdirectory(ext/base64)
add_subdirectory(ext/miniz)
add_subdirectory(ext/rapidxml)
add_subdirectory(ext/tmxlite)
# Main tmx2gba sources
add_subdirectory(src)

85
README.md
View File

@@ -1,43 +1,39 @@
# tmx2gba #
tmx2gba is a simple command line utility that converts [Tiled](http://www.mapeditor.org/) .tmx maps to Game Boy Advance formatted charmaps.
Originally developed for my own personal use, I've thrown it up in case this is of use to anyone else.
If you find a bug, please open an issue.
Enjoy!
### Features ###
* Exports to raw binary that can be easily memcpy'd into VRAM.
* Export raw charmaps that can be easily memcpy'd into VRAM.
* Preserves tile flipping.
* Supports per-tile palette specification.
* Custom collision layer support.
* Support for objects with id mapping.
### How do I use it? ###
## Usage ##
```
tmx2gba [-h] [-r offset] [-lyc name] [-p 0-15] <-i inpath> <-o outpath>
tmx2gba [-hv] [-r offset] [-lyc name] [-p 0-15] [-m name;id] <-i inpath> <-o outpath>
```
| Command | Required | Notes |
|--------------|----------|-----------------------------------------------------------------------|
| -h | N/A | Display help & command info. |
| -v | No | Display version & quit. |
| -l (name) | No | Name of layer to use (default first layer in TMX). |
| -y (name) | No | Layer for palette mappings. |
| -c (name) | No | Output a separate 8bit collision map of the specified layer. |
| -r (offset) | No | Offset tile indices (default 0). |
| -p (0-15) | No | Select which palette to use for 4-bit tilesets. |
| -m (name;id) | No | Map an object name to an ID, will enable object exports. |
| -i (path) | *Yes* | Path to input TMX file. |
| -o (path) | *Yes* | Path to output files. |
| -f <file> | No | Command line instructions list for easy integration with buildscripts |
| Command | Required | Notes |
|--------------|----------|------------------------------------------------------------------------------------|
| -h | N/A | Display help & command info |
| -v | No | Display version & quit |
| -l (name) | No | Name of layer to use (default first layer in TMX) |
| -y (name) | No | Layer for palette mappings |
| -c (name) | No | Output a separate 8bit collision map of the specified layer |
| -r (offset) | No | Offset tile indices (default 0) |
| -p (0-15) | No | Select which palette to use for 4-bit tilesets |
| -m (name;id) | No | Map an object name to an ID, will enable object exports |
| -i (path) | *Yes* | Path to input TMX file |
| -o (path) | *Yes* | Path to output files |
| -f <file> | No | Flag file containing command-line arguments for easy integration with buildscripts |
### How do I build it? ###
## Building ##
Dependencies for building are CMake 3.x and a C++11 compliant compiler,
Dependencies for building are CMake 3.15 and a C++20 compliant compiler,
all other dependencies are in-tree so you should be able to build with:
```bash
cmake -B build -DCMAKE_BUILD_TYPE=RelWithDebInfo
make -C build -j$(nproc --all)
cmake --build build
```
Optionally, you may install it to use it system wide:
@@ -45,10 +41,8 @@ Optionally, you may install it to use it system wide:
sudo cmake --install build
```
Which will copy the tmx2gba executable to /usr/local/bin/tmx2gba by default,
if you prefer to use /usr for some reason you may specify a prefix like so:
```bash
sudo cmake --install build --prefix /usr
```
`--prefix /usr` can be used to override install location.
If you're a devkitPro user and would prefer to keep all your development tools compartmentalised
you may optionally install to the tools directory with the `TMX2GBA_DKP_INSTALL` option (OFF by default).
The build scripts will respect your `DEVKITPRO` environment variable but if not set will install to
@@ -63,32 +57,11 @@ sudo cmake --install build
* Add support for multi-SBB prepared charmaps.
* Check if this works for NDS as well.
* Compression support.
* Support for less common TMX formats.
### License ###
[tmx2gba](https://github.com/ScrelliCopter/tmx2gba) is licensed under the zlib license.
[RapidXML](http://rapidxml.sourceforge.net/) is licensed under the Boost & MIT licenses.
[René Nyffenegger's base64.cpp](https://github.com/ReneNyffenegger/cpp-base64) is licensed under the zlib license.
[miniz](https://github.com/richgel999/miniz) is public domain software.
[ultragetopt](https://github.com/kevinoid/ultragetopt) is licensed under the MIT license.
```
Copyright (C) 2015-2023 a dinosaur
This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
```
## License ##
[tmx2gba](https://github.com/ScrelliCopter/tmx2gba) is licensed under the [Zlib license](COPYING.txt).
- A modified [tmxlite](https://github.com/fallahn/tmxlite) is licensed under the [Zlib license](ext/tmxlite/LICENSE).
- [pugixml](https://pugixml.org/) is licensed under the [MIT license](ext/pugixml/LICENSE.md).
- [René Nyffenegger's base64.cpp](https://github.com/ReneNyffenegger/cpp-base64) is licensed under the [Zlib license](ext/base64/LICENSE).
- [miniz](https://github.com/richgel999/miniz) is licensed under the [MIT license](ext/miniz/LICENSE).
- [ZStandard](https://facebook.github.io/zstd/) is licensed under the [BSD 3-clause license](ext/zstd/LICENSE).

171
cmake/modules/FindZstd.cmake Normal file
View File

@@ -0,0 +1,171 @@
# SPDX-License-Identifier: Zlib
# SPDX-FileCopyrightText: 2024 a dinosaur
#[=======================================================================[.rst:
FindZstd
--------
Find the Facebook Zstd library.
Imported Targets
^^^^^^^^^^^^^^^^
.. variable:: Zstd::Zstd
:prop_tgt:`IMPORTED` target for using Zstd, if Zstd is found.
Result Variables
^^^^^^^^^^^^^^^^
This module defines the following variables:
.. variable:: Zstd_FOUND
True if Zstd was found.
.. variable:: Zstd_INCLUDE_DIRS
Path to the directory containing the Zstd headers (zstd.h, etc.)
.. variable:: Zstd_LIBRARIES
Location of the Zstd library.
.. variable:: Zstd_VERSION
The version of Zstd found.
Legacy Variables
^^^^^^^^^^^^^^^^
The following variables are defined by the official Zstd CMakeLists.txt:
.. variable:: zstd_VERSION_MAJOR
The major version of Zstd.
.. variable:: zstd_VERSION_MINOR
The minor version of Zstd.
.. variable:: zstd_VERSION_PATCH
The patch/release version of Zstd.
The following variables are provided for compatibility with old find modules:
.. variable:: ZSTD_INCLUDE_DIR
Directory containing the Zstd header. (use ``Zstd_INCLUDE_DIRS`` instead)
.. variable:: ZSTD_LIBRARY
The Zstd library. (use ``Zstd_LIBRARIES`` instead)
Hints
^^^^^
.. variable:: Zstd_PREFER_STATIC_LIBS
Set to ``ON`` to prefer static libraries. Defaults to ``OFF``
Cache Variables
^^^^^^^^^^^^^^^
The following cache variables may also be set,
these are transitory and should not be relied upon:
.. variable:: Zstd_INCLUDE_DIR
Directory containing the Zstd header.
.. variable:: Zstd_LIBRARY_DEBUG
The Zstd debug library if found.
.. variable:: Zstd_LIBRARY_RELEASE
The Zstd release library if found.
.. variable:: Zstd_LIBRARY
The Zstd library.
#]=======================================================================]
#TODO: define Zstd::static & Zstd::shared and alias Zstd::Zstd based on preference
find_path(Zstd_INCLUDE_DIR NAMES zstd.h)
mark_as_advanced(Zstd_INCLUDE_DIR)
if (Zstd_PREFER_STATIC_LIBS)
find_library(Zstd_LIBRARY_DEBUG NAMES zstd_staticd zstdd)
find_library(Zstd_LIBRARY_RELEASE NAMES zstd_static zstd)
else()
find_library(Zstd_LIBRARY_DEBUG NAMES zstdd zstd_staticd)
find_library(Zstd_LIBRARY_RELEASE NAMES zstd zstd_static)
endif()
include(SelectLibraryConfigurations)
select_library_configurations(Zstd)
mark_as_advanced(Zstd_LIBRARY Zstd_LIBRARY_DEBUG Zstd_LIBRARY_RELEASE)
if (Zstd_INCLUDE_DIR AND EXISTS "${Zstd_INCLUDE_DIR}/zstd.h")
function (_zstd_read_define _variable _define)
set(_file "${Zstd_INCLUDE_DIR}/zstd.h")
set(_regex "#define[ \t]+${_define}[ \t]+([0-9]+)")
file(STRINGS "${_file}" _line LIMIT_COUNT 1 REGEX "${_regex}")
if (CMAKE_VERSION VERSION_LESS "3.29")
string(REGEX MATCH "${_regex}" _line "${_line}")
endif()
set(${_variable} ${CMAKE_MATCH_1} PARENT_SCOPE)
endfunction()
_zstd_read_define(zstd_VERSION_MAJOR "ZSTD_VERSION_MAJOR")
_zstd_read_define(zstd_VERSION_MINOR "ZSTD_VERSION_MINOR")
_zstd_read_define(zstd_VERSION_PATCH "ZSTD_VERSION_RELEASE")
set(Zstd_VERSION "${zstd_VERSION_MAJOR}.${zstd_VERSION_MINOR}.${zstd_VERSION_PATCH}")
mark_as_advanced(zstd_VERSION_MAJOR zstd_VERSION_MINOR zstd_VERSION_PATCH Zstd_VERSION)
endif()
include(FindPackageHandleStandardArgs)
find_package_handle_standard_args(Zstd
REQUIRED_VARS Zstd_LIBRARY Zstd_INCLUDE_DIR
VERSION_VAR Zstd_VERSION)
mark_as_advanced(Zstd_FOUND)
if (Zstd_FOUND)
set(Zstd_INCLUDE_DIRS ${Zstd_INCLUDE_DIR})
set(Zstd_LIBRARIES ${Zstd_LIBRARY})
# Legacy variables
set(ZSTD_INCLUDE_DIR ${Zstd_INCLUDE_DIR})
set(ZSTD_LIBRARY ${Zstd_LIBRARY})
mark_as_advanced(ZSTD_INCLUDE_DIR ZSTD_LIBRARY)
if (NOT TARGET Zstd::Zstd)
add_library(Zstd::Zstd UNKNOWN IMPORTED)
set_property(TARGET Zstd::Zstd PROPERTY
INTERFACE_INCLUDE_DIRECTORIES "${Zstd_INCLUDE_DIR}")
endif()
if (NOT Zstd_LIBRARY_DEBUG AND NOT Zstd_LIBRARY_RELEASE)
set_property(TARGET Zstd::Zstd PROPERTY
IMPORTED_LOCATION "${Zstd_LIBRARY}")
endif()
if (Zstd_LIBRARY_DEBUG)
set_property(TARGET Zstd::Zstd APPEND PROPERTY
IMPORTED_CONFIGURATIONS DEBUG)
set_property(TARGET Zstd::Zstd PROPERTY
IMPORTED_LOCATION_DEBUG "${Zstd_LIBRARY_DEBUG}")
endif()
if (Zstd_LIBRARY_RELEASE)
set_property(TARGET Zstd::Zstd APPEND PROPERTY
IMPORTED_CONFIGURATIONS RELEASE)
set_property(TARGET Zstd::Zstd PROPERTY
IMPORTED_LOCATION_RELEASE "${Zstd_LIBRARY_RELEASE}")
endif()
endif()

2
ext/base64/CMakeLists.txt
View File

@@ -1,5 +1,5 @@
add_library(base64
base64.cpp base64.h)
add_library(External::base64 ALIAS base64)
add_library(base64::base64 ALIAS base64)
target_include_directories(base64
PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})

19
ext/base64/LICENSE Normal file
View File

@@ -0,0 +1,19 @@
Copyright © 2004-2017 by René Nyffenegger
This source code is provided 'as-is', without any express or implied
warranty. In no event will the author be held liable for any damages
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this source code must not be misrepresented; you must not
claim that you wrote the original source code. If you use this source code
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original source code.
3. This notice may not be removed or altered from any source distribution.

61
ext/base64/base64.cpp
View File

@@ -5,9 +5,9 @@
More information at
https://renenyffenegger.ch/notes/development/Base64/Encoding-and-decoding-base-64-with-cpp
Version: 2.rc.08 (release candidate)
Version: 2.rc.09 (release candidate)
Copyright (C) 2004-2017, 2020, 2021 René Nyffenegger
Copyright (C) 2004-2017, 2020-2022 René Nyffenegger
This source code is provided 'as-is', without any express or implied
warranty. In no event will the author be held liable for any damages
@@ -34,14 +34,19 @@
#include "base64.h"
#include <algorithm>
#if defined(__cpp_exceptions)
#include <stdexcept>
#else
#include <cassert>
#endif
//
// Depending on the url parameter in base64_chars, one of
// two sets of base64 characters needs to be chosen.
// They differ in their last two characters.
//
static const char* base64_chars[2] = {
static const char* to_base64_chars[2] = {
"ABCDEFGHIJKLMNOPQRSTUVWXYZ"
"abcdefghijklmnopqrstuvwxyz"
"0123456789"
@@ -52,22 +57,40 @@ static const char* base64_chars[2] = {
"0123456789"
"-_"};
static const unsigned char from_base64_chars[256] = {
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 62, 64, 62, 64, 63,
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 64, 64, 64, 64, 64, 64,
64, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 64, 64, 64, 64, 63,
64, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64,
64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64
};
static unsigned int pos_of_char(const unsigned char chr) {
//
// Return the position of chr within base64_encode()
//
if (chr >= 'A' && chr <= 'Z') return chr - 'A';
else if (chr >= 'a' && chr <= 'z') return chr - 'a' + ('Z' - 'A') + 1;
else if (chr >= '0' && chr <= '9') return chr - '0' + ('Z' - 'A') + ('z' - 'a') + 2;
else if (chr == '+' || chr == '-') return 62; // Be liberal with input and accept both url ('-') and non-url ('+') base 64 characters (
else if (chr == '/' || chr == '_') return 63; // Ditto for '/' and '_'
else
if (from_base64_chars[chr] != 64) return from_base64_chars[chr];
//
// 2020-10-23: Throw std::exception rather than const char*
//(Pablo Martin-Gomez, https://github.com/Bouska)
//
#if defined(__cpp_exceptions)
throw std::runtime_error("Input is not valid base64-encoded data.");
#else
assert(!"Input is not valid base64-encoded data.");
#endif
}
static std::string insert_linebreaks(std::string str, size_t distance) {
@@ -123,7 +146,7 @@ std::string base64_encode(unsigned char const* bytes_to_encode, size_t in_len, b
// the correct character set is chosen by subscripting
// base64_chars with url.
//
const char* base64_chars_ = base64_chars[url];
const char* base64_chars_ = to_base64_chars[url];
std::string ret;
ret.reserve(len_encoded);
@@ -160,7 +183,7 @@ std::string base64_encode(unsigned char const* bytes_to_encode, size_t in_len, b
}
template <typename String>
static std::string decode(String encoded_string, bool remove_linebreaks) {
static std::string decode(String const& encoded_string, bool remove_linebreaks) {
//
// decode(…) is templated so that it can be used with String = const std::string&
// or std::string_view (requires at least C++17)
@@ -204,33 +227,33 @@ static std::string decode(String encoded_string, bool remove_linebreaks) {
// The last chunk produces at least one and up to three bytes.
//
size_t pos_of_char_1 = pos_of_char(encoded_string[pos+1] );
size_t pos_of_char_1 = pos_of_char(encoded_string.at(pos+1) );
//
// Emit the first output byte that is produced in each chunk:
//
ret.push_back(static_cast<std::string::value_type>( ( (pos_of_char(encoded_string[pos+0]) ) << 2 ) + ( (pos_of_char_1 & 0x30 ) >> 4)));
ret.push_back(static_cast<std::string::value_type>( ( (pos_of_char(encoded_string.at(pos+0)) ) << 2 ) + ( (pos_of_char_1 & 0x30 ) >> 4)));
if ( ( pos + 2 < length_of_string ) && // Check for data that is not padded with equal signs (which is allowed by RFC 2045)
encoded_string[pos+2] != '=' &&
encoded_string[pos+2] != '.' // accept URL-safe base 64 strings, too, so check for '.' also.
encoded_string.at(pos+2) != '=' &&
encoded_string.at(pos+2) != '.' // accept URL-safe base 64 strings, too, so check for '.' also.
)
{
//
// Emit a chunk's second byte (which might not be produced in the last chunk).
//
unsigned int pos_of_char_2 = pos_of_char(encoded_string[pos+2] );
unsigned int pos_of_char_2 = pos_of_char(encoded_string.at(pos+2) );
ret.push_back(static_cast<std::string::value_type>( (( pos_of_char_1 & 0x0f) << 4) + (( pos_of_char_2 & 0x3c) >> 2)));
if ( ( pos + 3 < length_of_string ) &&
encoded_string[pos+3] != '=' &&
encoded_string[pos+3] != '.'
encoded_string.at(pos+3) != '=' &&
encoded_string.at(pos+3) != '.'
)
{
//
// Emit a chunk's third byte (which might not be produced in the last chunk).
//
ret.push_back(static_cast<std::string::value_type>( ( (pos_of_char_2 & 0x03 ) << 6 ) + pos_of_char(encoded_string[pos+3]) ));
ret.push_back(static_cast<std::string::value_type>( ( (pos_of_char_2 & 0x03 ) << 6 ) + pos_of_char(encoded_string.at(pos+3)) ));
}
}

2
ext/base64/base64.h
View File

@@ -1,6 +1,6 @@
//
// base64 encoding and decoding with C++.
// Version: 2.rc.08 (release candidate)
// Version: 2.rc.09 (release candidate)
//
#ifndef BASE64_H_C0CE2A47_D10E_42C9_A27C_C883944E704A

2
ext/miniz/CMakeLists.txt
View File

@@ -1,5 +1,5 @@
add_library(miniz
miniz.c miniz.h)
add_library(External::miniz ALIAS miniz)
add_library(miniz::miniz ALIAS miniz)
target_include_directories(miniz
PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})

22
ext/miniz/LICENSE Normal file
View File

@@ -0,0 +1,22 @@
Copyright 2013-2014 RAD Game Tools and Valve Software
Copyright 2010-2014 Rich Geldreich and Tenacious Software LLC
All Rights Reserved.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

334
ext/miniz/miniz.c vendored
View File

@@ -187,6 +187,8 @@ const char *mz_version(void)
#ifndef MINIZ_NO_ZLIB_APIS
#ifndef MINIZ_NO_DEFLATE_APIS
int mz_deflateInit(mz_streamp pStream, int level)
{
return mz_deflateInit2(pStream, level, MZ_DEFLATED, MZ_DEFAULT_WINDOW_BITS, 9, MZ_DEFAULT_STRATEGY);
@@ -321,7 +323,7 @@ int mz_compress2(unsigned char *pDest, mz_ulong *pDest_len, const unsigned char
memset(&stream, 0, sizeof(stream));
/* In case mz_ulong is 64-bits (argh I hate longs). */
if ((source_len | *pDest_len) > 0xFFFFFFFFU)
if ((mz_uint64)(source_len | *pDest_len) > 0xFFFFFFFFU)
return MZ_PARAM_ERROR;
stream.next_in = pSource;
@@ -354,6 +356,10 @@ mz_ulong mz_compressBound(mz_ulong source_len)
return mz_deflateBound(NULL, source_len);
}
#endif /*#ifndef MINIZ_NO_DEFLATE_APIS*/
#ifndef MINIZ_NO_INFLATE_APIS
typedef struct
{
tinfl_decompressor m_decomp;
@@ -560,7 +566,7 @@ int mz_uncompress2(unsigned char *pDest, mz_ulong *pDest_len, const unsigned cha
memset(&stream, 0, sizeof(stream));
/* In case mz_ulong is 64-bits (argh I hate longs). */
if ((*pSource_len | *pDest_len) > 0xFFFFFFFFU)
if ((mz_uint64)(*pSource_len | *pDest_len) > 0xFFFFFFFFU)
return MZ_PARAM_ERROR;
stream.next_in = pSource;
@@ -589,6 +595,8 @@ int mz_uncompress(unsigned char *pDest, mz_ulong *pDest_len, const unsigned char
return mz_uncompress2(pDest, pDest_len, pSource, &source_len);
}
#endif /*#ifndef MINIZ_NO_INFLATE_APIS*/
const char *mz_error(int err)
{
static struct
@@ -666,6 +674,8 @@ const char *mz_error(int err)
#ifndef MINIZ_NO_DEFLATE_APIS
#ifdef __cplusplus
extern "C" {
#endif
@@ -744,7 +754,7 @@ static tdefl_sym_freq *tdefl_radix_sort_syms(mz_uint num_syms, tdefl_sym_freq *p
{
mz_uint32 total_passes = 2, pass_shift, pass, i, hist[256 * 2];
tdefl_sym_freq *pCur_syms = pSyms0, *pNew_syms = pSyms1;
MZ_CLEAR_OBJ(hist);
MZ_CLEAR_ARR(hist);
for (i = 0; i < num_syms; i++)
{
mz_uint freq = pSyms0[i].m_key;
@@ -862,7 +872,7 @@ static void tdefl_optimize_huffman_table(tdefl_compressor *d, int table_num, int
{
int i, j, l, num_codes[1 + TDEFL_MAX_SUPPORTED_HUFF_CODESIZE];
mz_uint next_code[TDEFL_MAX_SUPPORTED_HUFF_CODESIZE + 1];
MZ_CLEAR_OBJ(num_codes);
MZ_CLEAR_ARR(num_codes);
if (static_table)
{
for (i = 0; i < table_len; i++)
@@ -888,8 +898,8 @@ static void tdefl_optimize_huffman_table(tdefl_compressor *d, int table_num, int
tdefl_huffman_enforce_max_code_size(num_codes, num_used_syms, code_size_limit);
MZ_CLEAR_OBJ(d->m_huff_code_sizes[table_num]);
MZ_CLEAR_OBJ(d->m_huff_codes[table_num]);
MZ_CLEAR_ARR(d->m_huff_code_sizes[table_num]);
MZ_CLEAR_ARR(d->m_huff_codes[table_num]);
for (i = 1, j = num_used_syms; i <= code_size_limit; i++)
for (l = num_codes[i]; l > 0; l--)
d->m_huff_code_sizes[table_num][pSyms[--j].m_sym_index] = (mz_uint8)(i);
@@ -975,7 +985,7 @@ static void tdefl_optimize_huffman_table(tdefl_compressor *d, int table_num, int
} \
}
static mz_uint8 s_tdefl_packed_code_size_syms_swizzle[] = { 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15 };
static const mz_uint8 s_tdefl_packed_code_size_syms_swizzle[] = { 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15 };
static void tdefl_start_dynamic_block(tdefl_compressor *d)
{
@@ -1113,7 +1123,8 @@ static mz_bool tdefl_compress_lz_codes(tdefl_compressor *d)
if (flags & 1)
{
mz_uint s0, s1, n0, n1, sym, num_extra_bits;
mz_uint match_len = pLZ_codes[0], match_dist = *(const mz_uint16 *)(pLZ_codes + 1);
mz_uint match_len = pLZ_codes[0];
mz_uint match_dist = (pLZ_codes[1] | (pLZ_codes[2] << 8));
pLZ_codes += 3;
MZ_ASSERT(d->m_huff_code_sizes[0][s_tdefl_len_sym[match_len]]);
@@ -1158,7 +1169,7 @@ static mz_bool tdefl_compress_lz_codes(tdefl_compressor *d)
if (pOutput_buf >= d->m_pOutput_buf_end)
return MZ_FALSE;
*(mz_uint64 *)pOutput_buf = bit_buffer;
memcpy(pOutput_buf, &bit_buffer, sizeof(mz_uint64));
pOutput_buf += (bits_in >> 3);
bit_buffer >>= (bits_in & ~7);
bits_in &= 7;
@@ -1240,6 +1251,8 @@ static mz_bool tdefl_compress_block(tdefl_compressor *d, mz_bool static_block)
return tdefl_compress_lz_codes(d);
}
static const mz_uint s_tdefl_num_probes[11];
static int tdefl_flush_block(tdefl_compressor *d, int flush)
{
mz_uint saved_bit_buf, saved_bits_in;
@@ -1260,8 +1273,27 @@ static int tdefl_flush_block(tdefl_compressor *d, int flush)
if ((d->m_flags & TDEFL_WRITE_ZLIB_HEADER) && (!d->m_block_index))
{
TDEFL_PUT_BITS(0x78, 8);
TDEFL_PUT_BITS(0x01, 8);
const mz_uint8 cmf = 0x78;
mz_uint8 flg, flevel = 3;
mz_uint header, i, mz_un = sizeof(s_tdefl_num_probes) / sizeof(mz_uint);
/* Determine compression level by reversing the process in tdefl_create_comp_flags_from_zip_params() */
for (i = 0; i < mz_un; i++)
if (s_tdefl_num_probes[i] == (d->m_flags & 0xFFF)) break;
if (i < 2)
flevel = 0;
else if (i < 6)
flevel = 1;
else if (i == 6)
flevel = 2;
header = cmf << 8 | (flevel << 6);
header += 31 - (header % 31);
flg = header & 0xFF;
TDEFL_PUT_BITS(cmf, 8);
TDEFL_PUT_BITS(flg, 8);
}
TDEFL_PUT_BITS(flush == TDEFL_FINISH, 1);
@@ -1732,7 +1764,7 @@ static mz_bool tdefl_compress_normal(tdefl_compressor *d)
mz_uint dst_pos = (d->m_lookahead_pos + d->m_lookahead_size) & TDEFL_LZ_DICT_SIZE_MASK, ins_pos = d->m_lookahead_pos + d->m_lookahead_size - 2;
mz_uint hash = (d->m_dict[ins_pos & TDEFL_LZ_DICT_SIZE_MASK] << TDEFL_LZ_HASH_SHIFT) ^ d->m_dict[(ins_pos + 1) & TDEFL_LZ_DICT_SIZE_MASK];
mz_uint num_bytes_to_process = (mz_uint)MZ_MIN(src_buf_left, TDEFL_MAX_MATCH_LEN - d->m_lookahead_size);
const mz_uint8 *pSrc_end = pSrc + num_bytes_to_process;
const mz_uint8 *pSrc_end = pSrc ? pSrc + num_bytes_to_process : NULL;
src_buf_left -= num_bytes_to_process;
d->m_lookahead_size += num_bytes_to_process;
while (pSrc != pSrc_end)
@@ -1942,8 +1974,8 @@ tdefl_status tdefl_compress(tdefl_compressor *d, const void *pIn_buf, size_t *pI
d->m_finished = (flush == TDEFL_FINISH);
if (flush == TDEFL_FULL_FLUSH)
{
MZ_CLEAR_OBJ(d->m_hash);
MZ_CLEAR_OBJ(d->m_next);
MZ_CLEAR_ARR(d->m_hash);
MZ_CLEAR_ARR(d->m_next);
d->m_dict_size = 0;
}
}
@@ -1966,7 +1998,7 @@ tdefl_status tdefl_init(tdefl_compressor *d, tdefl_put_buf_func_ptr pPut_buf_fun
d->m_greedy_parsing = (flags & TDEFL_GREEDY_PARSING_FLAG) != 0;
d->m_max_probes[1] = 1 + (((flags & 0xFFF) >> 2) + 2) / 3;
if (!(flags & TDEFL_NONDETERMINISTIC_PARSING_FLAG))
MZ_CLEAR_OBJ(d->m_hash);
MZ_CLEAR_ARR(d->m_hash);
d->m_lookahead_pos = d->m_lookahead_size = d->m_dict_size = d->m_total_lz_bytes = d->m_lz_code_buf_dict_pos = d->m_bits_in = 0;
d->m_output_flush_ofs = d->m_output_flush_remaining = d->m_finished = d->m_block_index = d->m_bit_buffer = d->m_wants_to_finish = 0;
d->m_pLZ_code_buf = d->m_lz_code_buf + 1;
@@ -1987,7 +2019,7 @@ tdefl_status tdefl_init(tdefl_compressor *d, tdefl_put_buf_func_ptr pPut_buf_fun
d->m_src_buf_left = 0;
d->m_out_buf_ofs = 0;
if (!(flags & TDEFL_NONDETERMINISTIC_PARSING_FLAG))
MZ_CLEAR_OBJ(d->m_dict);
MZ_CLEAR_ARR(d->m_dict);
memset(&d->m_huff_count[0][0], 0, sizeof(d->m_huff_count[0][0]) * TDEFL_MAX_HUFF_SYMBOLS_0);
memset(&d->m_huff_count[1][0], 0, sizeof(d->m_huff_count[1][0]) * TDEFL_MAX_HUFF_SYMBOLS_1);
return TDEFL_STATUS_OKAY;
@@ -2197,7 +2229,7 @@ void *tdefl_write_image_to_png_file_in_memory(const void *pImage, int w, int h,
/* Allocate the tdefl_compressor and tinfl_decompressor structures in C so that */
/* non-C language bindings to tdefL_ and tinfl_ API don't need to worry about */
/* structure size and allocation mechanism. */
tdefl_compressor *tdefl_compressor_alloc()
tdefl_compressor *tdefl_compressor_alloc(void)
{
return (tdefl_compressor *)MZ_MALLOC(sizeof(tdefl_compressor));
}
@@ -2215,6 +2247,8 @@ void tdefl_compressor_free(tdefl_compressor *pComp)
#ifdef __cplusplus
}
#endif
#endif /*#ifndef MINIZ_NO_DEFLATE_APIS*/
/**************************************************************************
*
* Copyright 2013-2014 RAD Game Tools and Valve Software
@@ -2243,6 +2277,8 @@ void tdefl_compressor_free(tdefl_compressor *pComp)
#ifndef MINIZ_NO_INFLATE_APIS
#ifdef __cplusplus
extern "C" {
#endif
@@ -2323,10 +2359,10 @@ extern "C" {
/* It reads just enough bytes from the input stream that are needed to decode the next Huffman code (and absolutely no more). It works by trying to fully decode a */
/* Huffman code by using whatever bits are currently present in the bit buffer. If this fails, it reads another byte, and tries again until it succeeds or until the */
/* bit buffer contains >=15 bits (deflate's max. Huffman code size). */
#define TINFL_HUFF_BITBUF_FILL(state_index, pHuff) \
#define TINFL_HUFF_BITBUF_FILL(state_index, pLookUp, pTree) \
do \
{ \
temp = (pHuff)->m_look_up[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]; \
temp = pLookUp[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]; \
if (temp >= 0) \
{ \
code_len = temp >> 9; \
@@ -2338,7 +2374,7 @@ extern "C" {
code_len = TINFL_FAST_LOOKUP_BITS; \
do \
{ \
temp = (pHuff)->m_tree[~temp + ((bit_buf >> code_len++) & 1)]; \
temp = pTree[~temp + ((bit_buf >> code_len++) & 1)]; \
} while ((temp < 0) && (num_bits >= (code_len + 1))); \
if (temp >= 0) \
break; \
@@ -2354,7 +2390,7 @@ extern "C" {
/* The slow path is only executed at the very end of the input buffer. */
/* v1.16: The original macro handled the case at the very end of the passed-in input buffer, but we also need to handle the case where the user passes in 1+zillion bytes */
/* following the deflate data and our non-conservative read-ahead path won't kick in here on this code. This is much trickier. */
#define TINFL_HUFF_DECODE(state_index, sym, pHuff) \
#define TINFL_HUFF_DECODE(state_index, sym, pLookUp, pTree) \
do \
{ \
int temp; \
@@ -2363,7 +2399,7 @@ extern "C" {
{ \
if ((pIn_buf_end - pIn_buf_cur) < 2) \
{ \
TINFL_HUFF_BITBUF_FILL(state_index, pHuff); \
TINFL_HUFF_BITBUF_FILL(state_index, pLookUp, pTree); \
} \
else \
{ \
@@ -2372,14 +2408,14 @@ extern "C" {
num_bits += 16; \
} \
} \
if ((temp = (pHuff)->m_look_up[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0) \
if ((temp = pLookUp[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0) \
code_len = temp >> 9, temp &= 511; \
else \
{ \
code_len = TINFL_FAST_LOOKUP_BITS; \
do \
{ \
temp = (pHuff)->m_tree[~temp + ((bit_buf >> code_len++) & 1)]; \
temp = pTree[~temp + ((bit_buf >> code_len++) & 1)]; \
} while (temp < 0); \
} \
sym = temp; \
@@ -2388,20 +2424,33 @@ extern "C" {
} \
MZ_MACRO_END
static void tinfl_clear_tree(tinfl_decompressor *r)
{
if (r->m_type == 0)
MZ_CLEAR_ARR(r->m_tree_0);
else if (r->m_type == 1)
MZ_CLEAR_ARR(r->m_tree_1);
else
MZ_CLEAR_ARR(r->m_tree_2);
}
tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_next, size_t *pIn_buf_size, mz_uint8 *pOut_buf_start, mz_uint8 *pOut_buf_next, size_t *pOut_buf_size, const mz_uint32 decomp_flags)
{
static const int s_length_base[31] = { 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0 };
static const int s_length_extra[31] = { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 0, 0 };
static const int s_dist_base[32] = { 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577, 0, 0 };
static const int s_dist_extra[32] = { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13 };
static const mz_uint16 s_length_base[31] = { 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0 };
static const mz_uint8 s_length_extra[31] = { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 0, 0 };
static const mz_uint16 s_dist_base[32] = { 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577, 0, 0 };
static const mz_uint8 s_dist_extra[32] = { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13 };
static const mz_uint8 s_length_dezigzag[19] = { 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15 };
static const int s_min_table_sizes[3] = { 257, 1, 4 };
static const mz_uint16 s_min_table_sizes[3] = { 257, 1, 4 };
mz_int16 *pTrees[3];
mz_uint8 *pCode_sizes[3];
tinfl_status status = TINFL_STATUS_FAILED;
mz_uint32 num_bits, dist, counter, num_extra;
tinfl_bit_buf_t bit_buf;
const mz_uint8 *pIn_buf_cur = pIn_buf_next, *const pIn_buf_end = pIn_buf_next + *pIn_buf_size;
mz_uint8 *pOut_buf_cur = pOut_buf_next, *const pOut_buf_end = pOut_buf_next + *pOut_buf_size;
mz_uint8 *pOut_buf_cur = pOut_buf_next, *const pOut_buf_end = pOut_buf_next ? pOut_buf_next + *pOut_buf_size : NULL;
size_t out_buf_size_mask = (decomp_flags & TINFL_FLAG_USING_NON_WRAPPING_OUTPUT_BUF) ? (size_t)-1 : ((pOut_buf_next - pOut_buf_start) + *pOut_buf_size) - 1, dist_from_out_buf_start;
/* Ensure the output buffer's size is a power of 2, unless the output buffer is large enough to hold the entire output file (in which case it doesn't matter). */
@@ -2411,6 +2460,13 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
return TINFL_STATUS_BAD_PARAM;
}
pTrees[0] = r->m_tree_0;
pTrees[1] = r->m_tree_1;
pTrees[2] = r->m_tree_2;
pCode_sizes[0] = r->m_code_size_0;
pCode_sizes[1] = r->m_code_size_1;
pCode_sizes[2] = r->m_code_size_2;
num_bits = r->m_num_bits;
bit_buf = r->m_bit_buf;
dist = r->m_dist;
@@ -2427,7 +2483,7 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
TINFL_GET_BYTE(2, r->m_zhdr1);
counter = (((r->m_zhdr0 * 256 + r->m_zhdr1) % 31 != 0) || (r->m_zhdr1 & 32) || ((r->m_zhdr0 & 15) != 8));
if (!(decomp_flags & TINFL_FLAG_USING_NON_WRAPPING_OUTPUT_BUF))
counter |= (((1U << (8U + (r->m_zhdr0 >> 4))) > 32768U) || ((out_buf_size_mask + 1) < (size_t)(1U << (8U + (r->m_zhdr0 >> 4)))));
counter |= (((1U << (8U + (r->m_zhdr0 >> 4))) > 32768U) || ((out_buf_size_mask + 1) < (size_t)((size_t)1 << (8U + (r->m_zhdr0 >> 4)))));
if (counter)
{
TINFL_CR_RETURN_FOREVER(36, TINFL_STATUS_FAILED);
@@ -2488,11 +2544,11 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
{
if (r->m_type == 1)
{
mz_uint8 *p = r->m_tables[0].m_code_size;
mz_uint8 *p = r->m_code_size_0;
mz_uint i;
r->m_table_sizes[0] = 288;
r->m_table_sizes[1] = 32;
TINFL_MEMSET(r->m_tables[1].m_code_size, 5, 32);
TINFL_MEMSET(r->m_code_size_1, 5, 32);
for (i = 0; i <= 143; ++i)
*p++ = 8;
for (; i <= 255; ++i)
@@ -2509,26 +2565,30 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
TINFL_GET_BITS(11, r->m_table_sizes[counter], "\05\05\04"[counter]);
r->m_table_sizes[counter] += s_min_table_sizes[counter];
}
MZ_CLEAR_OBJ(r->m_tables[2].m_code_size);
MZ_CLEAR_ARR(r->m_code_size_2);
for (counter = 0; counter < r->m_table_sizes[2]; counter++)
{
mz_uint s;
TINFL_GET_BITS(14, s, 3);
r->m_tables[2].m_code_size[s_length_dezigzag[counter]] = (mz_uint8)s;
r->m_code_size_2[s_length_dezigzag[counter]] = (mz_uint8)s;
}
r->m_table_sizes[2] = 19;
}
for (; (int)r->m_type >= 0; r->m_type--)
{
int tree_next, tree_cur;
tinfl_huff_table *pTable;
mz_int16 *pLookUp;
mz_int16 *pTree;
mz_uint8 *pCode_size;
mz_uint i, j, used_syms, total, sym_index, next_code[17], total_syms[16];
pTable = &r->m_tables[r->m_type];
MZ_CLEAR_OBJ(total_syms);
MZ_CLEAR_OBJ(pTable->m_look_up);
MZ_CLEAR_OBJ(pTable->m_tree);
pLookUp = r->m_look_up[r->m_type];
pTree = pTrees[r->m_type];
pCode_size = pCode_sizes[r->m_type];
MZ_CLEAR_ARR(total_syms);
TINFL_MEMSET(pLookUp, 0, sizeof(r->m_look_up[0]));
tinfl_clear_tree(r);
for (i = 0; i < r->m_table_sizes[r->m_type]; ++i)
total_syms[pTable->m_code_size[i]]++;
total_syms[pCode_size[i]]++;
used_syms = 0, total = 0;
next_code[0] = next_code[1] = 0;
for (i = 1; i <= 15; ++i)
@@ -2542,7 +2602,7 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
}
for (tree_next = -1, sym_index = 0; sym_index < r->m_table_sizes[r->m_type]; ++sym_index)
{
mz_uint rev_code = 0, l, cur_code, code_size = pTable->m_code_size[sym_index];
mz_uint rev_code = 0, l, cur_code, code_size = pCode_size[sym_index];
if (!code_size)
continue;
cur_code = next_code[code_size]++;
@@ -2553,14 +2613,14 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
mz_int16 k = (mz_int16)((code_size << 9) | sym_index);
while (rev_code < TINFL_FAST_LOOKUP_SIZE)
{
pTable->m_look_up[rev_code] = k;
pLookUp[rev_code] = k;
rev_code += (1 << code_size);
}
continue;
}
if (0 == (tree_cur = pTable->m_look_up[rev_code & (TINFL_FAST_LOOKUP_SIZE - 1)]))
if (0 == (tree_cur = pLookUp[rev_code & (TINFL_FAST_LOOKUP_SIZE - 1)]))
{
pTable->m_look_up[rev_code & (TINFL_FAST_LOOKUP_SIZE - 1)] = (mz_int16)tree_next;
pLookUp[rev_code & (TINFL_FAST_LOOKUP_SIZE - 1)] = (mz_int16)tree_next;
tree_cur = tree_next;
tree_next -= 2;
}
@@ -2568,24 +2628,24 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
for (j = code_size; j > (TINFL_FAST_LOOKUP_BITS + 1); j--)
{
tree_cur -= ((rev_code >>= 1) & 1);
if (!pTable->m_tree[-tree_cur - 1])
if (!pTree[-tree_cur - 1])
{
pTable->m_tree[-tree_cur - 1] = (mz_int16)tree_next;
pTree[-tree_cur - 1] = (mz_int16)tree_next;
tree_cur = tree_next;
tree_next -= 2;
}
else
tree_cur = pTable->m_tree[-tree_cur - 1];
tree_cur = pTree[-tree_cur - 1];
}
tree_cur -= ((rev_code >>= 1) & 1);
pTable->m_tree[-tree_cur - 1] = (mz_int16)sym_index;
pTree[-tree_cur - 1] = (mz_int16)sym_index;
}
if (r->m_type == 2)
{
for (counter = 0; counter < (r->m_table_sizes[0] + r->m_table_sizes[1]);)
{
mz_uint s;
TINFL_HUFF_DECODE(16, dist, &r->m_tables[2]);
TINFL_HUFF_DECODE(16, dist, r->m_look_up[2], r->m_tree_2);
if (dist < 16)
{
r->m_len_codes[counter++] = (mz_uint8)dist;
@@ -2605,8 +2665,8 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
{
TINFL_CR_RETURN_FOREVER(21, TINFL_STATUS_FAILED);
}
TINFL_MEMCPY(r->m_tables[0].m_code_size, r->m_len_codes, r->m_table_sizes[0]);
TINFL_MEMCPY(r->m_tables[1].m_code_size, r->m_len_codes + r->m_table_sizes[0], r->m_table_sizes[1]);
TINFL_MEMCPY(r->m_code_size_0, r->m_len_codes, r->m_table_sizes[0]);
TINFL_MEMCPY(r->m_code_size_1, r->m_len_codes + r->m_table_sizes[0], r->m_table_sizes[1]);
}
}
for (;;)
@@ -2616,7 +2676,7 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
{
if (((pIn_buf_end - pIn_buf_cur) < 4) || ((pOut_buf_end - pOut_buf_cur) < 2))
{
TINFL_HUFF_DECODE(23, counter, &r->m_tables[0]);
TINFL_HUFF_DECODE(23, counter, r->m_look_up[0], r->m_tree_0);
if (counter >= 256)
break;
while (pOut_buf_cur >= pOut_buf_end)
@@ -2644,14 +2704,14 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
num_bits += 16;
}
#endif
if ((sym2 = r->m_tables[0].m_look_up[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0)
if ((sym2 = r->m_look_up[0][bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0)
code_len = sym2 >> 9;
else
{
code_len = TINFL_FAST_LOOKUP_BITS;
do
{
sym2 = r->m_tables[0].m_tree[~sym2 + ((bit_buf >> code_len++) & 1)];
sym2 = r->m_tree_0[~sym2 + ((bit_buf >> code_len++) & 1)];
} while (sym2 < 0);
}
counter = sym2;
@@ -2668,14 +2728,14 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
num_bits += 16;
}
#endif
if ((sym2 = r->m_tables[0].m_look_up[bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0)
if ((sym2 = r->m_look_up[0][bit_buf & (TINFL_FAST_LOOKUP_SIZE - 1)]) >= 0)
code_len = sym2 >> 9;
else
{
code_len = TINFL_FAST_LOOKUP_BITS;
do
{
sym2 = r->m_tables[0].m_tree[~sym2 + ((bit_buf >> code_len++) & 1)];
sym2 = r->m_tree_0[~sym2 + ((bit_buf >> code_len++) & 1)];
} while (sym2 < 0);
}
bit_buf >>= code_len;
@@ -2704,7 +2764,7 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
counter += extra_bits;
}
TINFL_HUFF_DECODE(26, dist, &r->m_tables[1]);
TINFL_HUFF_DECODE(26, dist, r->m_look_up[1], r->m_tree_1);
num_extra = s_dist_extra[dist];
dist = s_dist_base[dist];
if (num_extra)
@@ -2789,7 +2849,7 @@ tinfl_status tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_nex
--pIn_buf_cur;
num_bits -= 8;
}
bit_buf &= (tinfl_bit_buf_t)((((mz_uint64)1) << num_bits) - (mz_uint64)1);
bit_buf &= ~(~(tinfl_bit_buf_t)0 << num_bits);
MZ_ASSERT(!num_bits); /* if this assert fires then we've read beyond the end of non-deflate/zlib streams with following data (such as gzip streams). */
if (decomp_flags & TINFL_FLAG_PARSE_ZLIB_HEADER)
@@ -2821,7 +2881,7 @@ common_exit:
}
}
r->m_num_bits = num_bits;
r->m_bit_buf = bit_buf & (tinfl_bit_buf_t)((((mz_uint64)1) << num_bits) - (mz_uint64)1);
r->m_bit_buf = bit_buf & ~(~(tinfl_bit_buf_t)0 << num_bits);
r->m_dist = dist;
r->m_counter = counter;
r->m_num_extra = num_extra;
@@ -2916,6 +2976,7 @@ int tinfl_decompress_mem_to_callback(const void *pIn_buf, size_t *pIn_buf_size,
size_t in_buf_ofs = 0, dict_ofs = 0;
if (!pDict)
return TINFL_STATUS_FAILED;
memset(pDict,0,TINFL_LZ_DICT_SIZE);
tinfl_init(&decomp);
for (;;)
{
@@ -2938,7 +2999,7 @@ int tinfl_decompress_mem_to_callback(const void *pIn_buf, size_t *pIn_buf_size,
}
#ifndef MINIZ_NO_MALLOC
tinfl_decompressor *tinfl_decompressor_alloc()
tinfl_decompressor *tinfl_decompressor_alloc(void)
{
tinfl_decompressor *pDecomp = (tinfl_decompressor *)MZ_MALLOC(sizeof(tinfl_decompressor));
if (pDecomp)
@@ -2955,6 +3016,8 @@ void tinfl_decompressor_free(tinfl_decompressor *pDecomp)
#ifdef __cplusplus
}
#endif
#endif /*#ifndef MINIZ_NO_INFLATE_APIS*/
/**************************************************************************
*
* Copyright 2013-2014 RAD Game Tools and Valve Software
@@ -2997,19 +3060,48 @@ extern "C" {
#include <sys/stat.h>
#if defined(_MSC_VER) || defined(__MINGW64__)
#define WIN32_LEAN_AND_MEAN
#include <windows.h>
static WCHAR* mz_utf8z_to_widechar(const char* str)
{
int reqChars = MultiByteToWideChar(CP_UTF8, 0, str, -1, NULL, 0);
WCHAR* wStr = (WCHAR*)malloc(reqChars * sizeof(WCHAR));
MultiByteToWideChar(CP_UTF8, 0, str, -1, wStr, reqChars);
return wStr;
}
static FILE *mz_fopen(const char *pFilename, const char *pMode)
{
FILE *pFile = NULL;
fopen_s(&pFile, pFilename, pMode);
return pFile;
WCHAR* wFilename = mz_utf8z_to_widechar(pFilename);
WCHAR* wMode = mz_utf8z_to_widechar(pMode);
FILE* pFile = NULL;
errno_t err = _wfopen_s(&pFile, wFilename, wMode);
free(wFilename);
free(wMode);
return err ? NULL : pFile;
}
static FILE *mz_freopen(const char *pPath, const char *pMode, FILE *pStream)
{
FILE *pFile = NULL;
if (freopen_s(&pFile, pPath, pMode, pStream))
return NULL;
return pFile;
WCHAR* wPath = mz_utf8z_to_widechar(pPath);
WCHAR* wMode = mz_utf8z_to_widechar(pMode);
FILE* pFile = NULL;
errno_t err = _wfreopen_s(&pFile, wPath, wMode, pStream);
free(wPath);
free(wMode);
return err ? NULL : pFile;
}
static int mz_stat64(const char *path, struct __stat64 *buffer)
{
WCHAR* wPath = mz_utf8z_to_widechar(path);
int res = _wstat64(wPath, buffer);
free(wPath);
return res;
}
#ifndef MINIZ_NO_TIME
#include <sys/utime.h>
#endif
@@ -3020,11 +3112,12 @@ static FILE *mz_freopen(const char *pPath, const char *pMode, FILE *pStream)
#define MZ_FTELL64 _ftelli64
#define MZ_FSEEK64 _fseeki64
#define MZ_FILE_STAT_STRUCT _stat64
#define MZ_FILE_STAT _stat64
#define MZ_FILE_STAT mz_stat64
#define MZ_FFLUSH fflush
#define MZ_FREOPEN mz_freopen
#define MZ_DELETE_FILE remove
#elif defined(__MINGW32__)
#elif defined(__MINGW32__) || defined(__WATCOMC__)
#ifndef MINIZ_NO_TIME
#include <sys/utime.h>
#endif
@@ -3032,13 +3125,14 @@ static FILE *mz_freopen(const char *pPath, const char *pMode, FILE *pStream)
#define MZ_FCLOSE fclose
#define MZ_FREAD fread
#define MZ_FWRITE fwrite
#define MZ_FTELL64 ftello64
#define MZ_FSEEK64 fseeko64
#define MZ_FILE_STAT_STRUCT _stat
#define MZ_FILE_STAT _stat
#define MZ_FTELL64 _ftelli64
#define MZ_FSEEK64 _fseeki64
#define MZ_FILE_STAT_STRUCT stat
#define MZ_FILE_STAT stat
#define MZ_FFLUSH fflush
#define MZ_FREOPEN(f, m, s) freopen(f, m, s)
#define MZ_DELETE_FILE remove
#elif defined(__TINYC__)
#ifndef MINIZ_NO_TIME
#include <sys/utime.h>
@@ -3054,6 +3148,7 @@ static FILE *mz_freopen(const char *pPath, const char *pMode, FILE *pStream)
#define MZ_FFLUSH fflush
#define MZ_FREOPEN(f, m, s) freopen(f, m, s)
#define MZ_DELETE_FILE remove
#elif defined(__USE_LARGEFILE64) /* gcc, clang */
#ifndef MINIZ_NO_TIME
#include <utime.h>
@@ -3069,7 +3164,8 @@ static FILE *mz_freopen(const char *pPath, const char *pMode, FILE *pStream)
#define MZ_FFLUSH fflush
#define MZ_FREOPEN(p, m, s) freopen64(p, m, s)
#define MZ_DELETE_FILE remove
#elif defined(__APPLE__)
#elif defined(__APPLE__) || defined(__FreeBSD__)
#ifndef MINIZ_NO_TIME
#include <utime.h>
#endif
@@ -3215,7 +3311,7 @@ struct mz_zip_internal_state_tag
mz_zip_array m_sorted_central_dir_offsets;
/* The flags passed in when the archive is initially opened. */
uint32_t m_init_flags;
mz_uint32 m_init_flags;
/* MZ_TRUE if the archive has a zip64 end of central directory headers, etc. */
mz_bool m_zip64;
@@ -3651,7 +3747,7 @@ static mz_bool mz_zip_reader_read_central_dir(mz_zip_archive *pZip, mz_uint flag
if (((num_this_disk | cdir_disk_index) != 0) && ((num_this_disk != 1) || (cdir_disk_index != 1)))
return mz_zip_set_error(pZip, MZ_ZIP_UNSUPPORTED_MULTIDISK);
if (cdir_size < pZip->m_total_files * MZ_ZIP_CENTRAL_DIR_HEADER_SIZE)
if (cdir_size < (mz_uint64)pZip->m_total_files * MZ_ZIP_CENTRAL_DIR_HEADER_SIZE)
return mz_zip_set_error(pZip, MZ_ZIP_INVALID_HEADER_OR_CORRUPTED);
if ((cdir_ofs + (mz_uint64)cdir_size) > pZip->m_archive_size)
@@ -3802,7 +3898,7 @@ static mz_bool mz_zip_reader_read_central_dir(mz_zip_archive *pZip, mz_uint flag
void mz_zip_zero_struct(mz_zip_archive *pZip)
{
if (pZip)
MZ_CLEAR_OBJ(*pZip);
MZ_CLEAR_PTR(pZip);
}
static mz_bool mz_zip_reader_end_internal(mz_zip_archive *pZip, mz_bool set_last_error)
@@ -4276,7 +4372,7 @@ static mz_bool mz_zip_locate_file_binary_search(mz_zip_archive *pZip, const char
const mz_zip_array *pCentral_dir_offsets = &pState->m_central_dir_offsets;
const mz_zip_array *pCentral_dir = &pState->m_central_dir;
mz_uint32 *pIndices = &MZ_ZIP_ARRAY_ELEMENT(&pState->m_sorted_central_dir_offsets, mz_uint32, 0);
const uint32_t size = pZip->m_total_files;
const mz_uint32 size = pZip->m_total_files;
const mz_uint filename_len = (mz_uint)strlen(pFilename);
if (pIndex)
@@ -4291,7 +4387,7 @@ static mz_bool mz_zip_locate_file_binary_search(mz_zip_archive *pZip, const char
while (l <= h)
{
mz_int64 m = l + ((h - l) >> 1);
uint32_t file_index = pIndices[(uint32_t)m];
mz_uint32 file_index = pIndices[(mz_uint32)m];
int comp = mz_zip_filename_compare(pCentral_dir, pCentral_dir_offsets, file_index, pFilename, filename_len);
if (!comp)
@@ -4384,7 +4480,8 @@ mz_bool mz_zip_reader_locate_file_v2(mz_zip_archive *pZip, const char *pName, co
return mz_zip_set_error(pZip, MZ_ZIP_FILE_NOT_FOUND);
}
mz_bool mz_zip_reader_extract_to_mem_no_alloc(mz_zip_archive *pZip, mz_uint file_index, void *pBuf, size_t buf_size, mz_uint flags, void *pUser_read_buf, size_t user_read_buf_size)
static
mz_bool mz_zip_reader_extract_to_mem_no_alloc1(mz_zip_archive *pZip, mz_uint file_index, void *pBuf, size_t buf_size, mz_uint flags, void *pUser_read_buf, size_t user_read_buf_size, const mz_zip_archive_file_stat *st)
{
int status = TINFL_STATUS_DONE;
mz_uint64 needed_size, cur_file_ofs, comp_remaining, out_buf_ofs = 0, read_buf_size, read_buf_ofs = 0, read_buf_avail;
@@ -4397,6 +4494,9 @@ mz_bool mz_zip_reader_extract_to_mem_no_alloc(mz_zip_archive *pZip, mz_uint file
if ((!pZip) || (!pZip->m_pState) || ((buf_size) && (!pBuf)) || ((user_read_buf_size) && (!pUser_read_buf)) || (!pZip->m_pRead))
return mz_zip_set_error(pZip, MZ_ZIP_INVALID_PARAMETER);
if (st) {
file_stat = *st;
} else
if (!mz_zip_reader_file_stat(pZip, file_index, &file_stat))
return MZ_FALSE;
@@ -4527,17 +4627,22 @@ mz_bool mz_zip_reader_extract_to_mem_no_alloc(mz_zip_archive *pZip, mz_uint file
return status == TINFL_STATUS_DONE;
}
mz_bool mz_zip_reader_extract_to_mem_no_alloc(mz_zip_archive *pZip, mz_uint file_index, void *pBuf, size_t buf_size, mz_uint flags, void *pUser_read_buf, size_t user_read_buf_size)
{
return mz_zip_reader_extract_to_mem_no_alloc1(pZip, file_index, pBuf, buf_size, flags, pUser_read_buf, user_read_buf_size, NULL);
}
mz_bool mz_zip_reader_extract_file_to_mem_no_alloc(mz_zip_archive *pZip, const char *pFilename, void *pBuf, size_t buf_size, mz_uint flags, void *pUser_read_buf, size_t user_read_buf_size)
{
mz_uint32 file_index;
if (!mz_zip_reader_locate_file_v2(pZip, pFilename, NULL, flags, &file_index))
return MZ_FALSE;
return mz_zip_reader_extract_to_mem_no_alloc(pZip, file_index, pBuf, buf_size, flags, pUser_read_buf, user_read_buf_size);
return mz_zip_reader_extract_to_mem_no_alloc1(pZip, file_index, pBuf, buf_size, flags, pUser_read_buf, user_read_buf_size, NULL);
}
mz_bool mz_zip_reader_extract_to_mem(mz_zip_archive *pZip, mz_uint file_index, void *pBuf, size_t buf_size, mz_uint flags)
{
return mz_zip_reader_extract_to_mem_no_alloc(pZip, file_index, pBuf, buf_size, flags, NULL, 0);
return mz_zip_reader_extract_to_mem_no_alloc1(pZip, file_index, pBuf, buf_size, flags, NULL, 0, NULL);
}
mz_bool mz_zip_reader_extract_file_to_mem(mz_zip_archive *pZip, const char *pFilename, void *pBuf, size_t buf_size, mz_uint flags)
@@ -4547,23 +4652,17 @@ mz_bool mz_zip_reader_extract_file_to_mem(mz_zip_archive *pZip, const char *pFil
void *mz_zip_reader_extract_to_heap(mz_zip_archive *pZip, mz_uint file_index, size_t *pSize, mz_uint flags)
{
mz_uint64 comp_size, uncomp_size, alloc_size;
const mz_uint8 *p = mz_zip_get_cdh(pZip, file_index);
mz_zip_archive_file_stat file_stat;
mz_uint64 alloc_size;
void *pBuf;
if (pSize)
*pSize = 0;
if (!p)
{
mz_zip_set_error(pZip, MZ_ZIP_INVALID_PARAMETER);
if (!mz_zip_reader_file_stat(pZip, file_index, &file_stat))
return NULL;
}
comp_size = MZ_READ_LE32(p + MZ_ZIP_CDH_COMPRESSED_SIZE_OFS);
uncomp_size = MZ_READ_LE32(p + MZ_ZIP_CDH_DECOMPRESSED_SIZE_OFS);
alloc_size = (flags & MZ_ZIP_FLAG_COMPRESSED_DATA) ? comp_size : uncomp_size;
alloc_size = (flags & MZ_ZIP_FLAG_COMPRESSED_DATA) ? file_stat.m_comp_size : file_stat.m_uncomp_size;
if (((sizeof(size_t) == sizeof(mz_uint32))) && (alloc_size > 0x7FFFFFFF))
{
mz_zip_set_error(pZip, MZ_ZIP_INTERNAL_ERROR);
@@ -4576,7 +4675,7 @@ void *mz_zip_reader_extract_to_heap(mz_zip_archive *pZip, mz_uint file_index, si
return NULL;
}
if (!mz_zip_reader_extract_to_mem(pZip, file_index, pBuf, (size_t)alloc_size, flags))
if (!mz_zip_reader_extract_to_mem_no_alloc1(pZip, file_index, pBuf, (size_t)alloc_size, flags, NULL, 0, &file_stat))
{
pZip->m_pFree(pZip->m_pAlloc_opaque, pBuf);
return NULL;
@@ -5037,7 +5136,7 @@ size_t mz_zip_reader_extract_iter_read(mz_zip_reader_extract_iter_state* pState,
size_t to_copy = MZ_MIN( (buf_size - copied_to_caller), pState->out_blk_remain );
/* Copy data to caller's buffer */
memcpy( (uint8_t*)pvBuf + copied_to_caller, pWrite_buf_cur, to_copy );
memcpy( (mz_uint8*)pvBuf + copied_to_caller, pWrite_buf_cur, to_copy );
#ifndef MINIZ_DISABLE_ZIP_READER_CRC32_CHECKS
/* Perform CRC */
@@ -5406,7 +5505,7 @@ handle_failure:
mz_bool mz_zip_validate_archive(mz_zip_archive *pZip, mz_uint flags)
{
mz_zip_internal_state *pState;
uint32_t i;
mz_uint32 i;
if ((!pZip) || (!pZip->m_pState) || (!pZip->m_pAlloc) || (!pZip->m_pFree) || (!pZip->m_pRead))
return mz_zip_set_error(pZip, MZ_ZIP_INVALID_PARAMETER);
@@ -5424,9 +5523,6 @@ mz_bool mz_zip_validate_archive(mz_zip_archive *pZip, mz_uint flags)
}
else
{
if (pZip->m_total_files >= MZ_UINT32_MAX)
return mz_zip_set_error(pZip, MZ_ZIP_ARCHIVE_TOO_LARGE);
if (pState->m_central_dir.m_size >= MZ_UINT32_MAX)
return mz_zip_set_error(pZip, MZ_ZIP_ARCHIVE_TOO_LARGE);
}
@@ -5788,7 +5884,7 @@ mz_bool mz_zip_writer_init_file_v2(mz_zip_archive *pZip, const char *pFilename,
mz_uint64 cur_ofs = 0;
char buf[4096];
MZ_CLEAR_OBJ(buf);
MZ_CLEAR_ARR(buf);
do
{
@@ -6151,7 +6247,7 @@ mz_bool mz_zip_writer_add_mem_ex_v2(mz_zip_archive *pZip, const char *pArchive_n
pState->m_zip64 = MZ_TRUE;
/*return mz_zip_set_error(pZip, MZ_ZIP_TOO_MANY_FILES); */
}
if ((buf_size > 0xFFFFFFFF) || (uncomp_size > 0xFFFFFFFF))
if (((mz_uint64)buf_size > 0xFFFFFFFF) || (uncomp_size > 0xFFFFFFFF))
{
pState->m_zip64 = MZ_TRUE;
/*return mz_zip_set_error(pZip, MZ_ZIP_ARCHIVE_TOO_LARGE); */
@@ -6244,7 +6340,7 @@ mz_bool mz_zip_writer_add_mem_ex_v2(mz_zip_archive *pZip, const char *pArchive_n
}
cur_archive_file_ofs += num_alignment_padding_bytes;
MZ_CLEAR_OBJ(local_dir_header);
MZ_CLEAR_ARR(local_dir_header);
if (!store_data_uncompressed || (level_and_flags & MZ_ZIP_FLAG_COMPRESSED_DATA))
{
@@ -6394,7 +6490,7 @@ mz_bool mz_zip_writer_add_mem_ex_v2(mz_zip_archive *pZip, const char *pArchive_n
mz_bool mz_zip_writer_add_read_buf_callback(mz_zip_archive *pZip, const char *pArchive_name, mz_file_read_func read_callback, void* callback_opaque, mz_uint64 max_size, const MZ_TIME_T *pFile_time, const void *pComment, mz_uint16 comment_size, mz_uint level_and_flags,
const char *user_extra_data, mz_uint user_extra_data_len, const char *user_extra_data_central, mz_uint user_extra_data_central_len)
{
mz_uint16 gen_flags = (level_and_flags & MZ_ZIP_FLAG_WRITE_HEADER_SET_SIZE) ? 0 : MZ_ZIP_LDH_BIT_FLAG_HAS_LOCATOR;
mz_uint16 gen_flags;
mz_uint uncomp_crc32 = MZ_CRC32_INIT, level, num_alignment_padding_bytes;
mz_uint16 method = 0, dos_time = 0, dos_date = 0, ext_attributes = 0;
mz_uint64 local_dir_header_ofs, cur_archive_file_ofs = pZip->m_archive_size, uncomp_size = 0, comp_size = 0;
@@ -6406,13 +6502,15 @@ mz_bool mz_zip_writer_add_read_buf_callback(mz_zip_archive *pZip, const char *pA
mz_zip_internal_state *pState;
mz_uint64 file_ofs = 0, cur_archive_header_file_ofs;
if (!(level_and_flags & MZ_ZIP_FLAG_ASCII_FILENAME))
gen_flags |= MZ_ZIP_GENERAL_PURPOSE_BIT_FLAG_UTF8;
if ((int)level_and_flags < 0)
level_and_flags = MZ_DEFAULT_LEVEL;
level = level_and_flags & 0xF;
gen_flags = (level_and_flags & MZ_ZIP_FLAG_WRITE_HEADER_SET_SIZE) ? 0 : MZ_ZIP_LDH_BIT_FLAG_HAS_LOCATOR;
if (!(level_and_flags & MZ_ZIP_FLAG_ASCII_FILENAME))
gen_flags |= MZ_ZIP_GENERAL_PURPOSE_BIT_FLAG_UTF8;
/* Sanity checks */
if ((!pZip) || (!pZip->m_pState) || (pZip->m_zip_mode != MZ_ZIP_MODE_WRITING) || (!pArchive_name) || ((comment_size) && (!pComment)) || (level > MZ_UBER_COMPRESSION))
return mz_zip_set_error(pZip, MZ_ZIP_INVALID_PARAMETER);
@@ -6497,7 +6595,7 @@ mz_bool mz_zip_writer_add_read_buf_callback(mz_zip_archive *pZip, const char *pA
method = MZ_DEFLATED;
}
MZ_CLEAR_OBJ(local_dir_header);
MZ_CLEAR_ARR(local_dir_header);
if (pState->m_zip64)
{
if (max_size >= MZ_UINT32_MAX || local_dir_header_ofs >= MZ_UINT32_MAX)
@@ -6801,7 +6899,7 @@ mz_bool mz_zip_writer_add_file(mz_zip_archive *pZip, const char *pArchive_name,
}
#endif /* #ifndef MINIZ_NO_STDIO */
static mz_bool mz_zip_writer_update_zip64_extension_block(mz_zip_array *pNew_ext, mz_zip_archive *pZip, const mz_uint8 *pExt, uint32_t ext_len, mz_uint64 *pComp_size, mz_uint64 *pUncomp_size, mz_uint64 *pLocal_header_ofs, mz_uint32 *pDisk_start)
static mz_bool mz_zip_writer_update_zip64_extension_block(mz_zip_array *pNew_ext, mz_zip_archive *pZip, const mz_uint8 *pExt, mz_uint32 ext_len, mz_uint64 *pComp_size, mz_uint64 *pUncomp_size, mz_uint64 *pLocal_header_ofs, mz_uint32 *pDisk_start)
{
/* + 64 should be enough for any new zip64 data */
if (!mz_zip_array_reserve(pZip, pNew_ext, ext_len + 64, MZ_FALSE))
@@ -7117,10 +7215,10 @@ mz_bool mz_zip_writer_add_from_zip_reader(mz_zip_archive *pZip, mz_zip_archive *
if (pZip->m_pState->m_zip64)
{
/* dest is zip64, so upgrade the data descriptor */
const mz_uint32 *pSrc_descriptor = (const mz_uint32 *)((const mz_uint8 *)pBuf + (has_id ? sizeof(mz_uint32) : 0));
const mz_uint32 src_crc32 = pSrc_descriptor[0];
const mz_uint64 src_comp_size = pSrc_descriptor[1];
const mz_uint64 src_uncomp_size = pSrc_descriptor[2];
const mz_uint8 *pSrc_descriptor = (const mz_uint8 *)pBuf + (has_id ? sizeof(mz_uint32) : 0);
const mz_uint32 src_crc32 = MZ_READ_LE32(pSrc_descriptor);
const mz_uint64 src_comp_size = MZ_READ_LE32(pSrc_descriptor + sizeof(mz_uint32));
const mz_uint64 src_uncomp_size = MZ_READ_LE32(pSrc_descriptor + 2*sizeof(mz_uint32));
mz_write_le32((mz_uint8 *)pBuf, MZ_ZIP_DATA_DESCRIPTOR_ID);
mz_write_le32((mz_uint8 *)pBuf + sizeof(mz_uint32) * 1, src_crc32);
@@ -7256,7 +7354,7 @@ mz_bool mz_zip_writer_finalize_archive(mz_zip_archive *pZip)
if (pState->m_zip64)
{
if ((pZip->m_total_files > MZ_UINT32_MAX) || (pState->m_central_dir.m_size >= MZ_UINT32_MAX))
if ((mz_uint64)pState->m_central_dir.m_size >= MZ_UINT32_MAX)
return mz_zip_set_error(pZip, MZ_ZIP_TOO_MANY_FILES);
}
else
@@ -7284,7 +7382,7 @@ mz_bool mz_zip_writer_finalize_archive(mz_zip_archive *pZip)
/* Write zip64 end of central directory header */
mz_uint64 rel_ofs_to_zip64_ecdr = pZip->m_archive_size;
MZ_CLEAR_OBJ(hdr);
MZ_CLEAR_ARR(hdr);
MZ_WRITE_LE32(hdr + MZ_ZIP64_ECDH_SIG_OFS, MZ_ZIP64_END_OF_CENTRAL_DIR_HEADER_SIG);
MZ_WRITE_LE64(hdr + MZ_ZIP64_ECDH_SIZE_OF_RECORD_OFS, MZ_ZIP64_END_OF_CENTRAL_DIR_HEADER_SIZE - sizeof(mz_uint32) - sizeof(mz_uint64));
MZ_WRITE_LE16(hdr + MZ_ZIP64_ECDH_VERSION_MADE_BY_OFS, 0x031E); /* TODO: always Unix */
@@ -7299,7 +7397,7 @@ mz_bool mz_zip_writer_finalize_archive(mz_zip_archive *pZip)
pZip->m_archive_size += MZ_ZIP64_END_OF_CENTRAL_DIR_HEADER_SIZE;
/* Write zip64 end of central directory locator */
MZ_CLEAR_OBJ(hdr);
MZ_CLEAR_ARR(hdr);
MZ_WRITE_LE32(hdr + MZ_ZIP64_ECDL_SIG_OFS, MZ_ZIP64_END_OF_CENTRAL_DIR_LOCATOR_SIG);
MZ_WRITE_LE64(hdr + MZ_ZIP64_ECDL_REL_OFS_TO_ZIP64_ECDR_OFS, rel_ofs_to_zip64_ecdr);
MZ_WRITE_LE32(hdr + MZ_ZIP64_ECDL_TOTAL_NUMBER_OF_DISKS_OFS, 1);
@@ -7310,7 +7408,7 @@ mz_bool mz_zip_writer_finalize_archive(mz_zip_archive *pZip)
}
/* Write end of central directory record */
MZ_CLEAR_OBJ(hdr);
MZ_CLEAR_ARR(hdr);
MZ_WRITE_LE32(hdr + MZ_ZIP_ECDH_SIG_OFS, MZ_ZIP_END_OF_CENTRAL_DIR_HEADER_SIG);
MZ_WRITE_LE16(hdr + MZ_ZIP_ECDH_CDIR_NUM_ENTRIES_ON_DISK_OFS, MZ_MIN(MZ_UINT16_MAX, pZip->m_total_files));
MZ_WRITE_LE16(hdr + MZ_ZIP_ECDH_CDIR_TOTAL_ENTRIES_OFS, MZ_MIN(MZ_UINT16_MAX, pZip->m_total_files));
@@ -7626,7 +7724,9 @@ const char *mz_zip_get_error_string(mz_zip_error mz_err)
case MZ_ZIP_VALIDATION_FAILED:
return "validation failed";
case MZ_ZIP_WRITE_CALLBACK_FAILED:
return "write calledback failed";
return "write callback failed";
case MZ_ZIP_TOTAL_ERRORS:
return "total errors";
default:
break;
}

130
ext/miniz/miniz.h
View File

@@ -1,5 +1,7 @@
#ifndef MINIZ_EXPORT
#define MINIZ_EXPORT
/* miniz.c 2.2.0 - public domain deflate/inflate, zlib-subset, ZIP reading/writing/appending, PNG writing
#endif
/* miniz.c 3.0.2 - public domain deflate/inflate, zlib-subset, ZIP reading/writing/appending, PNG writing
See "unlicense" statement at the end of this file.
Rich Geldreich <richgel99@gmail.com>, last updated Oct. 13, 2013
Implements RFC 1950: http://www.ietf.org/rfc/rfc1950.txt and RFC 1951: http://www.ietf.org/rfc/rfc1951.txt
@@ -115,8 +117,8 @@
/* Defines to completely disable specific portions of miniz.c:
If all macros here are defined the only functionality remaining will be CRC-32, adler-32, tinfl, and tdefl. */
/* Defines to completely disable specific portions of miniz.c:
If all macros here are defined the only functionality remaining will be CRC-32 and adler-32. */
/* Define MINIZ_NO_STDIO to disable all usage and any functions which rely on stdio for file I/O. */
/*#define MINIZ_NO_STDIO */
@@ -126,6 +128,12 @@
/* The current downside is the times written to your archives will be from 1979. */
/*#define MINIZ_NO_TIME */
/* Define MINIZ_NO_DEFLATE_APIS to disable all compression API's. */
/*#define MINIZ_NO_DEFLATE_APIS */
/* Define MINIZ_NO_INFLATE_APIS to disable all decompression API's. */
/*#define MINIZ_NO_INFLATE_APIS */
/* Define MINIZ_NO_ARCHIVE_APIS to disable all ZIP archive API's. */
/*#define MINIZ_NO_ARCHIVE_APIS */
@@ -138,12 +146,20 @@
/* Define MINIZ_NO_ZLIB_COMPATIBLE_NAME to disable zlib names, to prevent conflicts against stock zlib. */
/*#define MINIZ_NO_ZLIB_COMPATIBLE_NAMES */
/* Define MINIZ_NO_MALLOC to disable all calls to malloc, free, and realloc.
/* Define MINIZ_NO_MALLOC to disable all calls to malloc, free, and realloc.
Note if MINIZ_NO_MALLOC is defined then the user must always provide custom user alloc/free/realloc
callbacks to the zlib and archive API's, and a few stand-alone helper API's which don't provide custom user
functions (such as tdefl_compress_mem_to_heap() and tinfl_decompress_mem_to_heap()) won't work. */
/*#define MINIZ_NO_MALLOC */
#ifdef MINIZ_NO_INFLATE_APIS
#define MINIZ_NO_ARCHIVE_APIS
#endif
#ifdef MINIZ_NO_DEFLATE_APIS
#define MINIZ_NO_ARCHIVE_WRITING_APIS
#endif
#if defined(__TINYC__) && (defined(__linux) || defined(__linux__))
/* TODO: Work around "error: include file 'sys\utime.h' when compiling with tcc on Linux */
#define MINIZ_NO_TIME
@@ -162,18 +178,40 @@
#define MINIZ_X86_OR_X64_CPU 0
#endif
#if (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__) || MINIZ_X86_OR_X64_CPU
/* Set MINIZ_LITTLE_ENDIAN only if not set */
#if !defined(MINIZ_LITTLE_ENDIAN)
#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__)
#if (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
/* Set MINIZ_LITTLE_ENDIAN to 1 if the processor is little endian. */
#define MINIZ_LITTLE_ENDIAN 1
#else
#define MINIZ_LITTLE_ENDIAN 0
#endif
#else
#if MINIZ_X86_OR_X64_CPU
#define MINIZ_LITTLE_ENDIAN 1
#else
#define MINIZ_LITTLE_ENDIAN 0
#endif
#endif
#endif
/* Using unaligned loads and stores causes errors when using UBSan */
#if defined(__has_feature)
#if __has_feature(undefined_behavior_sanitizer)
#define MINIZ_USE_UNALIGNED_LOADS_AND_STORES 0
#endif
#endif
/* Set MINIZ_USE_UNALIGNED_LOADS_AND_STORES only if not set */
#if !defined(MINIZ_USE_UNALIGNED_LOADS_AND_STORES)
#if MINIZ_X86_OR_X64_CPU
/* Set MINIZ_USE_UNALIGNED_LOADS_AND_STORES to 1 on CPU's that permit efficient integer loads and stores from unaligned addresses. */
#define MINIZ_USE_UNALIGNED_LOADS_AND_STORES 1
#define MINIZ_USE_UNALIGNED_LOADS_AND_STORES 0
#define MINIZ_UNALIGNED_USE_MEMCPY
#else
#define MINIZ_USE_UNALIGNED_LOADS_AND_STORES 0
@@ -237,9 +275,9 @@ enum
MZ_DEFAULT_COMPRESSION = -1
};
#define MZ_VERSION "10.2.0"
#define MZ_VERNUM 0xA100
#define MZ_VER_MAJOR 10
#define MZ_VERSION "11.0.2"
#define MZ_VERNUM 0xB002
#define MZ_VER_MAJOR 11
#define MZ_VER_MINOR 2
#define MZ_VER_REVISION 0
#define MZ_VER_SUBREVISION 0
@@ -305,6 +343,8 @@ typedef mz_stream *mz_streamp;
/* Returns the version string of miniz.c. */
MINIZ_EXPORT const char *mz_version(void);
#ifndef MINIZ_NO_DEFLATE_APIS
/* mz_deflateInit() initializes a compressor with default options: */
/* Parameters: */
/* pStream must point to an initialized mz_stream struct. */
@@ -357,6 +397,10 @@ MINIZ_EXPORT int mz_compress2(unsigned char *pDest, mz_ulong *pDest_len, const u
/* mz_compressBound() returns a (very) conservative upper bound on the amount of data that could be generated by calling mz_compress(). */
MINIZ_EXPORT mz_ulong mz_compressBound(mz_ulong source_len);
#endif /*#ifndef MINIZ_NO_DEFLATE_APIS*/
#ifndef MINIZ_NO_INFLATE_APIS
/* Initializes a decompressor. */
MINIZ_EXPORT int mz_inflateInit(mz_streamp pStream);
@@ -390,6 +434,7 @@ MINIZ_EXPORT int mz_inflateEnd(mz_streamp pStream);
/* Returns MZ_OK on success, or one of the error codes from mz_inflate() on failure. */
MINIZ_EXPORT int mz_uncompress(unsigned char *pDest, mz_ulong *pDest_len, const unsigned char *pSource, mz_ulong source_len);
MINIZ_EXPORT int mz_uncompress2(unsigned char *pDest, mz_ulong *pDest_len, const unsigned char *pSource, mz_ulong *pSource_len);
#endif /*#ifndef MINIZ_NO_INFLATE_APIS*/
/* Returns a string description of the specified error code, or NULL if the error code is invalid. */
MINIZ_EXPORT const char *mz_error(int err);
@@ -440,6 +485,8 @@ typedef void *const voidpc;
#define free_func mz_free_func
#define internal_state mz_internal_state
#define z_stream mz_stream
#ifndef MINIZ_NO_DEFLATE_APIS
#define deflateInit mz_deflateInit
#define deflateInit2 mz_deflateInit2
#define deflateReset mz_deflateReset
@@ -449,6 +496,9 @@ typedef void *const voidpc;
#define compress mz_compress
#define compress2 mz_compress2
#define compressBound mz_compressBound
#endif /*#ifndef MINIZ_NO_DEFLATE_APIS*/
#ifndef MINIZ_NO_INFLATE_APIS
#define inflateInit mz_inflateInit
#define inflateInit2 mz_inflateInit2
#define inflateReset mz_inflateReset
@@ -456,6 +506,8 @@ typedef void *const voidpc;
#define inflateEnd mz_inflateEnd
#define uncompress mz_uncompress
#define uncompress2 mz_uncompress2
#endif /*#ifndef MINIZ_NO_INFLATE_APIS*/
#define crc32 mz_crc32
#define adler32 mz_adler32
#define MAX_WBITS 15
@@ -519,7 +571,8 @@ typedef int mz_bool;
#ifdef MINIZ_NO_TIME
typedef struct mz_dummy_time_t_tag
{
int m_dummy;
mz_uint32 m_dummy1;
mz_uint32 m_dummy2;
} mz_dummy_time_t;
#define MZ_TIME_T mz_dummy_time_t
#else
@@ -541,6 +594,8 @@ typedef struct mz_dummy_time_t_tag
#define MZ_MAX(a, b) (((a) > (b)) ? (a) : (b))
#define MZ_MIN(a, b) (((a) < (b)) ? (a) : (b))
#define MZ_CLEAR_OBJ(obj) memset(&(obj), 0, sizeof(obj))
#define MZ_CLEAR_ARR(obj) memset((obj), 0, sizeof(obj))
#define MZ_CLEAR_PTR(obj) memset((obj), 0, sizeof(*obj))
#if MINIZ_USE_UNALIGNED_LOADS_AND_STORES && MINIZ_LITTLE_ENDIAN
#define MZ_READ_LE16(p) *((const mz_uint16 *)(p))
@@ -577,6 +632,8 @@ extern MINIZ_EXPORT void *miniz_def_realloc_func(void *opaque, void *address, si
#pragma once
#ifndef MINIZ_NO_DEFLATE_APIS
#ifdef __cplusplus
extern "C" {
#endif
@@ -764,10 +821,14 @@ MINIZ_EXPORT void tdefl_compressor_free(tdefl_compressor *pComp);
#ifdef __cplusplus
}
#endif
#endif /*#ifndef MINIZ_NO_DEFLATE_APIS*/
#pragma once
/* ------------------- Low-level Decompression API Definitions */
#ifndef MINIZ_NO_INFLATE_APIS
#ifdef __cplusplus
extern "C" {
#endif
@@ -876,12 +937,6 @@ enum
TINFL_FAST_LOOKUP_SIZE = 1 << TINFL_FAST_LOOKUP_BITS
};
typedef struct
{
mz_uint8 m_code_size[TINFL_MAX_HUFF_SYMBOLS_0];
mz_int16 m_look_up[TINFL_FAST_LOOKUP_SIZE], m_tree[TINFL_MAX_HUFF_SYMBOLS_0 * 2];
} tinfl_huff_table;
#if MINIZ_HAS_64BIT_REGISTERS
#define TINFL_USE_64BIT_BITBUF 1
#else
@@ -901,14 +956,22 @@ struct tinfl_decompressor_tag
mz_uint32 m_state, m_num_bits, m_zhdr0, m_zhdr1, m_z_adler32, m_final, m_type, m_check_adler32, m_dist, m_counter, m_num_extra, m_table_sizes[TINFL_MAX_HUFF_TABLES];
tinfl_bit_buf_t m_bit_buf;
size_t m_dist_from_out_buf_start;
tinfl_huff_table m_tables[TINFL_MAX_HUFF_TABLES];
mz_int16 m_look_up[TINFL_MAX_HUFF_TABLES][TINFL_FAST_LOOKUP_SIZE];
mz_int16 m_tree_0[TINFL_MAX_HUFF_SYMBOLS_0 * 2];
mz_int16 m_tree_1[TINFL_MAX_HUFF_SYMBOLS_1 * 2];
mz_int16 m_tree_2[TINFL_MAX_HUFF_SYMBOLS_2 * 2];
mz_uint8 m_code_size_0[TINFL_MAX_HUFF_SYMBOLS_0];
mz_uint8 m_code_size_1[TINFL_MAX_HUFF_SYMBOLS_1];
mz_uint8 m_code_size_2[TINFL_MAX_HUFF_SYMBOLS_2];
mz_uint8 m_raw_header[4], m_len_codes[TINFL_MAX_HUFF_SYMBOLS_0 + TINFL_MAX_HUFF_SYMBOLS_1 + 137];
};
#ifdef __cplusplus
}
#endif
#endif /*#ifndef MINIZ_NO_INFLATE_APIS*/
#pragma once
@@ -942,10 +1005,6 @@ typedef struct
mz_uint16 m_bit_flag;
mz_uint16 m_method;
#ifndef MINIZ_NO_TIME
MZ_TIME_T m_time;
#endif
/* CRC-32 of uncompressed data. */
mz_uint32 m_crc32;
@@ -982,6 +1041,11 @@ typedef struct
/* Guaranteed to be zero terminated, may be truncated to fit. */
char m_comment[MZ_ZIP_MAX_ARCHIVE_FILE_COMMENT_SIZE];
#ifdef MINIZ_NO_TIME
MZ_TIME_T m_padding;
#else
MZ_TIME_T m_time;
#endif
} mz_zip_archive_file_stat;
typedef size_t (*mz_file_read_func)(void *pOpaque, mz_uint64 file_ofs, void *pBuf, size_t n);
@@ -1093,9 +1157,7 @@ typedef struct
mz_uint flags;
int status;
#ifndef MINIZ_DISABLE_ZIP_READER_CRC32_CHECKS
mz_uint file_crc32;
#endif
mz_uint64 read_buf_size, read_buf_ofs, read_buf_avail, comp_remaining, out_buf_ofs, cur_file_ofs;
mz_zip_archive_file_stat file_stat;
void *pRead_buf;
@@ -1105,6 +1167,12 @@ typedef struct
tinfl_decompressor inflator;
#ifdef MINIZ_DISABLE_ZIP_READER_CRC32_CHECKS
mz_uint padding;
#else
mz_uint file_crc32;
#endif
} mz_zip_reader_extract_iter_state;
/* -------- ZIP reading */
@@ -1228,9 +1296,9 @@ MINIZ_EXPORT mz_bool mz_zip_reader_extract_file_to_cfile(mz_zip_archive *pZip, c
/* TODO */
typedef void *mz_zip_streaming_extract_state_ptr;
mz_zip_streaming_extract_state_ptr mz_zip_streaming_extract_begin(mz_zip_archive *pZip, mz_uint file_index, mz_uint flags);
uint64_t mz_zip_streaming_extract_get_size(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState);
uint64_t mz_zip_streaming_extract_get_cur_ofs(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState);
mz_bool mz_zip_streaming_extract_seek(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState, uint64_t new_ofs);
mz_uint64 mz_zip_streaming_extract_get_size(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState);
mz_uint64 mz_zip_streaming_extract_get_cur_ofs(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState);
mz_bool mz_zip_streaming_extract_seek(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState, mz_uint64 new_ofs);
size_t mz_zip_streaming_extract_read(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState, void *pBuf, size_t buf_size);
mz_bool mz_zip_streaming_extract_end(mz_zip_archive *pZip, mz_zip_streaming_extract_state_ptr pState);
#endif
@@ -1244,7 +1312,9 @@ MINIZ_EXPORT mz_bool mz_zip_validate_archive(mz_zip_archive *pZip, mz_uint flags
/* Misc utils/helpers, valid for ZIP reading or writing */
MINIZ_EXPORT mz_bool mz_zip_validate_mem_archive(const void *pMem, size_t size, mz_uint flags, mz_zip_error *pErr);
#ifndef MINIZ_NO_STDIO
MINIZ_EXPORT mz_bool mz_zip_validate_file_archive(const char *pFilename, mz_uint flags, mz_zip_error *pErr);
#endif
/* Universal end function - calls either mz_zip_reader_end() or mz_zip_writer_end(). */
MINIZ_EXPORT mz_bool mz_zip_end(mz_zip_archive *pZip);
@@ -1318,7 +1388,7 @@ MINIZ_EXPORT mz_bool mz_zip_writer_add_from_zip_reader(mz_zip_archive *pZip, mz_
/* An archive must be manually finalized by calling this function for it to be valid. */
MINIZ_EXPORT mz_bool mz_zip_writer_finalize_archive(mz_zip_archive *pZip);
/* Finalizes a heap archive, returning a poiner to the heap block and its size. */
/* Finalizes a heap archive, returning a pointer to the heap block and its size. */
/* The heap block will be allocated using the mz_zip_archive's alloc/realloc callbacks. */
MINIZ_EXPORT mz_bool mz_zip_writer_finalize_heap_archive(mz_zip_archive *pZip, void **ppBuf, size_t *pSize);
@@ -1335,11 +1405,13 @@ MINIZ_EXPORT mz_bool mz_zip_writer_end(mz_zip_archive *pZip);
MINIZ_EXPORT mz_bool mz_zip_add_mem_to_archive_file_in_place(const char *pZip_filename, const char *pArchive_name, const void *pBuf, size_t buf_size, const void *pComment, mz_uint16 comment_size, mz_uint level_and_flags);
MINIZ_EXPORT mz_bool mz_zip_add_mem_to_archive_file_in_place_v2(const char *pZip_filename, const char *pArchive_name, const void *pBuf, size_t buf_size, const void *pComment, mz_uint16 comment_size, mz_uint level_and_flags, mz_zip_error *pErr);
#ifndef MINIZ_NO_STDIO
/* Reads a single file from an archive into a heap block. */
/* If pComment is not NULL, only the file with the specified comment will be extracted. */
/* Returns NULL on failure. */
MINIZ_EXPORT void *mz_zip_extract_archive_file_to_heap(const char *pZip_filename, const char *pArchive_name, size_t *pSize, mz_uint flags);
MINIZ_EXPORT void *mz_zip_extract_archive_file_to_heap_v2(const char *pZip_filename, const char *pArchive_name, const char *pComment, size_t *pSize, mz_uint flags, mz_zip_error *pErr);
#endif
#endif /* #ifndef MINIZ_NO_ARCHIVE_WRITING_APIS */

47
ext/pugixml/CMakeLists.txt Normal file
View File

@@ -0,0 +1,47 @@
project(pugixml VERSION 1.14)
# Expose options from the pugiconfig.hpp
option(PUGIXML_WCHAR_MODE "Enable wchar_t mode" OFF)
option(PUGIXML_COMPACT "Enable compact mode" OFF)
# Advanced options from pugiconfig.hpp
option(PUGIXML_NO_XPATH "Disable XPath" OFF)
option(PUGIXML_NO_STL "Disable STL" OFF)
option(PUGIXML_NO_EXCEPTIONS "Disable Exceptions" OFF)
mark_as_advanced(PUGIXML_NO_XPATH PUGIXML_NO_STL PUGIXML_NO_EXCEPTIONS)
set(PUGIXML_PUBLIC_DEFINITIONS
$<$<BOOL:${PUGIXML_WCHAR_MODE}>:PUGIXML_WCHAR_MODE>
$<$<BOOL:${PUGIXML_COMPACT}>:PUGIXML_COMPACT>
$<$<BOOL:${PUGIXML_NO_XPATH}>:PUGIXML_NO_XPATH>
$<$<BOOL:${PUGIXML_NO_STL}>:PUGIXML_NO_STL>
$<$<BOOL:${PUGIXML_NO_EXCEPTIONS}>:PUGIXML_NO_EXCEPTIONS>)
add_library(pugixml STATIC
src/pugiconfig.hpp
src/pugixml.hpp
src/pugixml.cpp)
add_library(pugixml::static ALIAS pugixml)
set_target_properties(pugixml PROPERTIES
CXX_STANDARD_REQUIRED ON
CXX_STANDARD 11)
set_property(TARGET pugixml PROPERTY EXPORT_NAME static)
target_include_directories(pugixml PUBLIC
$<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/src>)
target_compile_definitions(pugixml PUBLIC
${PUGIXML_BUILD_DEFINES}
${PUGIXML_PUBLIC_DEFINITIONS})
add_library(pugixml::pugixml ALIAS pugixml)
set_target_properties(pugixml PROPERTIES
EXCLUDE_FROM_ALL ON
POSITION_INDEPENDENT_CODE ON
SOVERSION ${PROJECT_VERSION_MAJOR}
VERSION ${PROJECT_VERSION}
OUTPUT_NAME pugixml)
set_target_properties(pugixml PROPERTIES
EXCLUDE_FROM_ALL OFF)

24
ext/pugixml/LICENSE.md Normal file
View File

@@ -0,0 +1,24 @@
MIT License
Copyright (c) 2006-2023 Arseny Kapoulkine
Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation
files (the "Software"), to deal in the Software without
restriction, including without limitation the rights to use,
copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.

77
ext/pugixml/src/pugiconfig.hpp Normal file
View File

@@ -0,0 +1,77 @@
/**
* pugixml parser - version 1.14
* --------------------------------------------------------
* Copyright (C) 2006-2023, by Arseny Kapoulkine (arseny.kapoulkine@gmail.com)
* Report bugs and download new versions at https://pugixml.org/
*
* This library is distributed under the MIT License. See notice at the end
* of this file.
*
* This work is based on the pugxml parser, which is:
* Copyright (C) 2003, by Kristen Wegner (kristen@tima.net)
*/
#ifndef HEADER_PUGICONFIG_HPP
#define HEADER_PUGICONFIG_HPP
// Uncomment this to enable wchar_t mode
// #define PUGIXML_WCHAR_MODE
// Uncomment this to enable compact mode
// #define PUGIXML_COMPACT
// Uncomment this to disable XPath
// #define PUGIXML_NO_XPATH
// Uncomment this to disable STL
// #define PUGIXML_NO_STL
// Uncomment this to disable exceptions
// #define PUGIXML_NO_EXCEPTIONS
// Set this to control attributes for public classes/functions, i.e.:
// #define PUGIXML_API __declspec(dllexport) // to export all public symbols from DLL
// #define PUGIXML_CLASS __declspec(dllimport) // to import all classes from DLL
// #define PUGIXML_FUNCTION __fastcall // to set calling conventions to all public functions to fastcall
// In absence of PUGIXML_CLASS/PUGIXML_FUNCTION definitions PUGIXML_API is used instead
// Tune these constants to adjust memory-related behavior
// #define PUGIXML_MEMORY_PAGE_SIZE 32768
// #define PUGIXML_MEMORY_OUTPUT_STACK 10240
// #define PUGIXML_MEMORY_XPATH_PAGE_SIZE 4096
// Tune this constant to adjust max nesting for XPath queries
// #define PUGIXML_XPATH_DEPTH_LIMIT 1024
// Uncomment this to switch to header-only version
// #define PUGIXML_HEADER_ONLY
// Uncomment this to enable long long support
// #define PUGIXML_HAS_LONG_LONG
#endif
/**
* Copyright (c) 2006-2023 Arseny Kapoulkine
*
* Permission is hereby granted, free of charge, to any person
* obtaining a copy of this software and associated documentation
* files (the "Software"), to deal in the Software without
* restriction, including without limitation the rights to use,
* copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following
* conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
* OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
* HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
* WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*/

13226
ext/pugixml/src/pugixml.cpp Normal file
View File

File diff suppressed because it is too large Load Diff

1516
ext/pugixml/src/pugixml.hpp Normal file
View File

File diff suppressed because it is too large Load Diff

4
ext/rapidxml/CMakeLists.txt
View File

@@ -1,4 +0,0 @@
add_library(rapidxml INTERFACE)
add_library(External::rapidxml ALIAS rapidxml)
target_include_directories(rapidxml
INTERFACE ${CMAKE_CURRENT_SOURCE_DIR})

52
ext/rapidxml/license.txt
View File

@@ -1,52 +0,0 @@
Use of this software is granted under one of the following two licenses,
to be chosen freely by the user.
1. Boost Software License - Version 1.0 - August 17th, 2003
===============================================================================
Copyright (c) 2006, 2007 Marcin Kalicinski
Permission is hereby granted, free of charge, to any person or organization
obtaining a copy of the software and accompanying documentation covered by
this license (the "Software") to use, reproduce, display, distribute,
execute, and transmit the Software, and to prepare derivative works of the
Software, and to permit third-parties to whom the Software is furnished to
do so, all subject to the following:
The copyright notices in the Software and this entire statement, including
the above license grant, this restriction and the following disclaimer,
must be included in all copies of the Software, in whole or in part, and
all derivative works of the Software, unless such copies or derivative
works are solely in the form of machine-executable object code generated by
a source language processor.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
2. The MIT License
===============================================================================
Copyright (c) 2006, 2007 Marcin Kalicinski
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
of the Software, and to permit persons to whom the Software is furnished to do so,
subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
IN THE SOFTWARE.

2596
ext/rapidxml/rapidxml/rapidxml.hpp
View File

File diff suppressed because it is too large Load Diff

174
ext/rapidxml/rapidxml/rapidxml_iterators.hpp
View File

@@ -1,174 +0,0 @@
#ifndef RAPIDXML_ITERATORS_HPP_INCLUDED
#define RAPIDXML_ITERATORS_HPP_INCLUDED
// Copyright (C) 2006, 2009 Marcin Kalicinski
// Version 1.13
// Revision $DateTime: 2009/05/13 01:46:17 $
//! \file rapidxml_iterators.hpp This file contains rapidxml iterators
#include "rapidxml.hpp"
namespace rapidxml
{
//! Iterator of child nodes of xml_node
template<class Ch>
class node_iterator
{
public:
typedef typename xml_node<Ch> value_type;
typedef typename xml_node<Ch> &reference;
typedef typename xml_node<Ch> *pointer;
typedef std::ptrdiff_t difference_type;
typedef std::bidirectional_iterator_tag iterator_category;
node_iterator()
: m_node(0)
{
}
node_iterator(xml_node<Ch> *node)
: m_node(node->first_node())
{
}
reference operator *() const
{
assert(m_node);
return *m_node;
}
pointer operator->() const
{
assert(m_node);
return m_node;
}
node_iterator& operator++()
{
assert(m_node);
m_node = m_node->next_sibling();
return *this;
}
node_iterator operator++(int)
{
node_iterator tmp = *this;
++this;
return tmp;
}
node_iterator& operator--()
{
assert(m_node && m_node->previous_sibling());
m_node = m_node->previous_sibling();
return *this;
}
node_iterator operator--(int)
{
node_iterator tmp = *this;
++this;
return tmp;
}
bool operator ==(const node_iterator<Ch> &rhs)
{
return m_node == rhs.m_node;
}
bool operator !=(const node_iterator<Ch> &rhs)
{
return m_node != rhs.m_node;
}
private:
xml_node<Ch> *m_node;
};
//! Iterator of child attributes of xml_node
template<class Ch>
class attribute_iterator
{
public:
typedef typename xml_attribute<Ch> value_type;
typedef typename xml_attribute<Ch> &reference;
typedef typename xml_attribute<Ch> *pointer;
typedef std::ptrdiff_t difference_type;
typedef std::bidirectional_iterator_tag iterator_category;
attribute_iterator()
: m_attribute(0)
{
}
attribute_iterator(xml_node<Ch> *node)
: m_attribute(node->first_attribute())
{
}
reference operator *() const
{
assert(m_attribute);
return *m_attribute;
}
pointer operator->() const
{
assert(m_attribute);
return m_attribute;
}
attribute_iterator& operator++()
{
assert(m_attribute);
m_attribute = m_attribute->next_attribute();
return *this;
}
attribute_iterator operator++(int)
{
attribute_iterator tmp = *this;
++this;
return tmp;
}
attribute_iterator& operator--()
{
assert(m_attribute && m_attribute->previous_attribute());
m_attribute = m_attribute->previous_attribute();
return *this;
}
attribute_iterator operator--(int)
{
attribute_iterator tmp = *this;
++this;
return tmp;
}
bool operator ==(const attribute_iterator<Ch> &rhs)
{
return m_attribute == rhs.m_attribute;
}
bool operator !=(const attribute_iterator<Ch> &rhs)
{
return m_attribute != rhs.m_attribute;
}
private:
xml_attribute<Ch> *m_attribute;
};
}
#endif

421
ext/rapidxml/rapidxml/rapidxml_print.hpp
View File

@@ -1,421 +0,0 @@
#ifndef RAPIDXML_PRINT_HPP_INCLUDED
#define RAPIDXML_PRINT_HPP_INCLUDED
// Copyright (C) 2006, 2009 Marcin Kalicinski
// Version 1.13
// Revision $DateTime: 2009/05/13 01:46:17 $
//! \file rapidxml_print.hpp This file contains rapidxml printer implementation
#include "rapidxml.hpp"
// Only include streams if not disabled
#ifndef RAPIDXML_NO_STREAMS
#include <ostream>
#include <iterator>
#endif
namespace rapidxml
{
///////////////////////////////////////////////////////////////////////
// Printing flags
const int print_no_indenting = 0x1; //!< Printer flag instructing the printer to suppress indenting of XML. See print() function.
///////////////////////////////////////////////////////////////////////
// Internal
//! \cond internal
namespace internal
{
///////////////////////////////////////////////////////////////////////////
// Internal character operations
// Copy characters from given range to given output iterator
template<class OutIt, class Ch>
inline OutIt copy_chars(const Ch *begin, const Ch *end, OutIt out)
{
while (begin != end)
*out++ = *begin++;
return out;
}
// Copy characters from given range to given output iterator and expand
// characters into references (&lt; &gt; &apos; &quot; &amp;)
template<class OutIt, class Ch>
inline OutIt copy_and_expand_chars(const Ch *begin, const Ch *end, Ch noexpand, OutIt out)
{
while (begin != end)
{
if (*begin == noexpand)
{
*out++ = *begin; // No expansion, copy character
}
else
{
switch (*begin)
{
case Ch('<'):
*out++ = Ch('&'); *out++ = Ch('l'); *out++ = Ch('t'); *out++ = Ch(';');
break;
case Ch('>'):
*out++ = Ch('&'); *out++ = Ch('g'); *out++ = Ch('t'); *out++ = Ch(';');
break;
case Ch('\''):
*out++ = Ch('&'); *out++ = Ch('a'); *out++ = Ch('p'); *out++ = Ch('o'); *out++ = Ch('s'); *out++ = Ch(';');
break;
case Ch('"'):
*out++ = Ch('&'); *out++ = Ch('q'); *out++ = Ch('u'); *out++ = Ch('o'); *out++ = Ch('t'); *out++ = Ch(';');
break;
case Ch('&'):
*out++ = Ch('&'); *out++ = Ch('a'); *out++ = Ch('m'); *out++ = Ch('p'); *out++ = Ch(';');
break;
default:
*out++ = *begin; // No expansion, copy character
}
}
++begin; // Step to next character
}
return out;
}
// Fill given output iterator with repetitions of the same character
template<class OutIt, class Ch>
inline OutIt fill_chars(OutIt out, int n, Ch ch)
{
for (int i = 0; i < n; ++i)
*out++ = ch;
return out;
}
// Find character
template<class Ch, Ch ch>
inline bool find_char(const Ch *begin, const Ch *end)
{
while (begin != end)
if (*begin++ == ch)
return true;
return false;
}
///////////////////////////////////////////////////////////////////////////
// Internal printing operations
// Print node
template<class OutIt, class Ch>
inline OutIt print_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
// Print proper node type
switch (node->type())
{
// Document
case node_document:
out = print_children(out, node, flags, indent);
break;
// Element
case node_element:
out = print_element_node(out, node, flags, indent);
break;
// Data
case node_data:
out = print_data_node(out, node, flags, indent);
break;
// CDATA
case node_cdata:
out = print_cdata_node(out, node, flags, indent);
break;
// Declaration
case node_declaration:
out = print_declaration_node(out, node, flags, indent);
break;
// Comment
case node_comment:
out = print_comment_node(out, node, flags, indent);
break;
// Doctype
case node_doctype:
out = print_doctype_node(out, node, flags, indent);
break;
// Pi
case node_pi:
out = print_pi_node(out, node, flags, indent);
break;
// Unknown
default:
assert(0);
break;
}
// If indenting not disabled, add line break after node
if (!(flags & print_no_indenting))
*out = Ch('\n'), ++out;
// Return modified iterator
return out;
}
// Print children of the node
template<class OutIt, class Ch>
inline OutIt print_children(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
for (xml_node<Ch> *child = node->first_node(); child; child = child->next_sibling())
out = print_node(out, child, flags, indent);
return out;
}
// Print attributes of the node
template<class OutIt, class Ch>
inline OutIt print_attributes(OutIt out, const xml_node<Ch> *node, int flags)
{
for (xml_attribute<Ch> *attribute = node->first_attribute(); attribute; attribute = attribute->next_attribute())
{
if (attribute->name() && attribute->value())
{
// Print attribute name
*out = Ch(' '), ++out;
out = copy_chars(attribute->name(), attribute->name() + attribute->name_size(), out);
*out = Ch('='), ++out;
// Print attribute value using appropriate quote type
if (find_char<Ch, Ch('"')>(attribute->value(), attribute->value() + attribute->value_size()))
{
*out = Ch('\''), ++out;
out = copy_and_expand_chars(attribute->value(), attribute->value() + attribute->value_size(), Ch('"'), out);
*out = Ch('\''), ++out;
}
else
{
*out = Ch('"'), ++out;
out = copy_and_expand_chars(attribute->value(), attribute->value() + attribute->value_size(), Ch('\''), out);
*out = Ch('"'), ++out;
}
}
}
return out;
}
// Print data node
template<class OutIt, class Ch>
inline OutIt print_data_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_data);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
out = copy_and_expand_chars(node->value(), node->value() + node->value_size(), Ch(0), out);
return out;
}
// Print data node
template<class OutIt, class Ch>
inline OutIt print_cdata_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_cdata);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'); ++out;
*out = Ch('!'); ++out;
*out = Ch('['); ++out;
*out = Ch('C'); ++out;
*out = Ch('D'); ++out;
*out = Ch('A'); ++out;
*out = Ch('T'); ++out;
*out = Ch('A'); ++out;
*out = Ch('['); ++out;
out = copy_chars(node->value(), node->value() + node->value_size(), out);
*out = Ch(']'); ++out;
*out = Ch(']'); ++out;
*out = Ch('>'); ++out;
return out;
}
// Print element node
template<class OutIt, class Ch>
inline OutIt print_element_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_element);
// Print element name and attributes, if any
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'), ++out;
out = copy_chars(node->name(), node->name() + node->name_size(), out);
out = print_attributes(out, node, flags);
// If node is childless
if (node->value_size() == 0 && !node->first_node())
{
// Print childless node tag ending
*out = Ch('/'), ++out;
*out = Ch('>'), ++out;
}
else
{
// Print normal node tag ending
*out = Ch('>'), ++out;
// Test if node contains a single data node only (and no other nodes)
xml_node<Ch> *child = node->first_node();
if (!child)
{
// If node has no children, only print its value without indenting
out = copy_and_expand_chars(node->value(), node->value() + node->value_size(), Ch(0), out);
}
else if (child->next_sibling() == 0 && child->type() == node_data)
{
// If node has a sole data child, only print its value without indenting
out = copy_and_expand_chars(child->value(), child->value() + child->value_size(), Ch(0), out);
}
else
{
// Print all children with full indenting
if (!(flags & print_no_indenting))
*out = Ch('\n'), ++out;
out = print_children(out, node, flags, indent + 1);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
}
// Print node end
*out = Ch('<'), ++out;
*out = Ch('/'), ++out;
out = copy_chars(node->name(), node->name() + node->name_size(), out);
*out = Ch('>'), ++out;
}
return out;
}
// Print declaration node
template<class OutIt, class Ch>
inline OutIt print_declaration_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
// Print declaration start
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'), ++out;
*out = Ch('?'), ++out;
*out = Ch('x'), ++out;
*out = Ch('m'), ++out;
*out = Ch('l'), ++out;
// Print attributes
out = print_attributes(out, node, flags);
// Print declaration end
*out = Ch('?'), ++out;
*out = Ch('>'), ++out;
return out;
}
// Print comment node
template<class OutIt, class Ch>
inline OutIt print_comment_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_comment);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'), ++out;
*out = Ch('!'), ++out;
*out = Ch('-'), ++out;
*out = Ch('-'), ++out;
out = copy_chars(node->value(), node->value() + node->value_size(), out);
*out = Ch('-'), ++out;
*out = Ch('-'), ++out;
*out = Ch('>'), ++out;
return out;
}
// Print doctype node
template<class OutIt, class Ch>
inline OutIt print_doctype_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_doctype);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'), ++out;
*out = Ch('!'), ++out;
*out = Ch('D'), ++out;
*out = Ch('O'), ++out;
*out = Ch('C'), ++out;
*out = Ch('T'), ++out;
*out = Ch('Y'), ++out;
*out = Ch('P'), ++out;
*out = Ch('E'), ++out;
*out = Ch(' '), ++out;
out = copy_chars(node->value(), node->value() + node->value_size(), out);
*out = Ch('>'), ++out;
return out;
}
// Print pi node
template<class OutIt, class Ch>
inline OutIt print_pi_node(OutIt out, const xml_node<Ch> *node, int flags, int indent)
{
assert(node->type() == node_pi);
if (!(flags & print_no_indenting))
out = fill_chars(out, indent, Ch('\t'));
*out = Ch('<'), ++out;
*out = Ch('?'), ++out;
out = copy_chars(node->name(), node->name() + node->name_size(), out);
*out = Ch(' '), ++out;
out = copy_chars(node->value(), node->value() + node->value_size(), out);
*out = Ch('?'), ++out;
*out = Ch('>'), ++out;
return out;
}
}
//! \endcond
///////////////////////////////////////////////////////////////////////////
// Printing
//! Prints XML to given output iterator.
//! \param out Output iterator to print to.
//! \param node Node to be printed. Pass xml_document to print entire document.
//! \param flags Flags controlling how XML is printed.
//! \return Output iterator pointing to position immediately after last character of printed text.
template<class OutIt, class Ch>
inline OutIt print(OutIt out, const xml_node<Ch> &node, int flags = 0)
{
return internal::print_node(out, &node, flags, 0);
}
#ifndef RAPIDXML_NO_STREAMS
//! Prints XML to given output stream.
//! \param out Output stream to print to.
//! \param node Node to be printed. Pass xml_document to print entire document.
//! \param flags Flags controlling how XML is printed.
//! \return Output stream.
template<class Ch>
inline std::basic_ostream<Ch> &print(std::basic_ostream<Ch> &out, const xml_node<Ch> &node, int flags = 0)
{
print(std::ostream_iterator<Ch>(out), node, flags);
return out;
}
//! Prints formatted XML to given output stream. Uses default printing flags. Use print() function to customize printing process.
//! \param out Output stream to print to.
//! \param node Node to be printed.
//! \return Output stream.
template<class Ch>
inline std::basic_ostream<Ch> &operator <<(std::basic_ostream<Ch> &out, const xml_node<Ch> &node)
{
return print(out, node);
}
#endif
}
#endif

122
ext/rapidxml/rapidxml/rapidxml_utils.hpp
View File

@@ -1,122 +0,0 @@
#ifndef RAPIDXML_UTILS_HPP_INCLUDED
#define RAPIDXML_UTILS_HPP_INCLUDED
// Copyright (C) 2006, 2009 Marcin Kalicinski
// Version 1.13
// Revision $DateTime: 2009/05/13 01:46:17 $
//! \file rapidxml_utils.hpp This file contains high-level rapidxml utilities that can be useful
//! in certain simple scenarios. They should probably not be used if maximizing performance is the main objective.
#include "rapidxml.hpp"
#include <vector>
#include <string>
#include <fstream>
#include <stdexcept>
namespace rapidxml
{
//! Represents data loaded from a file
template<class Ch = char>
class file
{
public:
//! Loads file into the memory. Data will be automatically destroyed by the destructor.
//! \param filename Filename to load.
file(const char *filename)
{
using namespace std;
// Open stream
basic_ifstream<Ch> stream(filename, ios::binary);
if (!stream)
throw runtime_error(string("cannot open file ") + filename);
stream.unsetf(ios::skipws);
// Determine stream size
stream.seekg(0, ios::end);
size_t size = stream.tellg();
stream.seekg(0);
// Load data and add terminating 0
m_data.resize(size + 1);
stream.read(&m_data.front(), static_cast<streamsize>(size));
m_data[size] = 0;
}
//! Loads file into the memory. Data will be automatically destroyed by the destructor
//! \param stream Stream to load from
file(std::basic_istream<Ch> &stream)
{
using namespace std;
// Load data and add terminating 0
stream.unsetf(ios::skipws);
m_data.assign(istreambuf_iterator<Ch>(stream), istreambuf_iterator<Ch>());
if (stream.fail() || stream.bad())
throw runtime_error("error reading stream");
m_data.push_back(0);
}
//! Gets file data.
//! \return Pointer to data of file.
Ch *data()
{
return &m_data.front();
}
//! Gets file data.
//! \return Pointer to data of file.
const Ch *data() const
{
return &m_data.front();
}
//! Gets file data size.
//! \return Size of file data, in characters.
std::size_t size() const
{
return m_data.size();
}
private:
std::vector<Ch> m_data; // File data
};
//! Counts children of node. Time complexity is O(n).
//! \return Number of children of node
template<class Ch>
inline std::size_t count_children(xml_node<Ch> *node)
{
xml_node<Ch> *child = node->first_node();
std::size_t count = 0;
while (child)
{
++count;
child = child->next_sibling();
}
return count;
}
//! Counts attributes of node. Time complexity is O(n).
//! \return Number of attributes of node
template<class Ch>
inline std::size_t count_attributes(xml_node<Ch> *node)
{
xml_attribute<Ch> *attr = node->first_attribute();
std::size_t count = 0;
while (attr)
{
++count;
attr = attr->next_attribute();
}
return count;
}
}
#endif

52
ext/tmxlite/CMakeLists.txt Normal file
View File

@@ -0,0 +1,52 @@
project(tmxlite VERSION 1.3.1)
add_library(${PROJECT_NAME} STATIC
include/tmxlite/Config.hpp
include/tmxlite/FreeFuncs.hpp
include/tmxlite/ImageLayer.hpp
include/tmxlite/Layer.hpp
include/tmxlite/LayerGroup.hpp
include/tmxlite/Map.hpp
include/tmxlite/Object.hpp
include/tmxlite/ObjectGroup.hpp
include/tmxlite/ObjectTypes.hpp
include/tmxlite/Property.hpp
include/tmxlite/TileLayer.hpp
include/tmxlite/Tileset.hpp
include/tmxlite/Types.hpp
include/tmxlite/Types.inl
include/tmxlite/detail/Log.hpp
src/FreeFuncs.cpp
src/ImageLayer.cpp
src/Map.cpp
src/Object.cpp
src/ObjectGroup.cpp
src/Property.cpp
src/TileLayer.cpp
src/LayerGroup.cpp
src/Tileset.cpp
src/ObjectTypes.cpp)
if (NOT USE_ZLIB)
target_sources(${PROJECT_NAME} PRIVATE
include/tmxlite/detail/gzip.hpp
src/detail/gzip.cpp)
endif()
set_target_properties(${PROJECT_NAME} PROPERTIES
CXX_STANDARD 20
CXX_STANDARD_REQUIRED ON)
target_include_directories(${PROJECT_NAME} PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/include)
target_compile_options(${PROJECT_NAME} PRIVATE -Wall)
target_compile_definitions(${PROJECT_NAME} PRIVATE
$<$<BOOL:${MSVC}>:_CRT_SECURE_NO_WARNINGS> # disable msvc warning
$<$<TARGET_EXISTS:ZLIB::ZLIB>:USE_ZLIB>)
target_link_libraries(${PROJECT_NAME} base64::base64 pugixml Zstd::Zstd
$<$<TARGET_EXISTS:ZLIB::ZLIB>:ZLIB::ZLIB>
$<$<TARGET_EXISTS:miniz::miniz>:miniz::miniz>)

13
ext/tmxlite/LICENSE Normal file
View File

@@ -0,0 +1,13 @@
(c)Matt Marchant & contributors 2016 - 2021 http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or implied warranty.
In no event will the authors be held liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose, including commercial applications,
and to alter it and redistribute it freely, subject to the following restrictions:
The origin of this software must not be misrepresented; you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment in the product documentation would be appreciated but is not required.
Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
This notice may not be removed or altered from any source distribution.

31
ext/tmxlite/include/tmxlite/Config.hpp Normal file
View File

@@ -0,0 +1,31 @@
/*********************************************************************
(c) Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
//static build doesn't need import/export macros
#define TMXLITE_EXPORT_API

126
ext/tmxlite/include/tmxlite/FreeFuncs.hpp Normal file
View File

@@ -0,0 +1,126 @@
/*********************************************************************
Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/detail/Log.hpp"
#include "tmxlite/Types.hpp"
#include <string>
#include <sstream>
#include <vector>
#include <functional>
#include <algorithm>
namespace tmx
{
//using inline here just to supress unused warnings on gcc
bool decompress(const char* source, std::vector<unsigned char>& dest, std::size_t inSize, std::size_t expectedSize);
static inline Colour colourFromString(std::string str)
{
//removes preceding #
auto result = str.find_last_of('#');
if (result != std::string::npos)
{
str = str.substr(result + 1);
}
if (str.size() == 6 || str.size() == 8)
{
unsigned int value, r, g, b;
unsigned int a = 255;
std::stringstream input(str);
input >> std::hex >> value;
r = (value >> 16) & 0xff;
g = (value >> 8) & 0xff;
b = value & 0xff;
if (str.size() == 8)
{
a = (value >> 24) & 0xff;
}
return{ std::uint8_t(r), std::uint8_t(g), std::uint8_t(b), std::uint8_t(a) };
}
Logger::log(str + ": not a valid colour string", Logger::Type::Error);
return{};
}
static inline std::string resolveFilePath(std::string path, const std::string& workingDir)
{
static const std::string match("../");
std::size_t result = path.find(match);
std::size_t count = 0;
while (result != std::string::npos)
{
count++;
path = path.substr(result + match.size());
result = path.find(match);
}
if (workingDir.empty()) return path;
std::string outPath = workingDir;
for (auto i = 0u; i < count; ++i)
{
result = outPath.find_last_of('/');
if (result != std::string::npos)
{
outPath = outPath.substr(0, result);
}
}
return outPath + '/' + path;
}
static inline std::string getFilePath(const std::string& path)
{
//TODO this doesn't actually check that there is a file at the
//end of the path, or that it's even a valid path...
static auto searchFunc = [](const char separator, const std::string& path)->std::string
{
std::size_t i = path.rfind(separator, path.length());
if (i != std::string::npos)
{
return(path.substr(0, i + 1));
}
return "";
};
#ifdef _WIN32 //try windows formatted paths first
std::string retVal = searchFunc('\\', path);
if (!retVal.empty()) return retVal;
#endif
return searchFunc('/', path);
}
} //namespacec tmx

107
ext/tmxlite/include/tmxlite/ImageLayer.hpp Normal file
View File

@@ -0,0 +1,107 @@
/*********************************************************************
Matt Marchant 2016 - 2022
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Layer.hpp"
#include "tmxlite/Types.hpp"
namespace tmx
{
/*!
\brief Image layers contain a single image which make up that
layer. The parser contains the fully resolved path to the image
relative to the working directory.
*/
class TMXLITE_EXPORT_API ImageLayer final : public Layer
{
public:
explicit ImageLayer(const std::string&);
Type getType() const override { return Layer::Type::Image; }
void parse(const pugi::xml_node&, Map*) override;
/*!
\brief Returns the path, relative to the working directory,
of the image used by the image layer.
*/
const std::string& getImagePath() const { return m_filePath; }
/*!
\brief Returns the colour used by the image to represent transparent
pixels. By default this is (0, 0, 0, 0)
*/
const Colour& getTransparencyColour() const { return m_transparencyColour; }
/*!
\brief Returns true if the image used by this layer specifically states a
colour to use as transparency
*/
bool hasTransparency() const { return m_hasTransparency; }
/*!
\brief Returns the size of the image of the image layer in pixels.
*/
const Vector2u& getImageSize() const { return m_imageSize; }
/*!
\brief Returns true if the image drawn by this layer is repeated along
the X axis.
*/
bool hasRepeatX() const { return m_hasRepeatX; }
/*!
\brief Returns true if the image drawn by this layer is repeated along
the Y axis.
*/
bool hasRepeatY() const { return m_hasRepeatY; }
private:
std::string m_workingDir;
std::string m_filePath;
Colour m_transparencyColour;
bool m_hasTransparency;
Vector2u m_imageSize;
bool m_hasRepeatX;
bool m_hasRepeatY;
};
template <>
inline ImageLayer& Layer::getLayerAs<ImageLayer>()
{
assert(getType() == Type::Image);
return *static_cast<ImageLayer*>(this);
}
template <>
inline const ImageLayer& Layer::getLayerAs<ImageLayer>() const
{
assert(getType() == Type::Image);
return *static_cast<const ImageLayer*>(this);
}
}

175
ext/tmxlite/include/tmxlite/Layer.hpp Normal file
View File

@@ -0,0 +1,175 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Property.hpp"
#include "tmxlite/Types.hpp"
#include <string>
#include <memory>
#include <vector>
namespace pugi
{
class xml_node;
}
namespace tmx
{
class Map;
class TileLayer;
class ObjectGroup;
class ImageLayer;
class LayerGroup;
/*!
\brief Represents a layer of a tmx format tile map.
This is an abstract base class from which all layer
types are derived.
*/
class TMXLITE_EXPORT_API Layer
{
public:
using Ptr = std::unique_ptr<Layer>;
Layer() : m_opacity(1.f), m_visible(true) {};
virtual ~Layer() = default;
/*!
\brief Layer type as returned by getType()
Tile: this layer is a TileLayer type
Object: This layer is an ObjectGroup type
Image: This layer is an ImageLayer type
Group: This layer is a LayerGroup type
*/
enum class Type
{
Tile,
Object,
Image,
Group
};
/*!
\brief Returns a Type value representing the concrete type.
Use this when deciding which conrete layer type to use when
calling the templated function getLayerAs<T>()
*/
virtual Type getType() const = 0;
/*!
\brief Returns the class of the Layer, as defined in the editor Tiled 1.9+
*/
const std::string& getClass() const { return m_class; }
/*!
\brief Use this to get a reference to the concrete layer type
which this layer points to.
Use getType() to return the type value of this layer and determine
if the concrete type is TileLayer, ObjectGroup, ImageLayer, or LayerGroup
*/
template <typename T>
T& getLayerAs();
template <typename T>
const T& getLayerAs() const;
/*!
\brief Attempts to parse the specific node layer type
*/
virtual void parse(const pugi::xml_node&, Map* = nullptr) = 0;
/*!
\brief Returns the name of the layer
*/
const std::string& getName() const { return m_name; }
/*!
\brief Returns the opacity value for the layer
*/
float getOpacity() const { return m_opacity; }
/*!
\brief Returns whether this layer is visible or not
*/
bool getVisible() const { return m_visible; }
/*!
\brief Returns the offset from the top left corner
of the layer, in pixels
*/
const Vector2i& getOffset() const { return m_offset; }
/*!
\brief Returns the parallax factor
*/
const Vector2f& getParallaxFactor() const { return m_parallaxFactor; }
/*!
\brief Returns the tint colour of the layer.
Defaults to 0xFFFFFFFF - pure white
*/
Colour getTintColour() const { return m_tintColour; }
/*!
\brief Returns the size of the layer, in pixels.
This will be the same as the map size for fixed size maps.
*/
const Vector2u& getSize() const { return m_size; }
/*!
\brief Returns the list of properties of this layer
*/
const std::vector<Property>& getProperties() const { return m_properties; }
protected:
void setName(const std::string& name) { m_name = name; }
void setClass(const std::string& cls) { m_class = cls; }
void setOpacity(float opacity) { m_opacity = opacity; }
void setVisible(bool visible) { m_visible = visible; }
void setOffset(std::int32_t x, std::int32_t y) { m_offset = Vector2i(x, y); }
void setParallaxFactor(float x, float y) { m_parallaxFactor.x = x; m_parallaxFactor.y = y; }
void setTintColour(Colour c) { m_tintColour = c; }
void setSize(std::uint32_t width, std::uint32_t height) { m_size = Vector2u(width, height); }
void addProperty(const pugi::xml_node& node) { m_properties.emplace_back(); m_properties.back().parse(node); }
private:
std::string m_name;
std::string m_class;
float m_opacity;
bool m_visible;
Vector2i m_offset;
Vector2f m_parallaxFactor;
Colour m_tintColour = { 255,255,255,255 };
Vector2u m_size;
std::vector<Property> m_properties;
};
}

86
ext/tmxlite/include/tmxlite/LayerGroup.hpp Normal file
View File

@@ -0,0 +1,86 @@
/*********************************************************************
Grant Gangi 2019 - 2022
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Layer.hpp"
#include "tmxlite/Types.hpp"
#include <vector>
namespace tmx
{
/*!
\brief Layer groups are used to organize the layers of
the map in a hierarchy. They can contain all other layer
types including more layer groups to further nest layers.
*/
class TMXLITE_EXPORT_API LayerGroup final : public Layer
{
public:
LayerGroup(const std::string& workDir, const Vector2u& tileCount);
~LayerGroup() = default;
LayerGroup(const LayerGroup&) = delete;
const LayerGroup& operator = (const LayerGroup&) = delete;
LayerGroup(LayerGroup&&) = default;
LayerGroup& operator = (LayerGroup&&) = default;
Type getType() const override { return Layer::Type::Group; }
void parse(const pugi::xml_node&, Map*) override;
/*!
\brief Returns a reference to the vector containing the layer data.
Layers are pointer-to-baseclass, the concrete type of which can be
found via Layer::getType()
\see Layer
*/
const std::vector<Layer::Ptr>& getLayers() const { return m_layers; }
private:
std::vector<Layer::Ptr> m_layers;
std::string m_workingDir;
Vector2u m_tileCount;
};
template <>
inline LayerGroup& Layer::getLayerAs<LayerGroup>()
{
assert(getType() == Type::Group);
return *static_cast<LayerGroup*>(this);
}
template <>
inline const LayerGroup& Layer::getLayerAs<LayerGroup>() const
{
assert(getType() == Type::Group);
return *static_cast<const LayerGroup*>(this);
}
}

282
ext/tmxlite/include/tmxlite/Map.hpp Normal file
View File

@@ -0,0 +1,282 @@
/*********************************************************************
Matt Marchant 2016 -2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Tileset.hpp"
#include "tmxlite/Layer.hpp"
#include "tmxlite/Property.hpp"
#include "tmxlite/Types.hpp"
#include "tmxlite/Object.hpp"
#include <string>
#include <vector>
#include <map>
#include <unordered_map>
namespace tmx
{
/*!
\brief Holds the xml version of the loaded map
*/
struct TMXLITE_EXPORT_API Version
{
//major/minor are apparently reserved by gcc
std::uint16_t upper;
std::uint16_t lower;
Version(std::uint16_t maj = 0, std::uint16_t min = 0)
: upper(maj), lower(min) {}
};
enum class Orientation
{
Orthogonal,
Isometric,
Staggered,
Hexagonal,
None
};
enum class RenderOrder
{
RightDown,
RightUp,
LeftDown,
LeftUp,
None
};
enum class StaggerAxis
{
X, Y, None
};
enum class StaggerIndex
{
Even, Odd, None
};
/*!
\brief Parser for TMX format tile maps.
This class can be used to parse the XML format tile maps created
with the Tiled map editor, providing an interface to create drawable and
physics objects. Typical usage would be to create an instance of this
class before calling load() providing a path to the *.tmx file to be
loaded. Then layers or objects can be requested from the Map class
to be interpreted as needed.
\see https://doc.mapeditor.org/en/stable/reference/tmx-map-format/#map
*/
class TMXLITE_EXPORT_API Map final
{
public:
Map();
~Map() = default;
Map(const Map&) = delete;
Map& operator = (const Map&) = delete;
Map(Map&&) = default;
Map& operator = (Map&&) = default;
/*!
\brief Attempts to parse the tilemap at the given location.
\param std::string Path to map file to try to parse
\returns true if map was parsed successfully else returns false.
In debug mode this will attempt to log any errors to the console.
*/
bool load(const std::string&);
/*!
\brief Loads a map from a document stored in a string
\param data A std::string containing the map data to load
\param workingDir A std::string containing the working directory
in which to find assets such as tile sets or images
\returns true if successful, else false
*/
bool loadFromString(const std::string& data, const std::string& workingDir);
/*!
\brief Returns the version of the tile map last parsed.
If no tile map has yet been parsed the version will read 0, 0
*/
const Version& getVersion() const { return m_version; }
/*!
\brief Returns the orientation of the map if one is loaded,
else returns None
*/
Orientation getOrientation() const { return m_orientation; }
/*!
\brief Returns the RenderOrder of the map if one is loaded,
else returns None
*/
RenderOrder getRenderOrder() const { return m_renderOrder; }
/*!
\brief Returns the tile count of the map in the X and Y directions
*/
const Vector2u& getTileCount() const { return m_tileCount; }
/*!
\brief Returns the size of the tile grid in this map.
Actual tile sizes may vary and will be extended / shrunk about
the bottom left corner of the tile.
*/
const Vector2u& getTileSize() const { return m_tileSize; }
/*!
\brief Returns the bounds of the map
*/
FloatRect getBounds() const { return FloatRect(0.f, 0.f, static_cast<float>(m_tileCount.x * m_tileSize.x), static_cast<float>(m_tileCount.y * m_tileSize.y)); }
/*!
\brief Returns the length of an edge of a tile if a Hexagonal
map is loaded.
The length returned is in pixels of the straight edge running
along the axis returned by getStaggerAxis(). If no map is loaded
or the loaded map is not of Hexagonal orientation this function
returns 0.f
*/
float getHexSideLength() const { return m_hexSideLength; }
/*!
\brief Stagger axis of the map.
If either a Staggered or Hexagonal tile map is loaded this returns
which axis the map is staggered along, else returns None.
*/
StaggerAxis getStaggerAxis() const { return m_staggerAxis; }
/*!
\brief Stagger Index of the loaded map.
If a Staggered or Hexagonal map is loaded this returns whether
the even or odd rows of tiles are staggered, otherwise it returns None.
*/
StaggerIndex getStaggerIndex() const { return m_staggerIndex; }
/*!
\brief Returns the background colour of the map.
*/
const Colour& getBackgroundColour() const { return m_backgroundColour; }
/*!
\brief Returns a reference to the vector of tile sets used by the map
*/
const std::vector<Tileset>& getTilesets() const { return m_tilesets; }
/*!
\brief Returns a reference to the vector containing the layer data.
Layers are pointer-to-baseclass, the concrete type of which can be
found via Layer::getType()
\see Layer
*/
const std::vector<Layer::Ptr>& getLayers() const { return m_layers; }
/*!
\brief Returns the class of the Map, as defined in the editor Tiled 1.9+
*/
const std::string& getClass() const { return m_class; }
/*!
\brief Returns a vector of Property objects loaded by the map
*/
const std::vector<Property>& getProperties() const { return m_properties; }
/*!
\brief Returns a Hashmap of all animated tiles accessible by TileID
*/
const std::map<std::uint32_t, Tileset::Tile>& getAnimatedTiles() const { return m_animTiles; }
/*!
\brief Returns the current working directory of the map. Images and
other resources are loaded relative to this.
*/
const std::string& getWorkingDirectory() const { return m_workingDirectory; }
/*!
\brief Returns an unordered_map of template objects indexed by file name
*/
std::unordered_map<std::string, Object>& getTemplateObjects() { return m_templateObjects; }
const std::unordered_map<std::string, Object>& getTemplateObjects() const { return m_templateObjects; }
/*!
\brief Returns an unordered_map of tilesets used by templated objects.
If Object::getTilesetName() is not empty it can be used to retreive a tileset
from this map. Otherwise the object's tileset can be found from in the map's
global tilesets returned by getTilesets().
*/
std::unordered_map<std::string, Tileset>& getTemplateTilesets() { return m_templateTilesets; }
const std::unordered_map<std::string, Tileset>& getTemplateTilesets() const { return m_templateTilesets; }
/*!
\brief Returns true if this is in infinite tile map.
Infinite maps store their tile data in for tile layers in chunks. If
this is an infinite map use TileLayer::getChunks() to get tile IDs
rather than TileLayer::getTiles().
\see TileLayer
*/
bool isInfinite() const { return m_infinite; }
/*
\brief Returns the origin of each layer's parallax offset value
*/
Vector2f getParallaxOrigin() const { return m_parallaxOrigin; }
private:
Version m_version;
std::string m_class;
Orientation m_orientation;
RenderOrder m_renderOrder;
bool m_infinite;
Vector2u m_tileCount;
Vector2u m_tileSize;
float m_hexSideLength;
StaggerAxis m_staggerAxis;
StaggerIndex m_staggerIndex;
Vector2f m_parallaxOrigin;
Colour m_backgroundColour;
std::string m_workingDirectory;
std::vector<Tileset> m_tilesets;
std::vector<Layer::Ptr> m_layers;
std::vector<Property> m_properties;
std::map<std::uint32_t, Tileset::Tile> m_animTiles;
std::unordered_map<std::string, Object> m_templateObjects;
std::unordered_map<std::string, Tileset> m_templateTilesets;
bool parseMapNode(const pugi::xml_node&);
//always returns false so we can return this
//on load failure
bool reset();
};
}

221
ext/tmxlite/include/tmxlite/Object.hpp Normal file
View File

@@ -0,0 +1,221 @@
/*********************************************************************
(c) Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Property.hpp"
#include "tmxlite/Types.hpp"
#include <string>
#include <vector>
namespace pugi
{
class xml_node;
}
namespace tmx
{
class Map;
/*!
\brief Contains the text information stored in a Text object.
*/
struct TMXLITE_EXPORT_API Text final
{
std::string fontFamily;
std::uint32_t pixelSize = 16; //!< pixels, not points
bool wrap = false;
Colour colour;
bool bold = false;
bool italic = false;
bool underline = false;
bool strikethough = false;
bool kerning = true;
enum class HAlign
{
Left, Centre, Right
}hAlign = HAlign::Left;
enum class VAlign
{
Top, Centre, Bottom
}vAlign = VAlign::Top;
std::string content; //!< actual string content
};
/*!
\brief Objects are stored in ObjectGroup layers.
Objects may be rectangular, elliptical, polygonal or
a polyline. Rectangular and elliptical Objects have their
size determined via the AABB, whereas polygon and polyline
shapes are defined by a list of points. Objects are
rectangular by default. Since version 1.0 Objects also
support Text nodes.
*/
class TMXLITE_EXPORT_API Object final
{
public:
enum class Shape
{
Rectangle,
Ellipse,
Point,
Polygon,
Polyline,
Text
};
Object();
/*!
\brief Attempts to parse the given xml node and
read the Object properties if it is valid.
*/
void parse(const pugi::xml_node&, Map*);
/*!
\brief Returns the unique ID of the Object
*/
std::uint32_t getUID() const { return m_UID; }
/*!
\brief Returns the name of the Object
*/
const std::string& getName() const { return m_name; }
/*!
\brief Returns the type (equal to class) of the Object, as defined in the editor Tiled < 1.9
*/
const std::string& getType() const { return m_class; }
/*!
\brief Returns the class (equal to type) of the Object, as defined in the editor Tiled 1.9+
*/
const std::string& getClass() const { return m_class; }
/*!
\brief Returns the position of the Object in pixels
*/
const Vector2f& getPosition() const { return m_position; }
/*!
\brief Returns the global Axis Aligned Bounding Box.
The AABB is positioned via the left and top properties, and
define the Object's width and height. This can be used to derive
the shape of the Object if it is rectangular or elliptical.
*/
const FloatRect& getAABB() const { return m_AABB; }
/*!
\brief Returns the rotation of the Object in degrees clockwise
*/
float getRotation() const { return m_rotation; }
/*!
\brief Returns the global tile ID associated with the Object
if there is one. This is used to draw the Object (and therefore
the Object must be rectangular)
*/
std::uint32_t getTileID() const { return m_tileID; }
/*!
\brief Returns the flip flags if the objects uses a TileID to
draw it.
Returns 0 otherwise.
*/
std::uint8_t getFlipFlags() const { return m_flipFlags; }
/*!
\brief Returns whether or not the Object is visible
*/
bool visible() const { return m_visible; }
/*!
\brief Returns the Shape type of the Object
*/
Shape getShape() const { return m_shape; }
/*!
\brief Returns a reference to the vector of points which
make up the Object. If the Object is rectangular or elliptical
then the vector will be empty. Point coordinates are in pixels,
relative to the object position.
*/
const std::vector<Vector2f>& getPoints() const { return m_points; }
/*!
\brief Returns a reference to the vector of properties belonging to
the Object.
*/
const std::vector<Property>& getProperties() const { return m_properties; }
/*!
\brief Returns a Text struct containing information about any text
this object may have, such as font data and formatting.
If an object does not contain any text information this struct will
be populated with default values. Use getShape() to determine
if this object is in fact a text object.
*/
const Text& getText() const { return m_textData; }
Text& getText() { return m_textData; }
/*!
\brief Returns the tileset name used by this object if it is derived
from a template, else returns an empty string.
If the string is not empty use it to index the unordered_map returned
by Map::getTemplateTilesets()
*/
const std::string& getTilesetName() const { return m_tilesetName; }
private:
std::uint32_t m_UID;
std::string m_name;
std::string m_class;
Vector2f m_position;
FloatRect m_AABB;
float m_rotation;
std::uint32_t m_tileID;
std::uint8_t m_flipFlags;
bool m_visible;
Shape m_shape;
std::vector<Vector2f> m_points;
std::vector<Property> m_properties;
Text m_textData;
std::string m_tilesetName;
void parsePoints(const pugi::xml_node&);
void parseText(const pugi::xml_node&);
void parseTemplate(const std::string&, Map*);
};
}

99
ext/tmxlite/include/tmxlite/ObjectGroup.hpp Normal file
View File

@@ -0,0 +1,99 @@
/*********************************************************************
Matt Marchant 2016 - 2022
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Layer.hpp"
#include "tmxlite/Object.hpp"
#include <vector>
namespace tmx
{
/*!
\brief ObjectGroup layers contain a series of Objects
which may be made up of shapes or images.
*/
class TMXLITE_EXPORT_API ObjectGroup final : public Layer
{
public:
enum class DrawOrder
{
Index, //< objects should be drawn in the order in which they appear
TopDown //< objects should be drawn sorted by their Y position
};
ObjectGroup();
Type getType() const override { return Layer::Type::Object; }
void parse(const pugi::xml_node&, Map*) override;
/*!
\brief Returns the colour associated with this layer
*/
const Colour& getColour() const { return m_colour; }
/*!
\brief Returns the DrawOrder for the objects in this group.
Defaults to TopDown, where Objects are drawn sorted by Y position
*/
DrawOrder getDrawOrder() const { return m_drawOrder; }
/*!
\brief Returns a reference to the vector of properties for
the ObjectGroup
*/
const std::vector<Property>& getProperties() const { return m_properties; }
/*!
\brief Returns a reference to the vector of Objects which belong to the group
*/
const std::vector<Object>& getObjects() const { return m_objects; }
private:
Colour m_colour;
DrawOrder m_drawOrder;
std::vector<Property> m_properties;
std::vector<Object> m_objects;
};
template <>
inline ObjectGroup& Layer::getLayerAs<ObjectGroup>()
{
assert(getType() == Type::Object);
return *static_cast<ObjectGroup*>(this);
}
template <>
inline const ObjectGroup& Layer::getLayerAs<ObjectGroup>() const
{
assert(getType() == Type::Object);
return *static_cast<const ObjectGroup*>(this);
}
}

86
ext/tmxlite/include/tmxlite/ObjectTypes.hpp Normal file
View File

@@ -0,0 +1,86 @@
/*********************************************************************
Raphaël Frantz 2021
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Property.hpp"
#include <string>
#include <vector>
namespace tmx
{
/*!
\brief Parser for Tiled object types export format.
Link to the specification: https://doc.mapeditor.org/fr/latest/manual/custom-properties/#predefining-properties.
*/
class TMXLITE_EXPORT_API ObjectTypes final
{
public:
/*!
\brief Types that stores all predefined properties for all objects of this type.
To take less spaces, they are not exported by default into maps.
*/
struct Type
{
std::string name;
Colour colour;
std::vector<Property> properties;
};
/*!
\brief Attempts to parse the object types at the given location.
\param std::string Path to object types file to try to parse
\returns true if object types was parsed successfully else returns false.
In debug mode this will attempt to log any errors to the console.
*/
bool load(const std::string&);
/*!
\brief Loads an object types from a document stored in a string
\param data A std::string containing the object types to load
\param workingDir A std::string containing the working directory
in which to find files.
\returns true if successful, else false
*/
bool loadFromString(const std::string& data, const std::string& workingDir);
/*!
\brief Returns all predefined types and their default values.
*/
const std::vector<Type>& getTypes() const { return m_types; }
private:
std::string m_workingDirectory;
std::vector<Type> m_types;
bool parseObjectTypesNode(const pugi::xml_node&);
//always returns false so we can return this
//on load failure
bool reset();
};
}

144
ext/tmxlite/include/tmxlite/Property.hpp Normal file
View File

@@ -0,0 +1,144 @@
/*********************************************************************
Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Types.hpp"
#include <string>
#include <cassert>
namespace pugi
{
class xml_node;
}
namespace tmx
{
/*!
\brief Represents a custom property.
Tiles, objects and layers of a tmx map may have custom
properties assigned to them. This class represents a
single property and provides access to its value, the
type of which can be determined with getType()
*/
class TMXLITE_EXPORT_API Property final
{
public:
enum class Type
{
Boolean,
Float,
Int,
String,
Colour,
File,
Object,
Undef
};
Property();
static Property fromBoolean(bool value);
static Property fromFloat(float value);
static Property fromInt(int value);
static Property fromString(const std::string& value);
static Property fromColour(const Colour& value);
static Property fromFile(const std::string& value);
static Property fromObject(int value);
/*!
\brief Attempts to parse the given node as a property
\param isObjectTypes Set to true if the parsing is done from an object types files.
*/
void parse(const pugi::xml_node&, bool isObjectTypes = false);
/*!
\brief Returns the type of data stored in the property.
This should generally be called first before trying to
read the proprty value, as reading the incorrect type
will lead to undefined behaviour.
*/
Type getType() const { return m_type; }
/*!
\brief Returns the name of this property
*/
const std::string& getName() const { return m_name; }
/*!
\brief Returns the property's value as a boolean
*/
bool getBoolValue() const { assert(m_type == Type::Boolean); return m_boolValue; }
/*!
\brief Returns the property's value as a float
*/
float getFloatValue() const { assert(m_type == Type::Float); return m_floatValue; }
/*!
\brief Returns the property's value as an integer
*/
int getIntValue() const { assert(m_type == Type::Int || m_type == Type::Object); return m_intValue; }
/*!
\brief Returns the property's value as a string
*/
const std::string& getStringValue() const { assert(m_type == Type::String); return m_stringValue; }
/*!
\brief Returns the property's value as a Colour struct
*/
const Colour& getColourValue() const { assert(m_type == Type::Colour); return m_colourValue; }
/*!
\brief Returns the file path property as a string, relative to the map file
*/
const std::string& getFileValue() const { assert(m_type == Type::File); return m_stringValue; }
/*!
\brief Returns the property's value as an integer object handle
*/
int getObjectValue() const { assert(m_type == Type::Object); return m_intValue; }
private:
union
{
bool m_boolValue;
float m_floatValue;
int m_intValue;
};
std::string m_stringValue;
std::string m_name;
Colour m_colourValue;
Type m_type;
};
}

116
ext/tmxlite/include/tmxlite/TileLayer.hpp Normal file
View File

@@ -0,0 +1,116 @@
/*********************************************************************
Matt Marchant 2016 - 2022
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Layer.hpp"
#include "tmxlite/Types.hpp"
namespace tmx
{
/*!
\brief A layer made up from a series of tile sets
*/
class TMXLITE_EXPORT_API TileLayer final : public Layer
{
public:
/*!
\brief Tile information for a layer
*/
struct Tile final
{
std::uint32_t ID = 0; //!< Global ID of the tile
std::uint8_t flipFlags = 0; //!< Flags marking if the tile should be flipped when drawn
};
/*!
\brief Represents a chunk of tile data, if this is an infinite map
*/
struct Chunk final
{
Vector2i position; //<! coordinate in tiles, not pixels
Vector2i size; //!< size in tiles, not pixels
std::vector<Tile> tiles;
};
/*!
\brief Flags used to tell if a tile is flipped when drawn
*/
enum FlipFlag
{
Horizontal = 0x8,
Vertical = 0x4,
Diagonal = 0x2
};
explicit TileLayer(std::size_t);
Type getType() const override { return Layer::Type::Tile; }
void parse(const pugi::xml_node&, Map*) override;
/*!
\brief Returns the list of tiles used to make up the layer
If this is empty then the map is most likely infinite, in
which case the tile data is stored in chunks.
\see getChunks()
*/
const std::vector<Tile>& getTiles() const { return m_tiles; }
/*!
\brief Returns a vector of chunks which make up this layer
if the map is set to infinite. This will be empty if the map
is not infinite.
\see getTiles()
*/
const std::vector<Chunk>& getChunks() const { return m_chunks; }
private:
std::vector<Tile> m_tiles;
std::vector<Chunk> m_chunks;
std::size_t m_tileCount;
void parseBase64(const pugi::xml_node&);
void parseCSV(const pugi::xml_node&);
void parseUnencoded(const pugi::xml_node&);
void createTiles(const std::vector<std::uint32_t>&, std::vector<Tile>& destination);
};
template <>
inline TileLayer& Layer::getLayerAs<TileLayer>()
{
assert(getType() == Type::Tile);
return *static_cast<TileLayer*>(this);
}
template <>
inline const TileLayer& Layer::getLayerAs<TileLayer>() const
{
assert(getType() == Type::Tile);
return *static_cast<const TileLayer*>(this);
}
}

296
ext/tmxlite/include/tmxlite/Tileset.hpp Normal file
View File

@@ -0,0 +1,296 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include "tmxlite/Property.hpp"
#include "tmxlite/ObjectGroup.hpp"
#include <string>
#include <vector>
#include <array>
namespace pugi
{
class xml_node;
}
namespace tmx
{
class Map;
/*!
\brief Represents a Tileset node as loaded
from a *.tmx format tile map via the tmx::Map
class.
*/
class TMXLITE_EXPORT_API Tileset final
{
public:
explicit Tileset(const std::string& workingDir);
/*!
\brief Any tiles within a tile set which have special
data associated with them such as animation or terrain
information will have one of these stored in the tile set.
*/
struct Tile final
{
std::uint32_t ID = 0;
std::array<std::int32_t, 4u> terrainIndices{};
std::uint32_t probability = 100;
/*!
\brief a group of frames which make up an animation
*/
struct Animation final
{
/*!
\brief A frame within an animation
*/
struct Frame final
{
std::uint32_t tileID = 0;
std::uint32_t duration = 0;
bool operator == (const Frame& other) const
{
return (this == &other) ||
(tileID == other.tileID && duration == other.duration);
}
bool operator != (const Frame& other) const
{
return !(*this == other);
}
};
std::vector<Frame> frames;
}animation;
std::vector<Property> properties;
ObjectGroup objectGroup;
std::string imagePath;
Vector2u imageSize;
/*!
\brief The position of the tile within the image.
*/
Vector2u imagePosition;
std::string className;
};
/*!
\brief Terrain information with which one
or more tiles may be associated.
*/
struct Terrain final
{
std::string name;
std::uint32_t tileID = -1;
std::vector<Property> properties;
};
/*!
\brief Declares the alignment of tile Objects
*/
enum class ObjectAlignment
{
Unspecified,
TopLeft,
Top,
TopRight,
Left,
Center,
Right,
BottomLeft,
Bottom,
BottomRight
};
/*!
\brief Attempts to parse the given xml node.
If node parsing fails an error is printed in the console
and the Tileset remains in an uninitialised state.
*/
void parse(pugi::xml_node, Map*);
/*!
\brief Returns the first GID of this tile set.
This the ID of the first tile in the tile set, so that
each tile set guarantees a unique set of IDs
*/
std::uint32_t getFirstGID() const { return m_firstGID; }
/*!
\brief Returns the last GID of this tile set.
This is the ID of the last tile in the tile set.
*/
std::uint32_t getLastGID() const;
/*!
\brief Returns the name of this tile set.
*/
const std::string& getName() const { return m_name; }
/*!
\brief Returns the class of the Tileset, as defined in the editor Tiled 1.9+
*/
const std::string& getClass() const { return m_class; }
/*!
\brief Returns the width and height of a tile in the
tile set, in pixels.
*/
const Vector2u& getTileSize() const { return m_tileSize; }
/*!
\brief Returns the spacing, in pixels, between each tile in the set
*/
std::uint32_t getSpacing() const { return m_spacing; }
/*!
\brief Returns the margin, in pixels, around each tile in the set
*/
std::uint32_t getMargin() const { return m_margin; }
/*!
\brief Returns the number of tiles in the tile set
*/
std::uint32_t getTileCount() const { return m_tileCount; }
/*!
\brief Returns the number of columns which make up the tile set.
This is used when rendering collection of images sets
*/
std::uint32_t getColumnCount() const { return m_columnCount; }
/*!
\brief Returns the alignment of tile objects.
The default value is ObjectAlignment::Unspecified for compatibility.
When the alignment is Unspecified tile objects use BottomLeft in
orthogonal mode and Bottom in isometric mode.
\see ObjectAlignment
*/
ObjectAlignment getObjectAlignment() const { return m_objectAlignment; }
/*!
\brief Returns the tile offset in pixels.
Tile will draw tiles offset from the top left using this value.
*/
const Vector2u& getTileOffset() const { return m_tileOffset; }
/*!
\brief Returns a reference to the list of Property objects for this
tile set
*/
const std::vector<Property>& getProperties() const { return m_properties; }
/*!
\brief Returns the file path to the tile set image, relative to the
working directory. Use this to load the texture required by whichever
method you choose to render the map.
*/
const std::string& getImagePath() const { return m_imagePath; }
/*!
\brief Returns the size of the tile set image in pixels.
*/
const Vector2u& getImageSize() const { return m_imageSize; }
/*!
\brief Returns the colour used by the tile map image to represent transparency.
By default this is a transparent colour (0, 0, 0, 0)
*/
const Colour& getTransparencyColour() const { return m_transparencyColour; }
/*!
\brief Returns true if the image used by this tileset specifically requests
a colour to use as transparency.
*/
bool hasTransparency() const { return m_hasTransparency; }
/*!
\brief Returns a vector of Terrain types associated with one
or more tiles within this tile set
*/
const std::vector<Terrain>& getTerrainTypes() const { return m_terrainTypes; }
/*!
\brief Returns a reference to the vector of tile data used by
tiles which make up this tile set.
*/
const std::vector<Tile>& getTiles() const { return m_tiles; }
/*!
\brief Checks if a tiled ID is in the range of the first ID and the last ID
\param id Tile ID
\return
*/
bool hasTile(std::uint32_t id) const { return id >= m_firstGID && id <= getLastGID(); };
/*!
\brief queries tiles and returns a tile with the given ID. Checks if the TileID is part of the Tileset with `hasTile(id)`
\param id Tile ID. The Tile ID will be corrected internally.
\return In case of a success it returns the correct tile. In terms of failure it will return a nullptr.
*/
const Tile* getTile(std::uint32_t id) const;
private:
std::string m_workingDir;
std::uint32_t m_firstGID;
std::string m_source;
std::string m_name;
std::string m_class;
Vector2u m_tileSize;
std::uint32_t m_spacing;
std::uint32_t m_margin;
std::uint32_t m_tileCount;
std::uint32_t m_columnCount;
ObjectAlignment m_objectAlignment;
Vector2u m_tileOffset;
std::vector<Property> m_properties;
std::string m_imagePath;
Vector2u m_imageSize;
Colour m_transparencyColour;
bool m_hasTransparency;
std::vector<Terrain> m_terrainTypes;
std::vector<std::uint32_t> m_tileIndex;
std::vector<Tile> m_tiles;
void reset();
void parseOffsetNode(const pugi::xml_node&);
void parsePropertyNode(const pugi::xml_node&);
void parseTerrainNode(const pugi::xml_node&);
Tile& newTile(std::uint32_t ID);
void parseTileNode(const pugi::xml_node&, Map*);
void createMissingTile(std::uint32_t ID);
};
}

150
ext/tmxlite/include/tmxlite/Types.hpp Normal file
View File

@@ -0,0 +1,150 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#pragma once
#include "tmxlite/Config.hpp"
#include <cstdint>
#include <ostream>
namespace tmx
{
/*!
\brief Two dimensional vector used to store points and positions
*/
template <class T>
struct Vector2 final
{
Vector2() : x(0), y(0) {}
Vector2(T x, T y) :x(x), y(y) {}
T x, y;
};
using Vector2f = Vector2<float>;
using Vector2i = Vector2<int>;
using Vector2u = Vector2<unsigned>;
template <typename T>
Vector2<T> operator + (const Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T>& operator += (Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T> operator - (const Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T>& operator -= (Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T> operator * (const Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T>& operator *= (Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T> operator * (const Vector2<T>& l, T r);
template <typename T>
Vector2<T>& operator *= (Vector2<T>& l, T r);
template <typename T>
Vector2<T> operator / (const Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T>& operator /= (Vector2<T>& l, const Vector2<T>& r);
template <typename T>
Vector2<T> operator / (const Vector2<T>& l, T r);
template <typename T>
Vector2<T>& operator /= (Vector2<T>& l, T r);
#include "Types.inl"
/*!
\brief Describes a rectangular area, such as an AABB (axis aligned bounding box)
*/
template <class T>
struct Rectangle final
{
Rectangle() : left(0), top(0), width(0), height(0) {}
Rectangle(T l, T t, T w, T h) : left(l), top(t), width(w), height(h) {}
Rectangle(Vector2<T> position, Vector2<T> size) : left(position.x), top(position.y), width(size.x), height(size.y) {}
T left, top, width, height;
};
using FloatRect = Rectangle<float>;
using IntRect = Rectangle<int>;
/*!
\brief Contains the red, green, blue and alpha values of a colour
in the range 0 - 255.
*/
struct TMXLITE_EXPORT_API Colour final
{
Colour(std::uint8_t red = 0, std::uint8_t green = 0, std::uint8_t blue = 0, std::uint8_t alpha = 255)
: r(red), g(green), b(blue), a(alpha) {}
std::uint8_t r, g, b, a;
bool operator == (const Colour& other)
{
return other.r == r
&& other.g == g
&& other.b == b
&& other.a == a;
}
bool operator != (const Colour& other)
{
return !(*this == other);
}
explicit operator std::uint32_t() const
{
return (r << 24) | (g << 16) | (b << 8) | a;
}
};
}
template <typename T>
std::ostream& operator << (std::ostream& os, const tmx::Vector2<T>& t)
{
os << "{" << t.x << ", " << t.y << "}";
return os;
}
template <typename T>
std::ostream& operator << (std::ostream& os, const tmx::Rectangle<T>& t)
{
os << "{" << t.left << ", " << t.top << ", " << t.width << ", " << t.height << "}";
return os;
}
std::ostream& operator << (std::ostream& os, const tmx::Colour& c);

110
ext/tmxlite/include/tmxlite/Types.inl Normal file
View File

@@ -0,0 +1,110 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
template <typename T>
Vector2<T> operator + (const Vector2<T>& l, const Vector2<T>& r)
{
return { l.x + r.x, l.y + r.y };
}
template <typename T>
Vector2<T>& operator += (Vector2<T>& l, const Vector2<T>& r)
{
l.x += r.x;
l.y += r.y;
return l;
}
template <typename T>
Vector2<T> operator - (const Vector2<T>& l, const Vector2<T>& r)
{
return { l.x - r.x, l.y - r.y };
}
template <typename T>
Vector2<T>& operator -= (Vector2<T>& l, const Vector2<T>& r)
{
l.x -= r.x;
l.y -= r.y;
return l;
}
template <typename T>
Vector2<T> operator * (const Vector2<T>& l, const Vector2<T>& r)
{
return { l.x * r.x, l.y * r.y };
}
template <typename T>
Vector2<T>& operator *= (Vector2<T>& l, const Vector2<T>& r)
{
l.x *= r.x;
l.y *= r.y;
return l;
}
template <typename T>
Vector2<T> operator * (const Vector2<T>& l, T r)
{
return { l.x * r, l.y * r };
}
template <typename T>
Vector2<T>& operator *= (Vector2<T>& l, T r)
{
l.x *= r;
l.y *= r;
return l;
}
template <typename T>
Vector2<T> operator / (const Vector2<T>& l, const Vector2<T>& r)
{
return { l.x / r.x, l.y / r.y };
}
template <typename T>
Vector2<T>& operator /= (Vector2<T>& l, const Vector2<T>& r)
{
l.x /= r.x;
l.y /= r.y;
return l;
}
template <typename T>
Vector2<T> operator / (const Vector2<T>& l, T r)
{
return { l.x / r, l.y / r };
}
template <typename T>
Vector2<T>& operator /= (Vector2<T>& l, T r)
{
l.x /= r;
l.y /= r;
return l;
}

163
ext/tmxlite/include/tmxlite/detail/Log.hpp Normal file
View File

@@ -0,0 +1,163 @@
/*********************************************************************
Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
//flexible logging class, based on code at https://github.com/fallahn/xygine
#ifndef TMXLITE_LOGGER_HPP_
#define TMXLITE_LOGGER_HPP_
#include <string>
#include <iostream>
#include <iomanip>
#include <fstream>
#include <sstream>
#include <list>
#include <ctime>
#ifdef _MSC_VER
#define NOMINMAX
#include <windows.h>
#endif //_MSC_VER
namespace tmx
{
/*!
\brief Class allowing messages to be logged to a combination
of one or more destinations such as the console, log file or
output window in Visual Studio
*/
class Logger final
{
public:
enum class Output
{
Console,
File,
All
};
enum class Type
{
Info,
Warning,
Error
};
/*!
\brief Logs a message to a given destination.
\param message Message to log
\param type Whether this message gets tagged as information, a warning or an error
\param output Destination for the message. Can be the console via cout, a log file on disk, or both
*/
static void log(const std::string& message, Type type = Type::Info, Output output = Output::Console)
{
std::string outstring;
switch (type)
{
case Type::Info:
default:
outstring = "INFO: " + message;
break;
case Type::Error:
outstring = "ERROR: " + message;
break;
case Type::Warning:
outstring = "WARNING: " + message;
break;
}
if (output == Output::Console || output == Output::All)
{
if (type == Type::Error)
{
std::cerr << outstring << std::endl;
}
else
{
std::cout << outstring << std::endl;
}
const std::size_t maxBuffer = 30;
buffer().push_back(outstring);
if (buffer().size() > maxBuffer)buffer().pop_front(); //no majick here pl0x
updateOutString(maxBuffer);
#ifdef _MSC_VER
outstring += "\n";
OutputDebugStringA(outstring.c_str());
#endif //_MSC_VER
}
if (output == Output::File || output == Output::All)
{
//output to a log file
std::ofstream file("output.log", std::ios::app);
if (file.good())
{
std::time_t time = std::time(nullptr);
auto tm = *std::localtime(&time);
//put_time isn't implemented by the ndk versions of the stl
file.imbue(std::locale());
file << std::put_time(&tm, "%d/%m/%y-%H:%M:%S: ");
file << outstring << std::endl;
file.close();
}
else
{
log(message, type, Output::Console);
log("Above message was intended for log file. Opening file probably failed.", Type::Warning, Output::Console);
}
}
}
static const std::string& bufferString(){ return stringOutput(); }
private:
static std::list<std::string>& buffer(){ static std::list<std::string> buffer; return buffer; }
static std::string& stringOutput() { static std::string output; return output; }
static void updateOutString(std::size_t maxBuffer)
{
static size_t count = 0;
stringOutput().append(buffer().back());
stringOutput().append("\n");
count++;
if (count > maxBuffer)
{
stringOutput() = stringOutput().substr(stringOutput().find_first_of('\n') + 1, stringOutput().size());
count--;
}
}
};
}
#ifdef NDEBUG
#define LOG(message, type)
#else
#define LOG(message, type) {\
std::stringstream ss; \
ss << message << " (" << __FILE__ << ", " << __LINE__ << ")"; \
tmx::Logger::log(ss.str(), type);}
#endif //NDEBUG
#endif //TMXLITE_LOGGER_HPP_

39
ext/tmxlite/include/tmxlite/detail/gzip.hpp Normal file
View File

@@ -0,0 +1,39 @@
// gzip.hpp - portable memory miniz based gzip reader
// SPDX-License-Identifier: Zlib
// SPDX-FileCopyrightText: (c) 2024 a dinosaur
#ifndef GZIP_HPP
#define GZIP_HPP
#include "miniz.h"
#include <cstdint>
#include <span>
class GZipReader
{
static constexpr uint8_t
FTEXT = 1, FHCRC = 1<<1, FEXTRA = 1<<2, FNAME = 1<<3, FCOMMENT = 1<<4;
static constexpr uint8_t XFL_BEST = 2, XFL_FASTEST = 4;
tinfl_decompressor mState;
std::span<const uint8_t>::iterator mIt;
size_t mSourceLen, mBytesRead;
uint32_t mModificationTime, mCrc, mInputSize, mComputedCrc;
uint16_t crc16;
uint8_t mFlags, mXflags, mOsId;
public:
GZipReader() noexcept;
constexpr size_t SourceLength() const noexcept { return mSourceLen; }
constexpr uint32_t OutputLength() const noexcept { return mInputSize; }
bool OpenMemory(const std::span<const uint8_t> source) noexcept;
bool Read(std::span<uint8_t> out) noexcept;
bool Check() const noexcept;
};
#endif//GZIP_HPP

132
ext/tmxlite/src/FreeFuncs.cpp Normal file
View File

@@ -0,0 +1,132 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/Types.hpp"
#include "tmxlite/detail/Log.hpp"
#ifndef USE_ZLIB
# include "miniz.h"
#else
# include <zlib.h>
#endif
#include <cstring>
bool tmx::decompress(const char* source, std::vector<unsigned char>& dest, std::size_t inSize, std::size_t expectedSize)
{
if (!source)
{
LOG("Input string is empty, decompression failed.", Logger::Type::Error);
return false;
}
//#ifdef USE_EXTLIBS
//#else
int currentSize = static_cast<int>(expectedSize);
std::vector<unsigned char> byteArray(expectedSize / sizeof(unsigned char));
z_stream stream;
stream.zalloc = Z_NULL;
stream.zfree = Z_NULL;
stream.opaque = Z_NULL;
stream.next_in = (Bytef*)source;
stream.avail_in = static_cast<unsigned int>(inSize);
stream.next_out = (Bytef*)byteArray.data();
stream.avail_out = static_cast<unsigned int>(expectedSize);
//we'd prefer to use inflateInit2 but it appears
//to be incorrect in miniz. This is fine for zlib
//compressed data, but gzip compressed streams
//will fail to inflate.
#ifdef USE_ZLIB
if (inflateInit2(&stream, 15 + 32) != Z_OK)
#else
if (inflateInit(&stream) != Z_OK)
#endif
{
LOG("inflate init failed", Logger::Type::Error);
return false;
}
int result = 0;
do
{
result = inflate(&stream, Z_SYNC_FLUSH);
switch (result)
{
default: break;
case Z_NEED_DICT:
case Z_STREAM_ERROR:
result = Z_DATA_ERROR;
case Z_DATA_ERROR:
Logger::log("If using gzip or zstd compression try using zlib instead", Logger::Type::Info);
case Z_MEM_ERROR:
inflateEnd(&stream);
Logger::log("inflate() returned " + std::to_string(result), Logger::Type::Error);
return false;
}
if (result != Z_STREAM_END)
{
int oldSize = currentSize;
currentSize *= 2;
std::vector<unsigned char> newArray(currentSize / sizeof(unsigned char));
std::memcpy(newArray.data(), byteArray.data(), currentSize / 2);
byteArray = std::move(newArray);
stream.next_out = (Bytef*)(byteArray.data() + oldSize);
stream.avail_out = oldSize;
}
} while (result != Z_STREAM_END);
if (stream.avail_in != 0)
{
LOG("stream.avail_in is 0", Logger::Type::Error);
LOG("zlib decompression failed.", Logger::Type::Error);
return false;
}
const int outSize = currentSize - stream.avail_out;
inflateEnd(&stream);
std::vector<unsigned char> newArray(outSize / sizeof(unsigned char));
std::memcpy(newArray.data(), byteArray.data(), outSize);
byteArray = std::move(newArray);
//copy bytes to vector
dest.insert(dest.begin(), byteArray.begin(), byteArray.end());
//#endif
return true;
}
std::ostream& operator << (std::ostream& os, const tmx::Colour& c)
{
os << "RGBA: " << (int)c.r << ", " << (int)c.g << ", " << (int)c.b << ", " << (int)c.a;
return os;
}

107
ext/tmxlite/src/ImageLayer.cpp Normal file
View File

@@ -0,0 +1,107 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/ImageLayer.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
using namespace tmx;
ImageLayer::ImageLayer(const std::string& workingDir)
: m_workingDir (workingDir),
m_hasTransparency (false),
m_hasRepeatX (false),
m_hasRepeatY (false)
{
}
//public
void ImageLayer::parse(const pugi::xml_node& node, Map*)
{
std::string attribName = node.name();
if (attribName != "imagelayer")
{
Logger::log("Node not an image layer, node skipped", Logger::Type::Error);
return;
}
//TODO this gets repeated foreach layer type and could all be moved to base class...
setName(node.attribute("name").as_string());
setClass(node.attribute("class").as_string());
setOpacity(node.attribute("opacity").as_float(1.f));
setVisible(node.attribute("visible").as_bool(true));
setOffset(node.attribute("offsetx").as_int(0), node.attribute("offsety").as_int(0));
setSize(node.attribute("width").as_uint(0), node.attribute("height").as_uint(0));
setParallaxFactor(node.attribute("parallaxx").as_float(1.f), node.attribute("parallaxy").as_float(1.f));
std::string tintColour = node.attribute("tintcolor").as_string();
if (!tintColour.empty())
{
setTintColour(colourFromString(tintColour));
}
m_hasRepeatX = node.attribute("repeatx").as_bool(false);
m_hasRepeatY = node.attribute("repeaty").as_bool(false);
for (const auto& child : node.children())
{
attribName = child.name();
if (attribName == "image")
{
attribName = child.attribute("source").as_string();
if (attribName.empty())
{
Logger::log("Image Layer has missing source property", Logger::Type::Warning);
return;
}
if (child.attribute("width") && child.attribute("height"))
{
m_imageSize.x = child.attribute("width").as_uint();
m_imageSize.y = child.attribute("height").as_uint();
}
m_filePath = resolveFilePath(attribName, m_workingDir);
if (child.attribute("trans"))
{
attribName = child.attribute("trans").as_string();
m_transparencyColour = colourFromString(attribName);
m_hasTransparency = true;
}
}
else if (attribName == "properties")
{
for (const auto& p : child.children())
{
addProperty(p);
}
}
}
}

106
ext/tmxlite/src/LayerGroup.cpp Normal file
View File

@@ -0,0 +1,106 @@
/*********************************************************************
Grant Gangi 2019
Matt Marchant 2023
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/LayerGroup.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/ObjectGroup.hpp"
#include "tmxlite/ImageLayer.hpp"
#include "tmxlite/TileLayer.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
using namespace tmx;
LayerGroup::LayerGroup(const std::string& workingDir, const Vector2u& tileCount)
: m_workingDir(workingDir),
m_tileCount(tileCount)
{
}
//public
void LayerGroup::parse(const pugi::xml_node& node, Map* map)
{
assert(map);
std::string attribString = node.name();
if (attribString != "group")
{
Logger::log("Node was not a group layer, node will be skipped.", Logger::Type::Error);
return;
}
setName(node.attribute("name").as_string());
setClass(node.attribute("class").as_string());
setOpacity(node.attribute("opacity").as_float(1.f));
setVisible(node.attribute("visible").as_bool(true));
setOffset(node.attribute("offsetx").as_int(0), node.attribute("offsety").as_int(0));
setSize(node.attribute("width").as_uint(0), node.attribute("height").as_uint(0));
setParallaxFactor(node.attribute("parallaxx").as_float(1.f), node.attribute("parallaxy").as_float(1.f));
std::string tintColour = node.attribute("tintcolor").as_string();
if (!tintColour.empty())
{
setTintColour(colourFromString(tintColour));
}
// parse children
for (const auto& child : node.children())
{
attribString = child.name();
if (attribString == "properties")
{
for (const auto& p : child.children())
{
addProperty(p);
}
}
else if (attribString == "layer")
{
m_layers.emplace_back(std::make_unique<TileLayer>(m_tileCount.x * m_tileCount.y));
m_layers.back()->parse(child, map);
}
else if (attribString == "objectgroup")
{
m_layers.emplace_back(std::make_unique<ObjectGroup>());
m_layers.back()->parse(child, map);
}
else if (attribString == "imagelayer")
{
m_layers.emplace_back(std::make_unique<ImageLayer>(m_workingDir));
m_layers.back()->parse(child, map);
}
else if (attribString == "group")
{
m_layers.emplace_back(std::make_unique<LayerGroup>(m_workingDir, m_tileCount));
m_layers.back()->parse(child, map);
}
else
{
LOG("Unidentified name " + attribString + ": node skipped", Logger::Type::Warning);
}
}
}

362
ext/tmxlite/src/Map.cpp Normal file
View File

@@ -0,0 +1,362 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/Map.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/ObjectGroup.hpp"
#include "tmxlite/ImageLayer.hpp"
#include "tmxlite/TileLayer.hpp"
#include "tmxlite/LayerGroup.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
#include <queue>
using namespace tmx;
Map::Map()
: m_orientation (Orientation::None),
m_renderOrder (RenderOrder::None),
m_infinite (false),
m_hexSideLength (0.f),
m_staggerAxis (StaggerAxis::None),
m_staggerIndex (StaggerIndex::None)
{
}
//public
bool Map::load(const std::string& path)
{
reset();
//open the doc
pugi::xml_document doc;
auto result = doc.load_file(path.c_str());
if (!result)
{
Logger::log("Failed opening " + path, Logger::Type::Error);
Logger::log("Reason: " + std::string(result.description()), Logger::Type::Error);
return false;
}
//make sure we have consistent path separators
m_workingDirectory = path;
std::replace(m_workingDirectory.begin(), m_workingDirectory.end(), '\\', '/');
m_workingDirectory = getFilePath(m_workingDirectory);
if (!m_workingDirectory.empty() &&
m_workingDirectory.back() == '/')
{
m_workingDirectory.pop_back();
}
//find the map node and bail if it doesn't exist
auto mapNode = doc.child("map");
if (!mapNode)
{
Logger::log("Failed opening map: " + path + ", no map node found", Logger::Type::Error);
return reset();
}
return parseMapNode(mapNode);
}
bool Map::loadFromString(const std::string& data, const std::string& workingDir)
{
reset();
//open the doc
pugi::xml_document doc;
auto result = doc.load_string(data.c_str());
if (!result)
{
Logger::log("Failed opening map", Logger::Type::Error);
Logger::log("Reason: " + std::string(result.description()), Logger::Type::Error);
return false;
}
//make sure we have consistent path separators
m_workingDirectory = workingDir;
std::replace(m_workingDirectory.begin(), m_workingDirectory.end(), '\\', '/');
m_workingDirectory = getFilePath(m_workingDirectory);
if (!m_workingDirectory.empty() &&
m_workingDirectory.back() == '/')
{
m_workingDirectory.pop_back();
}
//find the map node and bail if it doesn't exist
auto mapNode = doc.child("map");
if (!mapNode)
{
Logger::log("Failed opening map: no map node found", Logger::Type::Error);
return reset();
}
return parseMapNode(mapNode);
}
//private
bool Map::parseMapNode(const pugi::xml_node& mapNode)
{
//parse map attributes
std::size_t pointPos = 0;
std::string attribString = mapNode.attribute("version").as_string();
if (attribString.empty() || (pointPos = attribString.find('.')) == std::string::npos)
{
Logger::log("Invalid map version value, map not loaded.", Logger::Type::Error);
return reset();
}
m_version.upper = std::stoi(attribString.substr(0, pointPos));
m_version.lower = std::stoi(attribString.substr(pointPos + 1));
m_class = mapNode.attribute("class").as_string();
attribString = mapNode.attribute("orientation").as_string();
if (attribString.empty())
{
Logger::log("Missing map orientation attribute, map not loaded.", Logger::Type::Error);
return reset();
}
if (attribString == "orthogonal")
{
m_orientation = Orientation::Orthogonal;
}
else if (attribString == "isometric")
{
m_orientation = Orientation::Isometric;
}
else if (attribString == "staggered")
{
m_orientation = Orientation::Staggered;
}
else if (attribString == "hexagonal")
{
m_orientation = Orientation::Hexagonal;
}
else
{
Logger::log(attribString + " format maps aren't supported yet, sorry! Map not loaded", Logger::Type::Error);
return reset();
}
attribString = mapNode.attribute("renderorder").as_string();
//this property is optional for older version of map files
if (!attribString.empty())
{
if (attribString == "right-down")
{
m_renderOrder = RenderOrder::RightDown;
}
else if (attribString == "right-up")
{
m_renderOrder = RenderOrder::RightUp;
}
else if (attribString == "left-down")
{
m_renderOrder = RenderOrder::LeftDown;
}
else if (attribString == "left-up")
{
m_renderOrder = RenderOrder::LeftUp;
}
else
{
Logger::log(attribString + ": invalid render order. Map not loaded.", Logger::Type::Error);
return reset();
}
}
if (mapNode.attribute("infinite"))
{
m_infinite = mapNode.attribute("infinite").as_int() != 0;
}
unsigned width = mapNode.attribute("width").as_int();
unsigned height = mapNode.attribute("height").as_int();
if (width && height)
{
m_tileCount = { width, height };
}
else
{
Logger::log("Invalid map tile count, map not loaded", Logger::Type::Error);
return reset();
}
width = mapNode.attribute("tilewidth").as_int();
height = mapNode.attribute("tileheight").as_int();
if (width && height)
{
m_tileSize = { width, height };
}
else
{
Logger::log("Invalid tile size, map not loaded", Logger::Type::Error);
return reset();
}
m_hexSideLength = mapNode.attribute("hexsidelength").as_float();
if (m_orientation == Orientation::Hexagonal && m_hexSideLength <= 0)
{
Logger::log("Invalid he side length found, map not loaded", Logger::Type::Error);
return reset();
}
attribString = mapNode.attribute("staggeraxis").as_string();
if (attribString == "x")
{
m_staggerAxis = StaggerAxis::X;
}
else if (attribString == "y")
{
m_staggerAxis = StaggerAxis::Y;
}
if ((m_orientation == Orientation::Staggered || m_orientation == Orientation::Hexagonal)
&& m_staggerAxis == StaggerAxis::None)
{
Logger::log("Map missing stagger axis property. Map not loaded.", Logger::Type::Error);
return reset();
}
attribString = mapNode.attribute("staggerindex").as_string();
if (attribString == "odd")
{
m_staggerIndex = StaggerIndex::Odd;
}
else if (attribString == "even")
{
m_staggerIndex = StaggerIndex::Even;
}
if ((m_orientation == Orientation::Staggered || m_orientation == Orientation::Hexagonal)
&& m_staggerIndex == StaggerIndex::None)
{
Logger::log("Map missing stagger index property. Map not loaded.", Logger::Type::Error);
return reset();
}
m_parallaxOrigin =
{
mapNode.attribute("parallaxoriginx").as_float(0.f),
mapNode.attribute("parallaxoriginy").as_float(0.f)
};
//colour property is optional
attribString = mapNode.attribute("backgroundcolor").as_string();
if (!attribString.empty())
{
m_backgroundColour = colourFromString(attribString);
}
//TODO do we need next object ID
//parse all child nodes
for (const auto& node : mapNode.children())
{
std::string name = node.name();
if (name == "tileset")
{
m_tilesets.emplace_back(m_workingDirectory);
m_tilesets.back().parse(node, this);
}
else if (name == "layer")
{
m_layers.emplace_back(std::make_unique<TileLayer>(m_tileCount.x * m_tileCount.y));
m_layers.back()->parse(node);
}
else if (name == "objectgroup")
{
m_layers.emplace_back(std::make_unique<ObjectGroup>());
m_layers.back()->parse(node, this);
}
else if (name == "imagelayer")
{
m_layers.emplace_back(std::make_unique<ImageLayer>(m_workingDirectory));
m_layers.back()->parse(node, this);
}
else if (name == "properties")
{
const auto& children = node.children();
for (const auto& child : children)
{
m_properties.emplace_back();
m_properties.back().parse(child);
}
}
else if (name == "group")
{
m_layers.emplace_back(std::make_unique<LayerGroup>(m_workingDirectory, m_tileCount));
m_layers.back()->parse(node, this);
}
else
{
LOG("Unidentified name " + name + ": node skipped", Logger::Type::Warning);
}
}
// fill animated tiles for easier lookup into map
for(const auto& ts : m_tilesets)
{
for(const auto& tile : ts.getTiles())
{
if (!tile.animation.frames.empty())
{
m_animTiles[tile.ID + ts.getFirstGID()] = tile;
}
}
}
return true;
}
bool Map::reset()
{
m_orientation = Orientation::None;
m_renderOrder = RenderOrder::None;
m_tileCount = { 0u, 0u };
m_tileSize = { 0u, 0u };
m_hexSideLength = 0.f;
m_staggerAxis = StaggerAxis::None;
m_staggerIndex = StaggerIndex::None;
m_backgroundColour = {};
m_workingDirectory = "";
m_tilesets.clear();
m_layers.clear();
m_properties.clear();
m_templateObjects.clear();
m_templateTilesets.clear();
m_animTiles.clear();
return false;
}

399
ext/tmxlite/src/Object.cpp Normal file
View File

@@ -0,0 +1,399 @@
/*********************************************************************
Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/Object.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/Map.hpp"
#include "tmxlite/Tileset.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
#include <sstream>
using namespace tmx;
Object::Object()
: m_UID (0),
m_rotation (0.f),
m_tileID (0),
m_flipFlags (0),
m_visible (true),
m_shape (Shape::Rectangle)
{
}
//public
void Object::parse(const pugi::xml_node& node, Map* map)
{
std::string attribString = node.name();
if (attribString != "object")
{
Logger::log("This not an Object node, parsing skipped.", Logger::Type::Error);
return;
}
m_UID = node.attribute("id").as_int();
m_name = node.attribute("name").as_string();
m_class = node.attribute("type").as_string();
if (m_class.empty())
{
m_class = node.attribute("class").as_string();
}
m_position.x = node.attribute("x").as_float();
m_AABB.left = m_position.x;
m_position.y = node.attribute("y").as_float();
m_AABB.top = m_position.y;
m_AABB.width = node.attribute("width").as_float();
m_AABB.height = node.attribute("height").as_float();
m_rotation = node.attribute("rotation").as_float();
m_visible = node.attribute("visible").as_bool(true);
m_tileID = node.attribute("gid").as_uint();
static const std::uint32_t mask = 0xf0000000;
m_flipFlags = ((m_tileID & mask) >> 28);
m_tileID = m_tileID & ~mask;
for (const auto& child : node.children())
{
attribString = child.name();
if (attribString == "properties")
{
for (const auto& p : child.children())
{
m_properties.emplace_back();
m_properties.back().parse(p);
}
}
else if (attribString == "ellipse")
{
m_shape = Shape::Ellipse;
}
else if (attribString == "point")
{
m_shape = Shape::Point;
}
else if (attribString == "polygon")
{
m_shape = Shape::Polygon;
parsePoints(child);
}
else if (attribString == "polyline")
{
m_shape = Shape::Polyline;
parsePoints(child);
}
else if (attribString == "text")
{
m_shape = Shape::Text;
parseText(child);
}
}
//parse templates last so we know which properties
//ought to be overridden
std::string templateStr = node.attribute("template").as_string();
if (!templateStr.empty() && map)
{
parseTemplate(templateStr, map);
}
}
//private
void Object::parsePoints(const pugi::xml_node& node)
{
if (node.attribute("points"))
{
std::string pointlist = node.attribute("points").as_string();
std::stringstream stream(pointlist);
std::vector<std::string> points;
std::string pointstring;
while (std::getline(stream, pointstring, ' '))
{
points.push_back(pointstring);
}
//parse each pair into sf::vector2f
for (unsigned int i = 0; i < points.size(); i++)
{
std::vector<float> coords;
std::stringstream coordstream(points[i]);
float j;
while (coordstream >> j)
{
coords.push_back(j);
//TODO this should really ignore anything non-numeric
if (coordstream.peek() == ',')
{
coordstream.ignore();
}
}
m_points.emplace_back(coords[0], coords[1]);
}
}
else
{
Logger::log("Points for polygon or polyline object are missing", Logger::Type::Warning);
}
}
void Object::parseText(const pugi::xml_node& node)
{
m_textData.bold = node.attribute("bold").as_bool(false);
m_textData.colour = colourFromString(node.attribute("color").as_string("#FFFFFFFF"));
m_textData.fontFamily = node.attribute("fontfamily").as_string();
m_textData.italic = node.attribute("italic").as_bool(false);
m_textData.kerning = node.attribute("kerning").as_bool(true);
m_textData.pixelSize = node.attribute("pixelsize").as_uint(16);
m_textData.strikethough = node.attribute("strikeout").as_bool(false);
m_textData.underline = node.attribute("underline").as_bool(false);
m_textData.wrap = node.attribute("wrap").as_bool(false);
std::string alignment = node.attribute("halign").as_string("left");
if (alignment == "left")
{
m_textData.hAlign = Text::HAlign::Left;
}
else if (alignment == "center")
{
m_textData.hAlign = Text::HAlign::Centre;
}
else if (alignment == "right")
{
m_textData.hAlign = Text::HAlign::Right;
}
alignment = node.attribute("valign").as_string("top");
if (alignment == "top")
{
m_textData.vAlign = Text::VAlign::Top;
}
else if (alignment == "center")
{
m_textData.vAlign = Text::VAlign::Centre;
}
else if (alignment == "bottom")
{
m_textData.vAlign = Text::VAlign::Bottom;
}
m_textData.content = node.text().as_string();
}
void Object::parseTemplate(const std::string& path, Map* map)
{
assert(map);
auto& templateObjects = map->getTemplateObjects();
auto& templateTilesets = map->getTemplateTilesets();
//load the template if not already loaded
if (templateObjects.count(path) == 0)
{
auto templatePath = map->getWorkingDirectory() + "/" + path;
pugi::xml_document doc;
if (!doc.load_file(templatePath.c_str()))
{
Logger::log("Failed opening template file " + path, Logger::Type::Error);
return;
}
auto templateNode = doc.child("template");
if (!templateNode)
{
Logger::log("Template node missing from " + path, Logger::Type::Error);
return;
}
//if the template has a tileset load that (if not already loaded)
std::string tilesetName;
auto tileset = templateNode.child("tileset");
if (tileset)
{
tilesetName = tileset.attribute("source").as_string();
if (!tilesetName.empty() &&
templateTilesets.count(tilesetName) == 0)
{
templateTilesets.insert(std::make_pair(tilesetName, Tileset(map->getWorkingDirectory())));
templateTilesets.at(tilesetName).parse(tileset, map);
}
}
//parse the object - don't pass the map pointer here so there's
//no recursion if someone tried to get clever and put a template in a template
auto obj = templateNode.child("object");
if (obj)
{
templateObjects.insert(std::make_pair(path, Object()));
templateObjects[path].parse(obj, nullptr);
templateObjects[path].m_tilesetName = tilesetName;
}
}
//apply any non-overridden object properties from the template
if (templateObjects.count(path) != 0)
{
const auto& obj = templateObjects[path];
if (m_AABB.width == 0)
{
m_AABB.width = obj.m_AABB.width;
}
if (m_AABB.height == 0)
{
m_AABB.height = obj.m_AABB.height;
}
m_tilesetName = obj.m_tilesetName;
if (m_name.empty())
{
m_name = obj.m_name;
}
if (m_class.empty())
{
m_class = obj.m_class;
}
if (m_rotation == 0)
{
m_rotation = obj.m_rotation;
}
if (m_tileID == 0)
{
m_tileID = obj.m_tileID;
}
if (m_flipFlags == 0)
{
m_flipFlags = obj.m_flipFlags;
}
if (m_shape == Shape::Rectangle)
{
m_shape = obj.m_shape;
}
if (m_points.empty())
{
m_points = obj.m_points;
}
//compare properties and only copy ones that don't exist
for (const auto& p : obj.m_properties)
{
auto result = std::find_if(m_properties.begin(), m_properties.end(),
[&p](const Property& a)
{
return a.getName() == p.getName();
});
if (result == m_properties.end())
{
m_properties.push_back(p);
}
}
if (m_shape == Shape::Text)
{
//check each text property and update as necessary
//TODO this makes he assumption we prefer the template
//properties over the default ones - this might not
//actually be the case....
const auto& otherText = obj.m_textData;
if (m_textData.fontFamily.empty())
{
m_textData.fontFamily = otherText.fontFamily;
}
if (m_textData.pixelSize == 16)
{
m_textData.pixelSize = otherText.pixelSize;
}
//TODO this isn't actually right if we *want* to be false
//and the template is set to true...
if (m_textData.wrap == false)
{
m_textData.wrap = otherText.wrap;
}
if (m_textData.colour == Colour())
{
m_textData.colour = otherText.colour;
}
if (m_textData.bold == false)
{
m_textData.bold = otherText.bold;
}
if (m_textData.italic == false)
{
m_textData.italic = otherText.italic;
}
if (m_textData.underline == false)
{
m_textData.underline = otherText.underline;
}
if (m_textData.strikethough == false)
{
m_textData.strikethough = otherText.strikethough;
}
if (m_textData.kerning == true)
{
m_textData.kerning = otherText.kerning;
}
if (m_textData.hAlign == Text::HAlign::Left)
{
m_textData.hAlign = otherText.hAlign;
}
if (m_textData.vAlign == Text::VAlign::Top)
{
m_textData.vAlign = otherText.vAlign;
}
if (m_textData.content.empty())
{
m_textData.content = otherText.content;
}
}
}
}

99
ext/tmxlite/src/ObjectGroup.cpp Normal file
View File

@@ -0,0 +1,99 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/ObjectGroup.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
using namespace tmx;
ObjectGroup::ObjectGroup()
: m_colour (127, 127, 127, 255),
m_drawOrder (DrawOrder::TopDown)
{
}
//public
void ObjectGroup::parse(const pugi::xml_node& node, Map* map)
{
assert(map);
std::string attribString = node.name();
if (attribString != "objectgroup")
{
Logger::log("Node was not an object group, node will be skipped.", Logger::Type::Error);
return;
}
setName(node.attribute("name").as_string());
setClass(node.attribute("class").as_string());
attribString = node.attribute("color").as_string();
if (!attribString.empty())
{
m_colour = colourFromString(attribString);
}
setOpacity(node.attribute("opacity").as_float(1.f));
setVisible(node.attribute("visible").as_bool(true));
setOffset(node.attribute("offsetx").as_int(0), node.attribute("offsety").as_int(0));
setSize(node.attribute("width").as_uint(0), node.attribute("height").as_uint(0));
setParallaxFactor(node.attribute("parallaxx").as_float(1.f), node.attribute("parallaxy").as_float(1.f));
std::string tintColour = node.attribute("tintcolor").as_string();
if (!tintColour.empty())
{
setTintColour(colourFromString(tintColour));
}
attribString = node.attribute("draworder").as_string();
if (attribString == "index")
{
m_drawOrder = DrawOrder::Index;
}
for (const auto& child : node.children())
{
attribString = child.name();
if (attribString == "properties")
{
for (const auto& p : child)
{
m_properties.emplace_back();
m_properties.back().parse(p);
}
}
else if (attribString == "object")
{
m_objects.emplace_back();
m_objects.back().parse(child, map);
}
}
}

151
ext/tmxlite/src/ObjectTypes.cpp Normal file
View File

@@ -0,0 +1,151 @@
/*********************************************************************
Raphaël Frantz 2021
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/ObjectTypes.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
using namespace tmx;
bool ObjectTypes::load(const std::string &path)
{
reset();
//open the doc
pugi::xml_document doc;
auto result = doc.load_file(path.c_str());
if (!result)
{
Logger::log("Failed opening " + path, Logger::Type::Error);
Logger::log("Reason: " + std::string(result.description()), Logger::Type::Error);
return false;
}
//make sure we have consistent path separators
m_workingDirectory = path;
std::replace(m_workingDirectory.begin(), m_workingDirectory.end(), '\\', '/');
m_workingDirectory = getFilePath(m_workingDirectory);
if (!m_workingDirectory.empty() &&
m_workingDirectory.back() == '/')
{
m_workingDirectory.pop_back();
}
//find the node and bail if it doesn't exist
auto node = doc.child("objecttypes");
if (!node)
{
Logger::log("Failed opening object types: " + path + ", no objecttype node found", Logger::Type::Error);
return reset();
}
return parseObjectTypesNode(node);
}
bool ObjectTypes::loadFromString(const std::string &data, const std::string &workingDir)
{
reset();
//open the doc
pugi::xml_document doc;
auto result = doc.load_string(data.c_str());
if (!result)
{
Logger::log("Failed opening object types", Logger::Type::Error);
Logger::log("Reason: " + std::string(result.description()), Logger::Type::Error);
return false;
}
//make sure we have consistent path separators
m_workingDirectory = workingDir;
std::replace(m_workingDirectory.begin(), m_workingDirectory.end(), '\\', '/');
m_workingDirectory = getFilePath(m_workingDirectory);
if (!m_workingDirectory.empty() &&
m_workingDirectory.back() == '/')
{
m_workingDirectory.pop_back();
}
//find the node and bail if it doesn't exist
auto node = doc.child("objecttypes");
if (!node)
{
Logger::log("Failed object types: no objecttypes node found", Logger::Type::Error);
return reset();
}
return parseObjectTypesNode(node);
}
bool ObjectTypes::parseObjectTypesNode(const pugi::xml_node &node)
{
//<objecttypes> <-- node
// <objecttype name="Character" color="#1e47ff">
// <property>...
//parse types
for(const auto& child : node.children())
{
std::string attribString = child.name();
if (attribString == "objecttype")
{
Type type;
//parse the metadata of the type
type.name = child.attribute("name").as_string();
type.colour = colourFromString(child.attribute("color").as_string("#FFFFFFFF"));;
//parse the default properties of the type
for (const auto& p : child.children())
{
Property prop;
prop.parse(p, true);
type.properties.push_back(prop);
}
m_types.push_back(type);
}
else
{
LOG("Unidentified name " + attribString + ": node skipped", Logger::Type::Warning);
}
}
return true;
}
bool ObjectTypes::reset()
{
m_workingDirectory.clear();
m_types.clear();
return false;
}

164
ext/tmxlite/src/Property.cpp Normal file
View File

@@ -0,0 +1,164 @@
/*********************************************************************
Matt Marchant 2016 - 2021
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/Property.hpp"
#include "tmxlite/detail/Log.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include <pugixml.hpp>
using namespace tmx;
Property::Property()
: m_type(Type::Undef)
{
}
Property Property::fromBoolean(bool value)
{
Property p;
p.m_type = Type::Boolean;
p.m_boolValue = value;
return p;
}
Property Property::fromFloat(float value)
{
Property p;
p.m_type = Type::Float;
p.m_floatValue = value;
return p;
}
Property Property::fromInt(int value)
{
Property p;
p.m_type = Type::Int;
p.m_intValue = value;
return p;
}
Property Property::fromString(const std::string& value)
{
Property p;
p.m_type = Type::String;
p.m_stringValue = value;
return p;
}
Property Property::fromColour(const Colour& value)
{
Property p;
p.m_type = Type::Colour;
p.m_colourValue = value;
return p;
}
Property Property::fromFile(const std::string& value)
{
Property p;
p.m_type = Type::File;
p.m_stringValue = value;
return p;
}
Property Property::fromObject(int value)
{
Property p;
p.m_type = Type::Object;
p.m_intValue = value;
return p;
}
//public
void Property::parse(const pugi::xml_node& node, bool isObjectTypes)
{
// The value attribute name is different in object types
const char *const valueAttribute = isObjectTypes ? "default" : "value";
std::string attribData = node.name();
if (attribData != "property")
{
Logger::log("Node was not a valid property, node will be skipped", Logger::Type::Error);
return;
}
m_name = node.attribute("name").as_string();
attribData = node.attribute("type").as_string("string");
if (attribData == "bool")
{
attribData = node.attribute(valueAttribute).as_string("false");
m_boolValue = (attribData == "true");
m_type = Type::Boolean;
return;
}
else if (attribData == "int")
{
m_intValue = node.attribute(valueAttribute).as_int(0);
m_type = Type::Int;
return;
}
else if (attribData == "float")
{
m_floatValue = node.attribute(valueAttribute).as_float(0.f);
m_type = Type::Float;
return;
}
else if (attribData == "string")
{
m_stringValue = node.attribute(valueAttribute).as_string();
//if value is empty, try getting the child value instead
//as this is how multiline string properties are stored.
if(m_stringValue.empty())
{
m_stringValue = node.child_value();
}
m_type = Type::String;
return;
}
else if (attribData == "color")
{
m_colourValue = colourFromString(node.attribute(valueAttribute).as_string("#FFFFFFFF"));
m_type = Type::Colour;
return;
}
else if (attribData == "file")
{
m_stringValue = node.attribute(valueAttribute).as_string();
m_type = Type::File;
return;
}
else if (attribData == "object")
{
m_intValue = node.attribute(valueAttribute).as_int(0);
m_type = Type::Object;
return;
}
}

342
ext/tmxlite/src/TileLayer.cpp Normal file
View File

@@ -0,0 +1,342 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "base64.h"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/TileLayer.hpp"
#include "tmxlite/detail/Log.hpp"
#ifndef USE_ZLIB
# include "tmxlite/detail/gzip.hpp"
#endif
#include <pugixml.hpp>
#include <zstd.h>
#include <sstream>
#include <span>
using namespace tmx;
namespace
{
struct CompressionType final
{
enum
{
Zlib, GZip, Zstd, None
};
};
}
TileLayer::TileLayer(std::size_t tileCount)
: m_tileCount (tileCount)
{
m_tiles.reserve(tileCount);
}
//public
void TileLayer::parse(const pugi::xml_node& node, Map*)
{
std::string attribName = node.name();
if (attribName != "layer")
{
Logger::log("node not a layer node, skipped parsing", Logger::Type::Error);
return;
}
setName(node.attribute("name").as_string());
setClass(node.attribute("class").as_string());
setOpacity(node.attribute("opacity").as_float(1.f));
setVisible(node.attribute("visible").as_bool(true));
setOffset(node.attribute("offsetx").as_int(0), node.attribute("offsety").as_int(0));
setSize(node.attribute("width").as_uint(0), node.attribute("height").as_uint(0));
setParallaxFactor(node.attribute("parallaxx").as_float(1.f), node.attribute("parallaxy").as_float(1.f));
std::string tintColour = node.attribute("tintcolor").as_string();
if (!tintColour.empty())
{
setTintColour(colourFromString(tintColour));
}
for (const auto& child : node.children())
{
attribName = child.name();
if (attribName == "data")
{
attribName = child.attribute("encoding").as_string();
if (attribName == "base64")
{
parseBase64(child);
}
else if (attribName == "csv")
{
parseCSV(child);
}
else
{
parseUnencoded(child);
}
}
else if (attribName == "properties")
{
for (const auto& p : child.children())
{
addProperty(p);
}
}
}
}
//private
void TileLayer::parseBase64(const pugi::xml_node& node)
{
auto processDataString = [](std::string dataString, std::size_t tileCount, std::int32_t compressionType)->std::vector<std::uint32_t>
{
std::stringstream ss;
ss << dataString;
ss >> dataString;
dataString = base64_decode(dataString);
std::size_t expectedSize = tileCount * 4; //4 bytes per tile
std::vector<unsigned char> byteData;
byteData.reserve(expectedSize);
switch (compressionType)
{
default:
byteData.insert(byteData.end(), dataString.begin(), dataString.end());
break;
case CompressionType::Zstd:
{
std::size_t dataSize = dataString.length() * sizeof(unsigned char);
std::size_t result = ZSTD_decompress(byteData.data(), expectedSize, &dataString[0], dataSize);
if (ZSTD_isError(result))
{
std::string err = ZSTD_getErrorName(result);
LOG("Failed to decompress layer data, node skipped.\nError: " + err, Logger::Type::Error);
}
}
break;
case CompressionType::GZip:
#ifndef USE_ZLIB
{
byteData.resize(expectedSize);
const auto source = std::span(reinterpret_cast<const uint8_t*>(dataString.data()), dataString.size());
GZipReader reader;
if (!reader.OpenMemory(source) || !reader.Read(byteData) || !reader.Check())
{
LOG("Failed to decompress layer data, node skipped.", Logger::Type::Error);
return {};
}
}
break;
#endif
//[[fallthrough]];
case CompressionType::Zlib:
{
//unzip
std::size_t dataSize = dataString.length() * sizeof(unsigned char);
if (!decompress(dataString.c_str(), byteData, dataSize, expectedSize))
{
LOG("Failed to decompress layer data, node skipped.", Logger::Type::Error);
return {};
}
}
break;
}
//data stream is in bytes so we need to OR into 32 bit values
std::vector<std::uint32_t> IDs;
IDs.reserve(tileCount);
for (auto i = 0u; i < expectedSize - 3u; i += 4u)
{
std::uint32_t id = byteData[i] | byteData[i + 1] << 8 | byteData[i + 2] << 16 | byteData[i + 3] << 24;
IDs.push_back(id);
}
return IDs;
};
std::int32_t compressionType = CompressionType::None;
std::string compression = node.attribute("compression").as_string();
if (compression == "gzip")
{
compressionType = CompressionType::GZip;
}
else if (compression == "zlib")
{
compressionType = CompressionType::Zlib;
}
else if (compression == "zstd")
{
compressionType = CompressionType::Zstd;
}
std::string data = node.text().as_string();
if (data.empty())
{
//check for chunk nodes
auto dataCount = 0;
for (const auto& childNode : node.children())
{
std::string childName = childNode.name();
if (childName == "chunk")
{
std::string dataString = childNode.text().as_string();
if (!dataString.empty())
{
Chunk chunk;
chunk.position.x = childNode.attribute("x").as_int();
chunk.position.y = childNode.attribute("y").as_int();
chunk.size.x = childNode.attribute("width").as_int();
chunk.size.y = childNode.attribute("height").as_int();
auto IDs = processDataString(dataString, (chunk.size.x * chunk.size.y), compressionType);
if (!IDs.empty())
{
createTiles(IDs, chunk.tiles);
m_chunks.push_back(chunk);
dataCount++;
}
}
}
}
if (dataCount == 0)
{
Logger::log("Layer " + getName() + " has no layer data. Layer skipped.", Logger::Type::Error);
return;
}
}
else
{
auto IDs = processDataString(data, m_tileCount, compressionType);
createTiles(IDs, m_tiles);
}
}
void TileLayer::parseCSV(const pugi::xml_node& node)
{
auto processDataString = [](const std::string dataString, std::size_t tileCount)->std::vector<std::uint32_t>
{
std::vector<std::uint32_t> IDs;
IDs.reserve(tileCount);
const char* ptr = dataString.c_str();
while (true)
{
char* end;
auto res = std::strtoul(ptr, &end, 10);
if (end == ptr) break;
ptr = end;
IDs.push_back(res);
if (*ptr == ',') ++ptr;
}
return IDs;
};
std::string data = node.text().as_string();
if (data.empty())
{
//check for chunk nodes
auto dataCount = 0;
for (const auto& childNode : node.children())
{
std::string childName = childNode.name();
if (childName == "chunk")
{
std::string dataString = childNode.text().as_string();
if (!dataString.empty())
{
Chunk chunk;
chunk.position.x = childNode.attribute("x").as_int();
chunk.position.y = childNode.attribute("y").as_int();
chunk.size.x = childNode.attribute("width").as_int();
chunk.size.y = childNode.attribute("height").as_int();
auto IDs = processDataString(dataString, chunk.size.x * chunk.size.y);
if (!IDs.empty())
{
createTiles(IDs, chunk.tiles);
m_chunks.push_back(chunk);
dataCount++;
}
}
}
}
if (dataCount == 0)
{
Logger::log("Layer " + getName() + " has no layer data. Layer skipped.", Logger::Type::Error);
return;
}
}
else
{
createTiles(processDataString(data, m_tileCount), m_tiles);
}
}
void TileLayer::parseUnencoded(const pugi::xml_node& node)
{
std::string attribName;
std::vector<std::uint32_t> IDs;
IDs.reserve(m_tileCount);
for (const auto& child : node.children())
{
attribName = child.name();
if (attribName == "tile")
{
IDs.push_back(child.attribute("gid").as_uint());
}
}
createTiles(IDs, m_tiles);
}
void TileLayer::createTiles(const std::vector<std::uint32_t>& IDs, std::vector<Tile>& destination)
{
//LOG(IDs.size() != m_tileCount, "Layer tile count does not match expected size. Found: "
// + std::to_string(IDs.size()) + ", expected: " + std::to_string(m_tileCount));
static const std::uint32_t mask = 0xf0000000;
for (const auto& id : IDs)
{
destination.emplace_back();
destination.back().flipFlags = ((id & mask) >> 28);
destination.back().ID = id & ~mask;
}
}

456
ext/tmxlite/src/Tileset.cpp Normal file
View File

@@ -0,0 +1,456 @@
/*********************************************************************
Matt Marchant 2016 - 2023
http://trederia.blogspot.com
tmxlite - Zlib license.
This software is provided 'as-is', without any express or
implied warranty. In no event will the authors be held
liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute
it freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented;
you must not claim that you wrote the original software.
If you use this software in a product, an acknowledgment
in the product documentation would be appreciated but
is not required.
2. Altered source versions must be plainly marked as such,
and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any
source distribution.
*********************************************************************/
#include "tmxlite/Tileset.hpp"
#include "tmxlite/FreeFuncs.hpp"
#include "tmxlite/detail/Log.hpp"
#include <pugixml.hpp>
#include <ctype.h>
using namespace tmx;
Tileset::Tileset(const std::string& workingDir)
: m_workingDir (workingDir),
m_firstGID (0),
m_spacing (0),
m_margin (0),
m_tileCount (0),
m_columnCount (0),
m_objectAlignment (ObjectAlignment::Unspecified),
m_transparencyColour (0, 0, 0, 0),
m_hasTransparency (false)
{
}
//public
void Tileset::parse(pugi::xml_node node, Map* map)
{
assert(map);
std::string attribString = node.name();
if (attribString != "tileset")
{
Logger::log(attribString + ": not a tileset node! Node will be skipped.", Logger::Type::Warning);
return;
}
m_firstGID = node.attribute("firstgid").as_int();
if (m_firstGID == 0)
{
Logger::log("Invalid first GID in tileset. Tileset node skipped.", Logger::Type::Warning);
return;
}
pugi::xml_document tsxDoc; //need to keep this in scope
if (node.attribute("source"))
{
//parse TSX doc
std::string path = node.attribute("source").as_string();
path = resolveFilePath(path, m_workingDir);
//as the TSX file now dictates the image path, the working
//directory is now that of the tsx file
auto position = path.find_last_of('/');
if (position != std::string::npos)
{
m_workingDir = path.substr(0, position);
}
else
{
m_workingDir = "";
}
//see if doc can be opened
auto result = tsxDoc.load_file(path.c_str());
if (!result)
{
Logger::log(path + ": Failed opening tsx file for tile set, tile set will be skipped", Logger::Type::Error);
return reset();
}
//if it can then replace the current node with tsx node
node = tsxDoc.child("tileset");
if (!node)
{
Logger::log("tsx file does not contain a tile set node, tile set will be skipped", Logger::Type::Error);
return reset();
}
}
m_name = node.attribute("name").as_string();
LOG("found tile set " + m_name, Logger::Type::Info);
m_class = node.attribute("class").as_string();
m_tileSize.x = node.attribute("tilewidth").as_int();
m_tileSize.y = node.attribute("tileheight").as_int();
if (m_tileSize.x == 0 || m_tileSize.y == 0)
{
Logger::log("Invalid tile size found in tile set node. Node will be skipped.", Logger::Type::Error);
return reset();
}
m_spacing = node.attribute("spacing").as_int();
m_margin = node.attribute("margin").as_int();
m_tileCount = node.attribute("tilecount").as_int();
m_columnCount = node.attribute("columns").as_int();
m_tileIndex.reserve(m_tileCount);
m_tiles.reserve(m_tileCount);
std::string objectAlignment = node.attribute("objectalignment").as_string();
if (!objectAlignment.empty())
{
if (objectAlignment == "unspecified")
{
m_objectAlignment = ObjectAlignment::Unspecified;
}
else if (objectAlignment == "topleft")
{
m_objectAlignment = ObjectAlignment::TopLeft;
}
else if (objectAlignment == "top")
{
m_objectAlignment = ObjectAlignment::Top;
}
else if (objectAlignment == "topright")
{
m_objectAlignment = ObjectAlignment::TopRight;
}
else if (objectAlignment == "left")
{
m_objectAlignment = ObjectAlignment::Left;
}
else if (objectAlignment == "center")
{
m_objectAlignment = ObjectAlignment::Center;
}
else if (objectAlignment == "right")
{
m_objectAlignment = ObjectAlignment::Right;
}
else if (objectAlignment == "bottomleft")
{
m_objectAlignment = ObjectAlignment::BottomLeft;
}
else if (objectAlignment == "bottom")
{
m_objectAlignment = ObjectAlignment::Bottom;
}
else if (objectAlignment == "bottomright")
{
m_objectAlignment = ObjectAlignment::BottomRight;
}
}
const auto& children = node.children();
for (const auto& node : children)
{
std::string name = node.name();
if (name == "image")
{
//TODO this currently doesn't cover embedded images
//mostly because I can't figure out how to export them
//from the Tiled editor... but also resource handling
//should be handled by the renderer, not the parser.
attribString = node.attribute("source").as_string();
if (attribString.empty())
{
Logger::log("Tileset image node has missing source property, tile set not loaded", Logger::Type::Error);
return reset();
}
m_imagePath = resolveFilePath(attribString, m_workingDir);
if (node.attribute("trans"))
{
attribString = node.attribute("trans").as_string();
m_transparencyColour = colourFromString(attribString);
m_hasTransparency = true;
}
if (node.attribute("width") && node.attribute("height"))
{
m_imageSize.x = node.attribute("width").as_int();
m_imageSize.y = node.attribute("height").as_int();
}
}
else if (name == "tileoffset")
{
parseOffsetNode(node);
}
else if (name == "properties")
{
parsePropertyNode(node);
}
else if (name == "terraintypes")
{
parseTerrainNode(node);
}
else if (name == "tile")
{
parseTileNode(node, map);
}
}
//if the tsx file does not declare every tile, we create the missing ones
if (m_tiles.size() != getTileCount())
{
for (std::uint32_t ID = 0; ID < getTileCount(); ID++)
{
createMissingTile(ID);
}
}
}
std::uint32_t Tileset::getLastGID() const
{
assert(!m_tileIndex.empty());
return m_firstGID + static_cast<std::uint32_t>(m_tileIndex.size()) - 1;
}
const Tileset::Tile* Tileset::getTile(std::uint32_t id) const
{
if (!hasTile(id))
{
return nullptr;
}
//corrects the ID. Indices and IDs are different.
id -= m_firstGID;
id = m_tileIndex[id];
return id ? &m_tiles[id - 1] : nullptr;
}
//private
void Tileset::reset()
{
m_firstGID = 0;
m_source = "";
m_name = "";
m_class = "";
m_tileSize = { 0,0 };
m_spacing = 0;
m_margin = 0;
m_tileCount = 0;
m_columnCount = 0;
m_objectAlignment = ObjectAlignment::Unspecified;
m_tileOffset = { 0,0 };
m_properties.clear();
m_imagePath = "";
m_transparencyColour = { 0, 0, 0, 0 };
m_hasTransparency = false;
m_terrainTypes.clear();
m_tileIndex.clear();
m_tiles.clear();
}
void Tileset::parseOffsetNode(const pugi::xml_node& node)
{
m_tileOffset.x = node.attribute("x").as_int();
m_tileOffset.y = node.attribute("y").as_int();
}
void Tileset::parsePropertyNode(const pugi::xml_node& node)
{
const auto& children = node.children();
for (const auto& child : children)
{
m_properties.emplace_back();
m_properties.back().parse(child);
}
}
void Tileset::parseTerrainNode(const pugi::xml_node& node)
{
const auto& children = node.children();
for (const auto& child : children)
{
std::string name = child.name();
if (name == "terrain")
{
m_terrainTypes.emplace_back();
auto& terrain = m_terrainTypes.back();
terrain.name = child.attribute("name").as_string();
terrain.tileID = child.attribute("tile").as_int();
auto properties = child.child("properties");
if (properties)
{
for (const auto& p : properties)
{
name = p.name();
if (name == "property")
{
terrain.properties.emplace_back();
terrain.properties.back().parse(p);
}
}
}
}
}
}
Tileset::Tile& Tileset::newTile(std::uint32_t ID)
{
Tile& tile = (m_tiles.emplace_back(), m_tiles.back());
if (m_tileIndex.size() <= ID)
{
m_tileIndex.resize(ID + 1, 0);
}
m_tileIndex[ID] = static_cast<std::uint32_t>(m_tiles.size());
tile.ID = ID;
return tile;
}
void Tileset::parseTileNode(const pugi::xml_node& node, Map* map)
{
assert(map);
Tile& tile = newTile(node.attribute("id").as_int());
if (node.attribute("terrain"))
{
std::string data = node.attribute("terrain").as_string();
bool lastWasChar = true;
std::size_t idx = 0u;
for (auto i = 0u; i < data.size() && idx < tile.terrainIndices.size(); ++i)
{
if (isdigit(data[i]))
{
tile.terrainIndices[idx++] = std::atoi(&data[i]);
lastWasChar = false;
}
else
{
if (!lastWasChar)
{
lastWasChar = true;
}
else
{
tile.terrainIndices[idx++] = -1;
lastWasChar = false;
}
}
}
if (lastWasChar)
{
tile.terrainIndices[idx] = -1;
}
}
tile.probability = node.attribute("probability").as_int(100);
tile.className = node.attribute("type").as_string();
if (tile.className.empty())
{
tile.className = node.attribute("class").as_string();
}
//by default we set the tile's values as in an Image tileset
tile.imagePath = m_imagePath;
tile.imageSize = m_tileSize;
if (m_columnCount != 0)
{
std::uint32_t rowIndex = tile.ID % m_columnCount;
std::uint32_t columnIndex = tile.ID / m_columnCount;
tile.imagePosition.x = m_margin + rowIndex * (m_tileSize.x + m_spacing);
tile.imagePosition.y = m_margin + columnIndex * (m_tileSize.y + m_spacing);
}
const auto& children = node.children();
for (const auto& child : children)
{
std::string name = child.name();
if (name == "properties")
{
for (const auto& prop : child.children())
{
tile.properties.emplace_back();
tile.properties.back().parse(prop);
}
}
else if (name == "objectgroup")
{
tile.objectGroup.parse(child, map);
}
else if (name == "image")
{
std::string attribString = child.attribute("source").as_string();
if (attribString.empty())
{
Logger::log("Tile image path missing", Logger::Type::Warning);
continue;
}
tile.imagePath = resolveFilePath(attribString, m_workingDir);
tile.imagePosition = tmx::Vector2u(0, 0);
if (child.attribute("trans"))
{
attribString = child.attribute("trans").as_string();
m_transparencyColour = colourFromString(attribString);
m_hasTransparency = true;
}
if (child.attribute("width"))
{
tile.imageSize.x = child.attribute("width").as_uint();
}
if (child.attribute("height"))
{
tile.imageSize.y = child.attribute("height").as_uint();
}
}
else if (name == "animation")
{
for (const auto& frameNode : child.children())
{
Tile::Animation::Frame frame;
frame.duration = frameNode.attribute("duration").as_int();
frame.tileID = frameNode.attribute("tileid").as_int() + m_firstGID;
tile.animation.frames.push_back(frame);
}
}
}
}
void Tileset::createMissingTile(std::uint32_t ID)
{
//first, we check if the tile does not yet exist
if (m_tileIndex.size() > ID && m_tileIndex[ID])
{
return;
}
Tile& tile = newTile(ID);
tile.imagePath = m_imagePath;
tile.imageSize = m_tileSize;
std::uint32_t rowIndex = ID % m_columnCount;
std::uint32_t columnIndex = ID / m_columnCount;
tile.imagePosition.x = m_margin + rowIndex * (m_tileSize.x + m_spacing);
tile.imagePosition.y = m_margin + columnIndex * (m_tileSize.y + m_spacing);
}

125
ext/tmxlite/src/detail/gzip.cpp Normal file
View File

@@ -0,0 +1,125 @@
// gzip.cpp - portable memory miniz based gzip reader
// SPDX-License-Identifier: Zlib
// SPDX-FileCopyrightText: (c) 2024 a dinosaur
#include "tmxlite/detail/gzip.hpp"
#include <string_view>
GZipReader::GZipReader() noexcept :
mSourceLen(0), mBytesRead(0),
mModificationTime(0), mCrc(0), mInputSize(0), mComputedCrc(0),
crc16(0), mFlags(0), mXflags(0), mOsId(0)
{
tinfl_init(&mState);
mComputedCrc = static_cast<uint32_t>(mz_crc32(0, nullptr, 0));
}
bool GZipReader::OpenMemory(const std::span<const uint8_t> source) noexcept
{
if (source.size() < 20)
return false;
auto it = std::cbegin(source), end = std::cend(source);
constexpr uint8_t magic[2] = { 0x1F, 0x8B };
if (*it++ != magic[0] || *it++ != magic[1])
return false;
constexpr uint8_t CM_DEFLATE = 8;
uint8_t compression = *it++;
if (compression != CM_DEFLATE)
return false;
mFlags = *it++;
mModificationTime = *it++;
mModificationTime |= *it++ << 8;
mModificationTime |= *it++ << 16;
mModificationTime |= *it++ << 24;
mXflags = *it++;
mOsId = *it++;
if (mFlags & FEXTRA)
{
// Skip "extra" field
if (it + 2 >= end)
return false;
uint16_t extraLen = *it++;
extraLen = *it++ << 8;
if (it + extraLen >= end)
return false;
it += extraLen;
}
if (mFlags & FNAME)
{
// Skip null-terminated name string
do
{
if (++it == end)
return false;
} while (*it != '\0');
if (++it == end)
return false;
}
if (mFlags & FCOMMENT)
{
// Skip null-terminated comment string
do
{
if (++it == end)
return false;
} while (*it != '\0');
if (++it == end)
return false;
}
if (mFlags & FHCRC)
{
if (it + 2 >= end)
return false;
crc16 = *it++;
crc16 |= *it++;
}
mIt = it;
mSourceLen = end - it - 8;
it += mSourceLen;
mCrc = *it++;
mCrc |= *it++ << 8;
mCrc |= *it++ << 16;
mCrc |= *it++ << 24;
mInputSize = *it++;
mInputSize |= *it++ << 8;
mInputSize |= *it++ << 16;
mInputSize |= *it++ << 24;
return true;
}
bool GZipReader::Read(std::span<uint8_t> out) noexcept
{
size_t outLen = out.size();
auto res = tinfl_decompress(&mState,
static_cast<const mz_uint8*>(&*mIt), &mSourceLen,
static_cast<mz_uint8*>(out.data()), static_cast<mz_uint8*>(out.data()), &outLen,
TINFL_FLAG_USING_NON_WRAPPING_OUTPUT_BUF);
if (res != TINFL_STATUS_DONE)
return false;
mIt += outLen;
mBytesRead += outLen;
mComputedCrc = static_cast<uint32_t>(mz_crc32(static_cast<mz_ulong>(mComputedCrc), out.data(), outLen));
return true;
}
bool GZipReader::Check() const noexcept
{
if (mComputedCrc != mCrc)
return false;
if (static_cast<uint32_t>(mBytesRead & UINT32_MAX) != mInputSize)
return false;
return true;
}

145
ext/zstd/CMakeLists.txt Normal file
View File

@@ -0,0 +1,145 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
set(LIBRARY_DIR ${CMAKE_CURRENT_SOURCE_DIR}/lib)
# Parse version
file(READ ${LIBRARY_DIR}/zstd.h CONTENT) # Read file content
string(REGEX MATCH ".*define ZSTD_VERSION_MAJOR *([0-9]+).*define ZSTD_VERSION_MINOR *([0-9]+).*define ZSTD_VERSION_RELEASE *([0-9]+)" VERSION_REGEX "${CONTENT}")
set(zstd_VERSION_MAJOR ${CMAKE_MATCH_1} PARENT_SCOPE)
set(zstd_VERSION_MINOR ${CMAKE_MATCH_2} PARENT_SCOPE)
set(zstd_VERSION_PATCH ${CMAKE_MATCH_3} PARENT_SCOPE)
enable_language(ASM)
#-----------------------------------------------------------------------------
# Add extra compilation flags
#-----------------------------------------------------------------------------
function (add_zstd_compilation_flags _target)
if (CMAKE_CXX_COMPILER_ID MATCHES "GNU|Clang" OR MINGW)
if (CMAKE_CXX_COMPILER_ID MATCHES "Clang" AND MSVC)
# clang-cl normally maps -Wall to -Weverything.
target_compile_options(${_target} PRIVATE "/clang:-Wall")
else()
target_compile_options(${_target} PRIVATE -Wall)
endif()
target_compile_options(${_target} PRIVATE -Wextra -Wundef -Wshadow -Wcast-align -Wcast-qual)
target_compile_options(${_target} PRIVATE $<$<COMPILE_LANGUAGE:CXX>:-Wstrict-prototypes>)
# Enable asserts in Debug mode
if (CMAKE_BUILD_TYPE MATCHES "Debug")
target_compile_options(${_target} PRIVATE DEBUGLEVEL=1)
endif()
# Add noexecstack flags
target_link_options(${_target} PRIVATE -z noexecstack) # LDFLAGS
target_compile_options(${_target} PRIVATE -Wunused-parameter -Wa,--noexecstack) # CFLAGS & CXXFLAGS
elseif (MSVC)
# UNICODE SUPPORT
target_compile_definitions(${_target} PRIVATE _UNICODE UNICODE)
# Enable asserts in Debug mode
if (CMAKE_BUILD_TYPE MATCHES "Debug")
target_compile_definitions(${_target} PRIVATE DEBUGLEVEL=1)
endif()
endif()
endfunction()
# Legacy support
option(ZSTD_LEGACY_SUPPORT "LEGACY SUPPORT" ON)
if (ZSTD_LEGACY_SUPPORT)
set(ZSTD_LEGACY_LEVEL 5 CACHE STRING "")
endif()
# Multi-threading support
if (ANDROID)
set(ZSTD_MULTITHREAD_SUPPORT_DEFAULT OFF)
else()
set(ZSTD_MULTITHREAD_SUPPORT_DEFAULT ON)
endif()
option(ZSTD_MULTITHREAD_SUPPORT "MULTITHREADING SUPPORT" ${ZSTD_MULTITHREAD_SUPPORT_DEFAULT})
file(GLOB CommonSources ${LIBRARY_DIR}/common/*.c)
file(GLOB CompressSources ${LIBRARY_DIR}/compress/*.c)
if (MSVC)
file(GLOB DecompressSources ${LIBRARY_DIR}/decompress/*.c)
else()
file(GLOB DecompressSources ${LIBRARY_DIR}/decompress/*.c ${LIBRARY_DIR}/decompress/*.S)
endif()
file(GLOB DictBuilderSources ${LIBRARY_DIR}/dictBuilder/*.c)
set(Sources
${CommonSources}
${CompressSources}
${DecompressSources}
${DictBuilderSources})
file(GLOB CommonHeaders ${LIBRARY_DIR}/common/*.h)
file(GLOB CompressHeaders ${LIBRARY_DIR}/compress/*.h)
file(GLOB DecompressHeaders ${LIBRARY_DIR}/decompress/*.h)
file(GLOB DictBuilderHeaders ${LIBRARY_DIR}/dictBuilder/*.h)
set(Headers
${LIBRARY_DIR}/zstd.h
${CommonHeaders}
${CompressHeaders}
${DecompressHeaders}
${DictBuilderHeaders})
if (ZSTD_LEGACY_SUPPORT)
foreach (SOURCE zstd_v01.c zstd_v02.c zstd_v03.c zstd_v04.c zstd_v05.c zstd_v06.c zstd_v07.c)
list(APPEND Sources ${LIBRARY_DIR}/legacy/${SOURCE})
endforeach()
foreach (HEADER zstd_legacy.h zstd_v01.h zstd_v02.h zstd_v03.h zstd_v04.h zstd_v05.h zstd_v06.h zstd_v07.h)
list(APPEND Headers ${LIBRARY_DIR}/legacy/${HEADER})
endforeach()
endif()
# Explicitly set the language to C for all files, including ASM files.
# Our assembly expects to be compiled by a C compiler, and is only enabled for
# __GNUC__ compatible compilers. Otherwise all the ASM code is disabled by
# macros.
set_source_files_properties(${Sources} PROPERTIES LANGUAGE C)
add_library(zstd STATIC ${Sources} ${Headers})
add_library(Zstd::static ALIAS zstd)
add_library(Zstd::Zstd ALIAS zstd)
add_zstd_compilation_flags(zstd)
# Define library directory, where sources and header files are located
target_include_directories(zstd PUBLIC ${LIBRARY_DIR})
target_include_directories(zstd PRIVATE ${LIBRARY_DIR}/common)
if (ZSTD_LEGACY_SUPPORT)
target_include_directories(zstd PRIVATE ${LIBRARY_DIR}/legacy)
target_compile_definitions(zstd PRIVATE ZSTD_LEGACY_SUPPORT=${ZSTD_LEGACY_LEVEL})
else()
target_compile_definitions(zstd PRIVATE ZSTD_LEGACY_SUPPORT=0)
endif()
if (ZSTD_MULTITHREAD_SUPPORT)
target_compile_definitions(zstd PRIVATE ZSTD_MULTITHREAD)
if (UNIX)
set(THREADS_PREFER_PTHREAD_FLAG ON)
find_package(Threads REQUIRED)
target_link_libraries(zstd Threads::Threads)
if (NOT CMAKE_USE_PTHREADS_INIT)
message(SEND_ERROR "ZSTD currently does not support thread libraries other than pthreads")
endif()
endif()
endif()
# Add specific compile definitions for MSVC project
if (MSVC)
target_compile_definitions(zstd PRIVATE ZSTD_HEAPMODE=0 ZSTD_DISABLE_ASM _CRT_SECURE_NO_WARNINGS)
endif()
# Define static library names
set_property(TARGET zstd PROPERTY POSITION_INDEPENDENT_CODE ON)

30
ext/zstd/LICENSE Normal file
View File

@@ -0,0 +1,30 @@
BSD License
For Zstandard software
Copyright (c) Meta Platforms, Inc. and affiliates. All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
* Neither the name Facebook, nor Meta, nor the names of its contributors may
be used to endorse or promote products derived from this software without
specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

55
ext/zstd/lib/common/allocations.h Normal file
View File

@@ -0,0 +1,55 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* This file provides custom allocation primitives
*/
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h" /* ZSTD_malloc, ZSTD_calloc, ZSTD_free, ZSTD_memset */
#include "mem.h" /* MEM_STATIC */
#define ZSTD_STATIC_LINKING_ONLY
#include "../zstd.h" /* ZSTD_customMem */
#ifndef ZSTD_ALLOCATIONS_H
#define ZSTD_ALLOCATIONS_H
/* custom memory allocation functions */
MEM_STATIC void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem)
{
if (customMem.customAlloc)
return customMem.customAlloc(customMem.opaque, size);
return ZSTD_malloc(size);
}
MEM_STATIC void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem)
{
if (customMem.customAlloc) {
/* calloc implemented as malloc+memset;
* not as efficient as calloc, but next best guess for custom malloc */
void* const ptr = customMem.customAlloc(customMem.opaque, size);
ZSTD_memset(ptr, 0, size);
return ptr;
}
return ZSTD_calloc(1, size);
}
MEM_STATIC void ZSTD_customFree(void* ptr, ZSTD_customMem customMem)
{
if (ptr!=NULL) {
if (customMem.customFree)
customMem.customFree(customMem.opaque, ptr);
else
ZSTD_free(ptr);
}
}
#endif /* ZSTD_ALLOCATIONS_H */

200
ext/zstd/lib/common/bits.h Normal file
View File

@@ -0,0 +1,200 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_BITS_H
#define ZSTD_BITS_H
#include "mem.h"
MEM_STATIC unsigned ZSTD_countTrailingZeros32_fallback(U32 val)
{
assert(val != 0);
{
static const U32 DeBruijnBytePos[32] = {0, 1, 28, 2, 29, 14, 24, 3,
30, 22, 20, 15, 25, 17, 4, 8,
31, 27, 13, 23, 21, 19, 16, 7,
26, 12, 18, 6, 11, 5, 10, 9};
return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countTrailingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return (unsigned)_tzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_ctz(val);
# else
return ZSTD_countTrailingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32_fallback(U32 val) {
assert(val != 0);
{
static const U32 DeBruijnClz[32] = {0, 9, 1, 10, 13, 21, 2, 29,
11, 14, 16, 18, 22, 25, 3, 30,
8, 12, 20, 28, 15, 17, 24, 7,
19, 27, 23, 6, 26, 5, 4, 31};
val |= val >> 1;
val |= val >> 2;
val |= val >> 4;
val |= val >> 8;
val |= val >> 16;
return 31 - DeBruijnClz[(val * 0x07C4ACDDU) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return (unsigned)_lzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse(&r, val);
return (unsigned)(31 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_clz(val);
# else
return ZSTD_countLeadingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countTrailingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return (unsigned)_tzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward64(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4) && defined(__LP64__)
return (unsigned)__builtin_ctzll(val);
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (leastSignificantWord == 0) {
return 32 + ZSTD_countTrailingZeros32(mostSignificantWord);
} else {
return ZSTD_countTrailingZeros32(leastSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return (unsigned)_lzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse64(&r, val);
return (unsigned)(63 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)(__builtin_clzll(val));
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (mostSignificantWord == 0) {
return 32 + ZSTD_countLeadingZeros32(leastSignificantWord);
} else {
return ZSTD_countLeadingZeros32(mostSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_NbCommonBytes(size_t val)
{
if (MEM_isLittleEndian()) {
if (MEM_64bits()) {
return ZSTD_countTrailingZeros64((U64)val) >> 3;
} else {
return ZSTD_countTrailingZeros32((U32)val) >> 3;
}
} else { /* Big Endian CPU */
if (MEM_64bits()) {
return ZSTD_countLeadingZeros64((U64)val) >> 3;
} else {
return ZSTD_countLeadingZeros32((U32)val) >> 3;
}
}
}
MEM_STATIC unsigned ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */
{
assert(val != 0);
return 31 - ZSTD_countLeadingZeros32(val);
}
/* ZSTD_rotateRight_*():
* Rotates a bitfield to the right by "count" bits.
* https://en.wikipedia.org/w/index.php?title=Circular_shift&oldid=991635599#Implementing_circular_shifts
*/
MEM_STATIC
U64 ZSTD_rotateRight_U64(U64 const value, U32 count) {
assert(count < 64);
count &= 0x3F; /* for fickle pattern recognition */
return (value >> count) | (U64)(value << ((0U - count) & 0x3F));
}
MEM_STATIC
U32 ZSTD_rotateRight_U32(U32 const value, U32 count) {
assert(count < 32);
count &= 0x1F; /* for fickle pattern recognition */
return (value >> count) | (U32)(value << ((0U - count) & 0x1F));
}
MEM_STATIC
U16 ZSTD_rotateRight_U16(U16 const value, U32 count) {
assert(count < 16);
count &= 0x0F; /* for fickle pattern recognition */
return (value >> count) | (U16)(value << ((0U - count) & 0x0F));
}
#endif /* ZSTD_BITS_H */

437
ext/zstd/lib/common/bitstream.h Normal file
View File

@@ -0,0 +1,437 @@
/* ******************************************************************
* bitstream
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#ifndef BITSTREAM_H_MODULE
#define BITSTREAM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/*
* This API consists of small unitary functions, which must be inlined for best performance.
* Since link-time-optimization is not available for all compilers,
* these functions are defined into a .h to be included.
*/
/*-****************************************
* Dependencies
******************************************/
#include "mem.h" /* unaligned access routines */
#include "compiler.h" /* UNLIKELY() */
#include "debug.h" /* assert(), DEBUGLOG(), RAWLOG() */
#include "error_private.h" /* error codes and messages */
#include "bits.h" /* ZSTD_highbit32 */
/*=========================================
* Target specific
=========================================*/
#ifndef ZSTD_NO_INTRINSICS
# if (defined(__BMI__) || defined(__BMI2__)) && defined(__GNUC__)
# include <immintrin.h> /* support for bextr (experimental)/bzhi */
# elif defined(__ICCARM__)
# include <intrinsics.h>
# endif
#endif
#define STREAM_ACCUMULATOR_MIN_32 25
#define STREAM_ACCUMULATOR_MIN_64 57
#define STREAM_ACCUMULATOR_MIN ((U32)(MEM_32bits() ? STREAM_ACCUMULATOR_MIN_32 : STREAM_ACCUMULATOR_MIN_64))
/*-******************************************
* bitStream encoding API (write forward)
********************************************/
/* bitStream can mix input from multiple sources.
* A critical property of these streams is that they encode and decode in **reverse** direction.
* So the first bit sequence you add will be the last to be read, like a LIFO stack.
*/
typedef struct {
size_t bitContainer;
unsigned bitPos;
char* startPtr;
char* ptr;
char* endPtr;
} BIT_CStream_t;
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t dstCapacity);
MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC);
MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
/* Start with initCStream, providing the size of buffer to write into.
* bitStream will never write outside of this buffer.
* `dstCapacity` must be >= sizeof(bitD->bitContainer), otherwise @return will be an error code.
*
* bits are first added to a local register.
* Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems.
* Writing data into memory is an explicit operation, performed by the flushBits function.
* Hence keep track how many bits are potentially stored into local register to avoid register overflow.
* After a flushBits, a maximum of 7 bits might still be stored into local register.
*
* Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers.
*
* Last operation is to close the bitStream.
* The function returns the final size of CStream in bytes.
* If data couldn't fit into `dstBuffer`, it will return a 0 ( == not storable)
*/
/*-********************************************
* bitStream decoding API (read backward)
**********************************************/
typedef struct {
size_t bitContainer;
unsigned bitsConsumed;
const char* ptr;
const char* start;
const char* limitPtr;
} BIT_DStream_t;
typedef enum { BIT_DStream_unfinished = 0,
BIT_DStream_endOfBuffer = 1,
BIT_DStream_completed = 2,
BIT_DStream_overflow = 3 } BIT_DStream_status; /* result of BIT_reloadDStream() */
/* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize);
MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits);
MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD);
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
/* Start by invoking BIT_initDStream().
* A chunk of the bitStream is then stored into a local register.
* Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
* You can then retrieve bitFields stored into the local register, **in reverse order**.
* Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
* A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished.
* Otherwise, it can be less than that, so proceed accordingly.
* Checking if DStream has reached its end can be performed with BIT_endOfDStream().
*/
/*-****************************************
* unsafe API
******************************************/
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
/* faster, but works only if value is "clean", meaning all high bits above nbBits are 0 */
MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC);
/* unsafe version; does not check buffer overflow */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
/* faster, but works only if nbBits >= 1 */
/*===== Local Constants =====*/
static const unsigned BIT_mask[] = {
0, 1, 3, 7, 0xF, 0x1F,
0x3F, 0x7F, 0xFF, 0x1FF, 0x3FF, 0x7FF,
0xFFF, 0x1FFF, 0x3FFF, 0x7FFF, 0xFFFF, 0x1FFFF,
0x3FFFF, 0x7FFFF, 0xFFFFF, 0x1FFFFF, 0x3FFFFF, 0x7FFFFF,
0xFFFFFF, 0x1FFFFFF, 0x3FFFFFF, 0x7FFFFFF, 0xFFFFFFF, 0x1FFFFFFF,
0x3FFFFFFF, 0x7FFFFFFF}; /* up to 31 bits */
#define BIT_MASK_SIZE (sizeof(BIT_mask) / sizeof(BIT_mask[0]))
/*-**************************************************************
* bitStream encoding
****************************************************************/
/*! BIT_initCStream() :
* `dstCapacity` must be > sizeof(size_t)
* @return : 0 if success,
* otherwise an error code (can be tested using ERR_isError()) */
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
void* startPtr, size_t dstCapacity)
{
bitC->bitContainer = 0;
bitC->bitPos = 0;
bitC->startPtr = (char*)startPtr;
bitC->ptr = bitC->startPtr;
bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->bitContainer);
if (dstCapacity <= sizeof(bitC->bitContainer)) return ERROR(dstSize_tooSmall);
return 0;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
{
#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 && !defined(ZSTD_NO_INTRINSICS)
return _bzhi_u64(bitContainer, nbBits);
#else
assert(nbBits < BIT_MASK_SIZE);
return bitContainer & BIT_mask[nbBits];
#endif
}
/*! BIT_addBits() :
* can add up to 31 bits into `bitC`.
* Note : does not check for register overflow ! */
MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
size_t value, unsigned nbBits)
{
DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
assert(nbBits < BIT_MASK_SIZE);
assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
bitC->bitContainer |= BIT_getLowerBits(value, nbBits) << bitC->bitPos;
bitC->bitPos += nbBits;
}
/*! BIT_addBitsFast() :
* works only if `value` is _clean_,
* meaning all high bits above nbBits are 0 */
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC,
size_t value, unsigned nbBits)
{
assert((value>>nbBits) == 0);
assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
bitC->bitContainer |= value << bitC->bitPos;
bitC->bitPos += nbBits;
}
/*! BIT_flushBitsFast() :
* assumption : bitContainer has not overflowed
* unsafe version; does not check buffer overflow */
MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC)
{
size_t const nbBytes = bitC->bitPos >> 3;
assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
assert(bitC->ptr <= bitC->endPtr);
MEM_writeLEST(bitC->ptr, bitC->bitContainer);
bitC->ptr += nbBytes;
bitC->bitPos &= 7;
bitC->bitContainer >>= nbBytes*8;
}
/*! BIT_flushBits() :
* assumption : bitContainer has not overflowed
* safe version; check for buffer overflow, and prevents it.
* note : does not signal buffer overflow.
* overflow will be revealed later on using BIT_closeCStream() */
MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC)
{
size_t const nbBytes = bitC->bitPos >> 3;
assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
assert(bitC->ptr <= bitC->endPtr);
MEM_writeLEST(bitC->ptr, bitC->bitContainer);
bitC->ptr += nbBytes;
if (bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr;
bitC->bitPos &= 7;
bitC->bitContainer >>= nbBytes*8;
}
/*! BIT_closeCStream() :
* @return : size of CStream, in bytes,
* or 0 if it could not fit into dstBuffer */
MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
{
BIT_addBitsFast(bitC, 1, 1); /* endMark */
BIT_flushBits(bitC);
if (bitC->ptr >= bitC->endPtr) return 0; /* overflow detected */
return (bitC->ptr - bitC->startPtr) + (bitC->bitPos > 0);
}
/*-********************************************************
* bitStream decoding
**********************************************************/
/*! BIT_initDStream() :
* Initialize a BIT_DStream_t.
* `bitD` : a pointer to an already allocated BIT_DStream_t structure.
* `srcSize` must be the *exact* size of the bitStream, in bytes.
* @return : size of stream (== srcSize), or an errorCode if a problem is detected
*/
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
{
if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
bitD->start = (const char*)srcBuffer;
bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer);
if (srcSize >= sizeof(bitD->bitContainer)) { /* normal case */
bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer);
bitD->bitContainer = MEM_readLEST(bitD->ptr);
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */
if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ }
} else {
bitD->ptr = bitD->start;
bitD->bitContainer = *(const BYTE*)(bitD->start);
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
ZSTD_FALLTHROUGH;
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
ZSTD_FALLTHROUGH;
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
ZSTD_FALLTHROUGH;
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24;
ZSTD_FALLTHROUGH;
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16;
ZSTD_FALLTHROUGH;
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) << 8;
ZSTD_FALLTHROUGH;
default: break;
}
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;
if (lastByte == 0) return ERROR(corruption_detected); /* endMark not present */
}
bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8;
}
return srcSize;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
{
return bitContainer >> start;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
{
U32 const regMask = sizeof(bitContainer)*8 - 1;
/* if start > regMask, bitstream is corrupted, and result is undefined */
assert(nbBits < BIT_MASK_SIZE);
/* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better
* than accessing memory. When bmi2 instruction is not present, we consider
* such cpus old (pre-Haswell, 2013) and their performance is not of that
* importance.
*/
#if defined(__x86_64__) || defined(_M_X86)
return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1);
#else
return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
#endif
}
/*! BIT_lookBits() :
* Provides next n bits from local register.
* local register is not modified.
* On 32-bits, maxNbBits==24.
* On 64-bits, maxNbBits==56.
* @return : value extracted */
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
{
/* arbitrate between double-shift and shift+mask */
#if 1
/* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8,
* bitstream is likely corrupted, and result is undefined */
return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits);
#else
/* this code path is slower on my os-x laptop */
U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask);
#endif
}
/*! BIT_lookBitsFast() :
* unsafe version; only works if nbBits >= 1 */
MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits)
{
U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
assert(nbBits >= 1);
return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask);
}
MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
{
bitD->bitsConsumed += nbBits;
}
/*! BIT_readBits() :
* Read (consume) next n bits from local register and update.
* Pay attention to not read more than nbBits contained into local register.
* @return : extracted value. */
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBits(bitD, nbBits);
BIT_skipBits(bitD, nbBits);
return value;
}
/*! BIT_readBitsFast() :
* unsafe version; only works if nbBits >= 1 */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBitsFast(bitD, nbBits);
assert(nbBits >= 1);
BIT_skipBits(bitD, nbBits);
return value;
}
/*! BIT_reloadDStreamFast() :
* Similar to BIT_reloadDStream(), but with two differences:
* 1. bitsConsumed <= sizeof(bitD->bitContainer)*8 must hold!
* 2. Returns BIT_DStream_overflow when bitD->ptr < bitD->limitPtr, at this
* point you must use BIT_reloadDStream() to reload.
*/
MEM_STATIC BIT_DStream_status BIT_reloadDStreamFast(BIT_DStream_t* bitD)
{
if (UNLIKELY(bitD->ptr < bitD->limitPtr))
return BIT_DStream_overflow;
assert(bitD->bitsConsumed <= sizeof(bitD->bitContainer)*8);
bitD->ptr -= bitD->bitsConsumed >> 3;
bitD->bitsConsumed &= 7;
bitD->bitContainer = MEM_readLEST(bitD->ptr);
return BIT_DStream_unfinished;
}
/*! BIT_reloadDStream() :
* Refill `bitD` from buffer previously set in BIT_initDStream() .
* This function is safe, it guarantees it will not read beyond src buffer.
* @return : status of `BIT_DStream_t` internal register.
* when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
MEM_STATIC FORCE_INLINE_ATTR BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
{
if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8)) /* overflow detected, like end of stream */
return BIT_DStream_overflow;
if (bitD->ptr >= bitD->limitPtr) {
return BIT_reloadDStreamFast(bitD);
}
if (bitD->ptr == bitD->start) {
if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
return BIT_DStream_completed;
}
/* start < ptr < limitPtr */
{ U32 nbBytes = bitD->bitsConsumed >> 3;
BIT_DStream_status result = BIT_DStream_unfinished;
if (bitD->ptr - nbBytes < bitD->start) {
nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */
result = BIT_DStream_endOfBuffer;
}
bitD->ptr -= nbBytes;
bitD->bitsConsumed -= nbBytes*8;
bitD->bitContainer = MEM_readLEST(bitD->ptr); /* reminder : srcSize > sizeof(bitD->bitContainer), otherwise bitD->ptr == bitD->start */
return result;
}
}
/*! BIT_endOfDStream() :
* @return : 1 if DStream has _exactly_ reached its end (all bits consumed).
*/
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream)
{
return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer)*8));
}
#if defined (__cplusplus)
}
#endif
#endif /* BITSTREAM_H_MODULE */

358
ext/zstd/lib/common/compiler.h Normal file
View File

@@ -0,0 +1,358 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPILER_H
#define ZSTD_COMPILER_H
#include "portability_macros.h"
/*-*******************************************************
* Compiler specifics
*********************************************************/
/* force inlining */
#if !defined(ZSTD_NO_INLINE)
#if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
# define INLINE_KEYWORD inline
#else
# define INLINE_KEYWORD
#endif
#if defined(__GNUC__) || defined(__ICCARM__)
# define FORCE_INLINE_ATTR __attribute__((always_inline))
#elif defined(_MSC_VER)
# define FORCE_INLINE_ATTR __forceinline
#else
# define FORCE_INLINE_ATTR
#endif
#else
#define INLINE_KEYWORD
#define FORCE_INLINE_ATTR
#endif
/**
On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
This explicitly marks such functions as __cdecl so that the code will still compile
if a CC other than __cdecl has been made the default.
*/
#if defined(_MSC_VER)
# define WIN_CDECL __cdecl
#else
# define WIN_CDECL
#endif
/**
* FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
* parameters. They must be inlined for the compiler to eliminate the constant
* branches.
*/
#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
/**
* HINT_INLINE is used to help the compiler generate better code. It is *not*
* used for "templates", so it can be tweaked based on the compilers
* performance.
*
* gcc-4.8 and gcc-4.9 have been shown to benefit from leaving off the
* always_inline attribute.
*
* clang up to 5.0.0 (trunk) benefit tremendously from the always_inline
* attribute.
*/
#if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 4 && __GNUC_MINOR__ >= 8 && __GNUC__ < 5
# define HINT_INLINE static INLINE_KEYWORD
#else
# define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR
#endif
/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
#if defined(__GNUC__)
# define UNUSED_ATTR __attribute__((unused))
#else
# define UNUSED_ATTR
#endif
/* force no inlining */
#ifdef _MSC_VER
# define FORCE_NOINLINE static __declspec(noinline)
#else
# if defined(__GNUC__) || defined(__ICCARM__)
# define FORCE_NOINLINE static __attribute__((__noinline__))
# else
# define FORCE_NOINLINE static
# endif
#endif
/* target attribute */
#if defined(__GNUC__) || defined(__ICCARM__)
# define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
#else
# define TARGET_ATTRIBUTE(target)
#endif
/* Target attribute for BMI2 dynamic dispatch.
* Enable lzcnt, bmi, and bmi2.
* We test for bmi1 & bmi2. lzcnt is included in bmi1.
*/
#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2")
/* prefetch
* can be disabled, by declaring NO_PREFETCH build macro */
#if defined(NO_PREFETCH)
# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
#else
# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) /* _mm_prefetch() is not defined outside of x86/x64 */
# include <mmintrin.h> /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
# define PREFETCH_L1(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
# define PREFETCH_L2(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
# elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
# define PREFETCH_L1(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
# define PREFETCH_L2(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
# elif defined(__aarch64__)
# define PREFETCH_L1(ptr) __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr)))
# define PREFETCH_L2(ptr) __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr)))
# else
# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
# endif
#endif /* NO_PREFETCH */
#define CACHELINE_SIZE 64
#define PREFETCH_AREA(p, s) { \
const char* const _ptr = (const char*)(p); \
size_t const _size = (size_t)(s); \
size_t _pos; \
for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) { \
PREFETCH_L2(_ptr + _pos); \
} \
}
/* vectorization
* older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax,
* and some compilers, like Intel ICC and MCST LCC, do not support it at all. */
#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__)
# if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
# define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
# else
# define DONT_VECTORIZE _Pragma("GCC optimize(\"no-tree-vectorize\")")
# endif
#else
# define DONT_VECTORIZE
#endif
/* Tell the compiler that a branch is likely or unlikely.
* Only use these macros if it causes the compiler to generate better code.
* If you can remove a LIKELY/UNLIKELY annotation without speed changes in gcc
* and clang, please do.
*/
#if defined(__GNUC__)
#define LIKELY(x) (__builtin_expect((x), 1))
#define UNLIKELY(x) (__builtin_expect((x), 0))
#else
#define LIKELY(x) (x)
#define UNLIKELY(x) (x)
#endif
#if __has_builtin(__builtin_unreachable) || (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 5)))
# define ZSTD_UNREACHABLE { assert(0), __builtin_unreachable(); }
#else
# define ZSTD_UNREACHABLE { assert(0); }
#endif
/* disable warnings */
#ifdef _MSC_VER /* Visual Studio */
# include <intrin.h> /* For Visual 2005 */
# pragma warning(disable : 4100) /* disable: C4100: unreferenced formal parameter */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */
# pragma warning(disable : 4214) /* disable: C4214: non-int bitfields */
# pragma warning(disable : 4324) /* disable: C4324: padded structure */
#endif
/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/
#ifndef STATIC_BMI2
# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))
# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
# define STATIC_BMI2 1
# endif
# elif defined(__BMI2__) && defined(__x86_64__) && defined(__GNUC__)
# define STATIC_BMI2 1
# endif
#endif
#ifndef STATIC_BMI2
#define STATIC_BMI2 0
#endif
/* compile time determination of SIMD support */
#if !defined(ZSTD_NO_INTRINSICS)
# if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2))
# define ZSTD_ARCH_X86_SSE2
# endif
# if defined(__ARM_NEON) || defined(_M_ARM64)
# define ZSTD_ARCH_ARM_NEON
# endif
#
# if defined(ZSTD_ARCH_X86_SSE2)
# include <emmintrin.h>
# elif defined(ZSTD_ARCH_ARM_NEON)
# include <arm_neon.h>
# endif
#endif
/* C-language Attributes are added in C23. */
#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute)
# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x)
#else
# define ZSTD_HAS_C_ATTRIBUTE(x) 0
#endif
/* Only use C++ attributes in C++. Some compilers report support for C++
* attributes when compiling with C.
*/
#if defined(__cplusplus) && defined(__has_cpp_attribute)
# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
#else
# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0
#endif
/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute.
* - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough
* - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough
* - Else: __attribute__((__fallthrough__))
*/
#ifndef ZSTD_FALLTHROUGH
# if ZSTD_HAS_C_ATTRIBUTE(fallthrough)
# define ZSTD_FALLTHROUGH [[fallthrough]]
# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough)
# define ZSTD_FALLTHROUGH [[fallthrough]]
# elif __has_attribute(__fallthrough__)
/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon
* gcc complains about: a label can only be part of a statement and a declaration is not a statement.
*/
# define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__))
# else
# define ZSTD_FALLTHROUGH
# endif
#endif
/*-**************************************************************
* Alignment check
*****************************************************************/
/* this test was initially positioned in mem.h,
* but this file is removed (or replaced) for linux kernel
* so it's now hosted in compiler.h,
* which remains valid for both user & kernel spaces.
*/
#ifndef ZSTD_ALIGNOF
# if defined(__GNUC__) || defined(_MSC_VER)
/* covers gcc, clang & MSVC */
/* note : this section must come first, before C11,
* due to a limitation in the kernel source generator */
# define ZSTD_ALIGNOF(T) __alignof(T)
# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)
/* C11 support */
# include <stdalign.h>
# define ZSTD_ALIGNOF(T) alignof(T)
# else
/* No known support for alignof() - imperfect backup */
# define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T))
# endif
#endif /* ZSTD_ALIGNOF */
/*-**************************************************************
* Sanitizer
*****************************************************************/
/* Issue #3240 reports an ASAN failure on an llvm-mingw build. Out of an
* abundance of caution, disable our custom poisoning on mingw. */
#ifdef __MINGW32__
#ifndef ZSTD_ASAN_DONT_POISON_WORKSPACE
#define ZSTD_ASAN_DONT_POISON_WORKSPACE 1
#endif
#ifndef ZSTD_MSAN_DONT_POISON_WORKSPACE
#define ZSTD_MSAN_DONT_POISON_WORKSPACE 1
#endif
#endif
#if ZSTD_MEMORY_SANITIZER && !defined(ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support msan provide sanitizers/msan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
#include <stddef.h> /* size_t */
#define ZSTD_DEPS_NEED_STDINT
#include "zstd_deps.h" /* intptr_t */
/* Make memory region fully initialized (without changing its contents). */
void __msan_unpoison(const volatile void *a, size_t size);
/* Make memory region fully uninitialized (without changing its contents).
This is a legacy interface that does not update origin information. Use
__msan_allocated_memory() instead. */
void __msan_poison(const volatile void *a, size_t size);
/* Returns the offset of the first (at least partially) poisoned byte in the
memory range, or -1 if the whole range is good. */
intptr_t __msan_test_shadow(const volatile void *x, size_t size);
/* Print shadow and origin for the memory range to stderr in a human-readable
format. */
void __msan_print_shadow(const volatile void *x, size_t size);
#endif
#if ZSTD_ADDRESS_SANITIZER && !defined(ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support asan provide sanitizers/asan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
#include <stddef.h> /* size_t */
/**
* Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
*
* This memory must be previously allocated by your program. Instrumented
* code is forbidden from accessing addresses in this region until it is
* unpoisoned. This function is not guaranteed to poison the entire region -
* it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
* alignment restrictions.
*
* \note This function is not thread-safe because no two threads can poison or
* unpoison memory in the same memory region simultaneously.
*
* \param addr Start of memory region.
* \param size Size of memory region. */
void __asan_poison_memory_region(void const volatile *addr, size_t size);
/**
* Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
*
* This memory must be previously allocated by your program. Accessing
* addresses in this region is allowed until this region is poisoned again.
* This function could unpoison a super-region of <c>[addr, addr+size)</c> due
* to ASan alignment restrictions.
*
* \note This function is not thread-safe because no two threads can
* poison or unpoison memory in the same memory region simultaneously.
*
* \param addr Start of memory region.
* \param size Size of memory region. */
void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
#endif
#endif /* ZSTD_COMPILER_H */

213
ext/zstd/lib/common/cpu.h Normal file
View File

@@ -0,0 +1,213 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMMON_CPU_H
#define ZSTD_COMMON_CPU_H
/**
* Implementation taken from folly/CpuId.h
* https://github.com/facebook/folly/blob/master/folly/CpuId.h
*/
#include "mem.h"
#ifdef _MSC_VER
#include <intrin.h>
#endif
typedef struct {
U32 f1c;
U32 f1d;
U32 f7b;
U32 f7c;
} ZSTD_cpuid_t;
MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
U32 f1c = 0;
U32 f1d = 0;
U32 f7b = 0;
U32 f7c = 0;
#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
int reg[4];
__cpuid((int*)reg, 0);
{
int const n = reg[0];
if (n >= 1) {
__cpuid((int*)reg, 1);
f1c = (U32)reg[2];
f1d = (U32)reg[3];
}
if (n >= 7) {
__cpuidex((int*)reg, 7, 0);
f7b = (U32)reg[1];
f7c = (U32)reg[2];
}
}
#elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__)
/* The following block like the normal cpuid branch below, but gcc
* reserves ebx for use of its pic register so we must specially
* handle the save and restore to avoid clobbering the register
*/
U32 n;
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"popl %%ebx\n\t"
: "=a"(n)
: "a"(0)
: "ecx", "edx");
if (n >= 1) {
U32 f1a;
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"popl %%ebx\n\t"
: "=a"(f1a), "=c"(f1c), "=d"(f1d)
: "a"(1));
}
if (n >= 7) {
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"movl %%ebx, %%eax\n\t"
"popl %%ebx"
: "=a"(f7b), "=c"(f7c)
: "a"(7), "c"(0)
: "edx");
}
#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386__)
U32 n;
__asm__("cpuid" : "=a"(n) : "a"(0) : "ebx", "ecx", "edx");
if (n >= 1) {
U32 f1a;
__asm__("cpuid" : "=a"(f1a), "=c"(f1c), "=d"(f1d) : "a"(1) : "ebx");
}
if (n >= 7) {
U32 f7a;
__asm__("cpuid"
: "=a"(f7a), "=b"(f7b), "=c"(f7c)
: "a"(7), "c"(0)
: "edx");
}
#endif
{
ZSTD_cpuid_t cpuid;
cpuid.f1c = f1c;
cpuid.f1d = f1d;
cpuid.f7b = f7b;
cpuid.f7c = f7c;
return cpuid;
}
}
#define X(name, r, bit) \
MEM_STATIC int ZSTD_cpuid_##name(ZSTD_cpuid_t const cpuid) { \
return ((cpuid.r) & (1U << bit)) != 0; \
}
/* cpuid(1): Processor Info and Feature Bits. */
#define C(name, bit) X(name, f1c, bit)
C(sse3, 0)
C(pclmuldq, 1)
C(dtes64, 2)
C(monitor, 3)
C(dscpl, 4)
C(vmx, 5)
C(smx, 6)
C(eist, 7)
C(tm2, 8)
C(ssse3, 9)
C(cnxtid, 10)
C(fma, 12)
C(cx16, 13)
C(xtpr, 14)
C(pdcm, 15)
C(pcid, 17)
C(dca, 18)
C(sse41, 19)
C(sse42, 20)
C(x2apic, 21)
C(movbe, 22)
C(popcnt, 23)
C(tscdeadline, 24)
C(aes, 25)
C(xsave, 26)
C(osxsave, 27)
C(avx, 28)
C(f16c, 29)
C(rdrand, 30)
#undef C
#define D(name, bit) X(name, f1d, bit)
D(fpu, 0)
D(vme, 1)
D(de, 2)
D(pse, 3)
D(tsc, 4)
D(msr, 5)
D(pae, 6)
D(mce, 7)
D(cx8, 8)
D(apic, 9)
D(sep, 11)
D(mtrr, 12)
D(pge, 13)
D(mca, 14)
D(cmov, 15)
D(pat, 16)
D(pse36, 17)
D(psn, 18)
D(clfsh, 19)
D(ds, 21)
D(acpi, 22)
D(mmx, 23)
D(fxsr, 24)
D(sse, 25)
D(sse2, 26)
D(ss, 27)
D(htt, 28)
D(tm, 29)
D(pbe, 31)
#undef D
/* cpuid(7): Extended Features. */
#define B(name, bit) X(name, f7b, bit)
B(bmi1, 3)
B(hle, 4)
B(avx2, 5)
B(smep, 7)
B(bmi2, 8)
B(erms, 9)
B(invpcid, 10)
B(rtm, 11)
B(mpx, 14)
B(avx512f, 16)
B(avx512dq, 17)
B(rdseed, 18)
B(adx, 19)
B(smap, 20)
B(avx512ifma, 21)
B(pcommit, 22)
B(clflushopt, 23)
B(clwb, 24)
B(avx512pf, 26)
B(avx512er, 27)
B(avx512cd, 28)
B(sha, 29)
B(avx512bw, 30)
B(avx512vl, 31)
#undef B
#define C(name, bit) X(name, f7c, bit)
C(prefetchwt1, 0)
C(avx512vbmi, 1)
#undef C
#undef X
#endif /* ZSTD_COMMON_CPU_H */

24
ext/zstd/lib/common/debug.c vendored Normal file
View File

@@ -0,0 +1,24 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/*
* This module only hosts one global variable
* which can be used to dynamically influence the verbosity of traces,
* such as DEBUGLOG and RAWLOG
*/
#include "debug.h"
int g_debuglevel = DEBUGLEVEL;

107
ext/zstd/lib/common/debug.h Normal file
View File

@@ -0,0 +1,107 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/*
* The purpose of this header is to enable debug functions.
* They regroup assert(), DEBUGLOG() and RAWLOG() for run-time,
* and DEBUG_STATIC_ASSERT() for compile-time.
*
* By default, DEBUGLEVEL==0, which means run-time debug is disabled.
*
* Level 1 enables assert() only.
* Starting level 2, traces can be generated and pushed to stderr.
* The higher the level, the more verbose the traces.
*
* It's possible to dynamically adjust level using variable g_debug_level,
* which is only declared if DEBUGLEVEL>=2,
* and is a global variable, not multi-thread protected (use with care)
*/
#ifndef DEBUG_H_12987983217
#define DEBUG_H_12987983217
#if defined (__cplusplus)
extern "C" {
#endif
/* static assert is triggered at compile time, leaving no runtime artefact.
* static assert only works with compile-time constants.
* Also, this variant can only be used inside a function. */
#define DEBUG_STATIC_ASSERT(c) (void)sizeof(char[(c) ? 1 : -1])
/* DEBUGLEVEL is expected to be defined externally,
* typically through compiler command line.
* Value must be a number. */
#ifndef DEBUGLEVEL
# define DEBUGLEVEL 0
#endif
/* recommended values for DEBUGLEVEL :
* 0 : release mode, no debug, all run-time checks disabled
* 1 : enables assert() only, no display
* 2 : reserved, for currently active debug path
* 3 : events once per object lifetime (CCtx, CDict, etc.)
* 4 : events once per frame
* 5 : events once per block
* 6 : events once per sequence (verbose)
* 7+: events at every position (*very* verbose)
*
* It's generally inconvenient to output traces > 5.
* In which case, it's possible to selectively trigger high verbosity levels
* by modifying g_debug_level.
*/
#if (DEBUGLEVEL>=1)
# define ZSTD_DEPS_NEED_ASSERT
# include "zstd_deps.h"
#else
# ifndef assert /* assert may be already defined, due to prior #include <assert.h> */
# define assert(condition) ((void)0) /* disable assert (default) */
# endif
#endif
#if (DEBUGLEVEL>=2)
# define ZSTD_DEPS_NEED_IO
# include "zstd_deps.h"
extern int g_debuglevel; /* the variable is only declared,
it actually lives in debug.c,
and is shared by the whole process.
It's not thread-safe.
It's useful when enabling very verbose levels
on selective conditions (such as position in src) */
# define RAWLOG(l, ...) { \
if (l<=g_debuglevel) { \
ZSTD_DEBUG_PRINT(__VA_ARGS__); \
} }
# define DEBUGLOG(l, ...) { \
if (l<=g_debuglevel) { \
ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \
ZSTD_DEBUG_PRINT(" \n"); \
} }
#else
# define RAWLOG(l, ...) {} /* disabled */
# define DEBUGLOG(l, ...) {} /* disabled */
#endif
#if defined (__cplusplus)
}
#endif
#endif /* DEBUG_H_12987983217 */

340
ext/zstd/lib/common/entropy_common.c vendored Normal file
View File

@@ -0,0 +1,340 @@
/* ******************************************************************
* Common functions of New Generation Entropy library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* *************************************
* Dependencies
***************************************/
#include "mem.h"
#include "error_private.h" /* ERR_*, ERROR */
#define FSE_STATIC_LINKING_ONLY /* FSE_MIN_TABLELOG */
#include "fse.h"
#include "huf.h"
#include "bits.h" /* ZSDT_highbit32, ZSTD_countTrailingZeros32 */
/*=== Version ===*/
unsigned FSE_versionNumber(void) { return FSE_VERSION_NUMBER; }
/*=== Error Management ===*/
unsigned FSE_isError(size_t code) { return ERR_isError(code); }
const char* FSE_getErrorName(size_t code) { return ERR_getErrorName(code); }
unsigned HUF_isError(size_t code) { return ERR_isError(code); }
const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); }
/*-**************************************************************
* FSE NCount encoding-decoding
****************************************************************/
FORCE_INLINE_TEMPLATE
size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
const BYTE* const istart = (const BYTE*) headerBuffer;
const BYTE* const iend = istart + hbSize;
const BYTE* ip = istart;
int nbBits;
int remaining;
int threshold;
U32 bitStream;
int bitCount;
unsigned charnum = 0;
unsigned const maxSV1 = *maxSVPtr + 1;
int previous0 = 0;
if (hbSize < 8) {
/* This function only works when hbSize >= 8 */
char buffer[8] = {0};
ZSTD_memcpy(buffer, headerBuffer, hbSize);
{ size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr,
buffer, sizeof(buffer));
if (FSE_isError(countSize)) return countSize;
if (countSize > hbSize) return ERROR(corruption_detected);
return countSize;
} }
assert(hbSize >= 8);
/* init */
ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */
bitStream = MEM_readLE32(ip);
nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */
if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
bitStream >>= 4;
bitCount = 4;
*tableLogPtr = nbBits;
remaining = (1<<nbBits)+1;
threshold = 1<<nbBits;
nbBits++;
for (;;) {
if (previous0) {
/* Count the number of repeats. Each time the
* 2-bit repeat code is 0b11 there is another
* repeat.
* Avoid UB by setting the high bit to 1.
*/
int repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
while (repeats >= 12) {
charnum += 3 * 12;
if (LIKELY(ip <= iend-7)) {
ip += 3;
} else {
bitCount -= (int)(8 * (iend - 7 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
}
charnum += 3 * repeats;
bitStream >>= 2 * repeats;
bitCount += 2 * repeats;
/* Add the final repeat which isn't 0b11. */
assert((bitStream & 3) < 3);
charnum += bitStream & 3;
bitCount += 2;
/* This is an error, but break and return an error
* at the end, because returning out of a loop makes
* it harder for the compiler to optimize.
*/
if (charnum >= maxSV1) break;
/* We don't need to set the normalized count to 0
* because we already memset the whole buffer to 0.
*/
if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
assert((bitCount >> 3) <= 3); /* For first condition to work */
ip += bitCount>>3;
bitCount &= 7;
} else {
bitCount -= (int)(8 * (iend - 4 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
}
{
int const max = (2*threshold-1) - remaining;
int count;
if ((bitStream & (threshold-1)) < (U32)max) {
count = bitStream & (threshold-1);
bitCount += nbBits-1;
} else {
count = bitStream & (2*threshold-1);
if (count >= threshold) count -= max;
bitCount += nbBits;
}
count--; /* extra accuracy */
/* When it matters (small blocks), this is a
* predictable branch, because we don't use -1.
*/
if (count >= 0) {
remaining -= count;
} else {
assert(count == -1);
remaining += count;
}
normalizedCounter[charnum++] = (short)count;
previous0 = !count;
assert(threshold > 1);
if (remaining < threshold) {
/* This branch can be folded into the
* threshold update condition because we
* know that threshold > 1.
*/
if (remaining <= 1) break;
nbBits = ZSTD_highbit32(remaining) + 1;
threshold = 1 << (nbBits - 1);
}
if (charnum >= maxSV1) break;
if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
ip += bitCount>>3;
bitCount &= 7;
} else {
bitCount -= (int)(8 * (iend - 4 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
} }
if (remaining != 1) return ERROR(corruption_detected);
/* Only possible when there are too many zeros. */
if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall);
if (bitCount > 32) return ERROR(corruption_detected);
*maxSVPtr = charnum-1;
ip += (bitCount+7)>>3;
return ip-istart;
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t FSE_readNCount_body_default(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#if DYNAMIC_BMI2
BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#endif
size_t FSE_readNCount_bmi2(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize, int bmi2)
{
#if DYNAMIC_BMI2
if (bmi2) {
return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#endif
(void)bmi2;
return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
size_t FSE_readNCount(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0);
}
/*! HUF_readStats() :
Read compact Huffman tree, saved by HUF_writeCTable().
`huffWeight` is destination buffer.
`rankStats` is assumed to be a table of at least HUF_TABLELOG_MAX U32.
@return : size read from `src` , or an error Code .
Note : Needed by HUF_readCTable() and HUF_readDTableX?() .
*/
size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize)
{
U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32];
return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* flags */ 0);
}
FORCE_INLINE_TEMPLATE size_t
HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize,
int bmi2)
{
U32 weightTotal;
const BYTE* ip = (const BYTE*) src;
size_t iSize;
size_t oSize;
if (!srcSize) return ERROR(srcSize_wrong);
iSize = ip[0];
/* ZSTD_memset(huffWeight, 0, hwSize); *//* is not necessary, even though some analyzer complain ... */
if (iSize >= 128) { /* special header */
oSize = iSize - 127;
iSize = ((oSize+1)/2);
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
if (oSize >= hwSize) return ERROR(corruption_detected);
ip += 1;
{ U32 n;
for (n=0; n<oSize; n+=2) {
huffWeight[n] = ip[n/2] >> 4;
huffWeight[n+1] = ip[n/2] & 15;
} } }
else { /* header compressed with FSE (normal case) */
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
/* max (hwSize-1) values decoded, as last one is implied */
oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2);
if (FSE_isError(oSize)) return oSize;
}
/* collect weight stats */
ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
weightTotal = 0;
{ U32 n; for (n=0; n<oSize; n++) {
if (huffWeight[n] > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
rankStats[huffWeight[n]]++;
weightTotal += (1 << huffWeight[n]) >> 1;
} }
if (weightTotal == 0) return ERROR(corruption_detected);
/* get last non-null symbol weight (implied, total must be 2^n) */
{ U32 const tableLog = ZSTD_highbit32(weightTotal) + 1;
if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
*tableLogPtr = tableLog;
/* determine last weight */
{ U32 const total = 1 << tableLog;
U32 const rest = total - weightTotal;
U32 const verif = 1 << ZSTD_highbit32(rest);
U32 const lastWeight = ZSTD_highbit32(rest) + 1;
if (verif != rest) return ERROR(corruption_detected); /* last value must be a clean power of 2 */
huffWeight[oSize] = (BYTE)lastWeight;
rankStats[lastWeight]++;
} }
/* check tree construction validity */
if ((rankStats[1] < 2) || (rankStats[1] & 1)) return ERROR(corruption_detected); /* by construction : at least 2 elts of rank 1, must be even */
/* results */
*nbSymbolsPtr = (U32)(oSize+1);
return iSize+1;
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize)
{
return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0);
}
#if DYNAMIC_BMI2
static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize)
{
return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1);
}
#endif
size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize,
int flags)
{
#if DYNAMIC_BMI2
if (flags & HUF_flags_bmi2) {
return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}
#endif
(void)flags;
return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}

63
ext/zstd/lib/common/error_private.c vendored Normal file
View File

@@ -0,0 +1,63 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* The purpose of this file is to have a single list of error strings embedded in binary */
#include "error_private.h"
const char* ERR_getErrorString(ERR_enum code)
{
#ifdef ZSTD_STRIP_ERROR_STRINGS
(void)code;
return "Error strings stripped";
#else
static const char* const notErrorCode = "Unspecified error code";
switch( code )
{
case PREFIX(no_error): return "No error detected";
case PREFIX(GENERIC): return "Error (generic)";
case PREFIX(prefix_unknown): return "Unknown frame descriptor";
case PREFIX(version_unsupported): return "Version not supported";
case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding";
case PREFIX(corruption_detected): return "Data corruption detected";
case PREFIX(checksum_wrong): return "Restored data doesn't match checksum";
case PREFIX(literals_headerWrong): return "Header of Literals' block doesn't respect format specification";
case PREFIX(parameter_unsupported): return "Unsupported parameter";
case PREFIX(parameter_combination_unsupported): return "Unsupported combination of parameters";
case PREFIX(parameter_outOfBound): return "Parameter is out of bound";
case PREFIX(init_missing): return "Context should be init first";
case PREFIX(memory_allocation): return "Allocation error : not enough memory";
case PREFIX(workSpace_tooSmall): return "workSpace buffer is not large enough";
case PREFIX(stage_wrong): return "Operation not authorized at current processing stage";
case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
case PREFIX(dictionary_wrong): return "Dictionary mismatch";
case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples";
case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
case PREFIX(srcSize_wrong): return "Src size is incorrect";
case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer";
case PREFIX(noForwardProgress_destFull): return "Operation made no progress over multiple calls, due to output buffer being full";
case PREFIX(noForwardProgress_inputEmpty): return "Operation made no progress over multiple calls, due to input being empty";
/* following error codes are not stable and may be removed or changed in a future version */
case PREFIX(frameIndex_tooLarge): return "Frame index is too large";
case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking";
case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong";
case PREFIX(srcBuffer_wrong): return "Source buffer is wrong";
case PREFIX(sequenceProducer_failed): return "Block-level external sequence producer returned an error code";
case PREFIX(externalSequences_invalid): return "External sequences are not valid";
case PREFIX(maxCode):
default: return notErrorCode;
}
#endif
}

159
ext/zstd/lib/common/error_private.h Normal file
View File

@@ -0,0 +1,159 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* Note : this module is expected to remain private, do not expose it */
#ifndef ERROR_H_MODULE
#define ERROR_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************************
* Dependencies
******************************************/
#include "../zstd_errors.h" /* enum list */
#include "compiler.h"
#include "debug.h"
#include "zstd_deps.h" /* size_t */
/* ****************************************
* Compiler-specific
******************************************/
#if defined(__GNUC__)
# define ERR_STATIC static __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define ERR_STATIC static inline
#elif defined(_MSC_VER)
# define ERR_STATIC static __inline
#else
# define ERR_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/*-****************************************
* Customization (error_public.h)
******************************************/
typedef ZSTD_ErrorCode ERR_enum;
#define PREFIX(name) ZSTD_error_##name
/*-****************************************
* Error codes handling
******************************************/
#undef ERROR /* already defined on Visual Studio */
#define ERROR(name) ZSTD_ERROR(name)
#define ZSTD_ERROR(name) ((size_t)-PREFIX(name))
ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
ERR_STATIC ERR_enum ERR_getErrorCode(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); }
/* check and forward error code */
#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e
#define CHECK_F(f) { CHECK_V_F(_var_err__, f); }
/*-****************************************
* Error Strings
******************************************/
const char* ERR_getErrorString(ERR_enum code); /* error_private.c */
ERR_STATIC const char* ERR_getErrorName(size_t code)
{
return ERR_getErrorString(ERR_getErrorCode(code));
}
/**
* Ignore: this is an internal helper.
*
* This is a helper function to help force C99-correctness during compilation.
* Under strict compilation modes, variadic macro arguments can't be empty.
* However, variadic function arguments can be. Using a function therefore lets
* us statically check that at least one (string) argument was passed,
* independent of the compilation flags.
*/
static INLINE_KEYWORD UNUSED_ATTR
void _force_has_format_string(const char *format, ...) {
(void)format;
}
/**
* Ignore: this is an internal helper.
*
* We want to force this function invocation to be syntactically correct, but
* we don't want to force runtime evaluation of its arguments.
*/
#define _FORCE_HAS_FORMAT_STRING(...) \
if (0) { \
_force_has_format_string(__VA_ARGS__); \
}
#define ERR_QUOTE(str) #str
/**
* Return the specified error if the condition evaluates to true.
*
* In debug modes, prints additional information.
* In order to do that (particularly, printing the conditional that failed),
* this can't just wrap RETURN_ERROR().
*/
#define RETURN_ERROR_IF(cond, err, ...) \
if (cond) { \
RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", \
__FILE__, __LINE__, ERR_QUOTE(cond), ERR_QUOTE(ERROR(err))); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return ERROR(err); \
}
/**
* Unconditionally return the specified error.
*
* In debug modes, prints additional information.
*/
#define RETURN_ERROR(err, ...) \
do { \
RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", \
__FILE__, __LINE__, ERR_QUOTE(ERROR(err))); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return ERROR(err); \
} while(0);
/**
* If the provided expression evaluates to an error code, returns that error code.
*
* In debug modes, prints additional information.
*/
#define FORWARD_IF_ERROR(err, ...) \
do { \
size_t const err_code = (err); \
if (ERR_isError(err_code)) { \
RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", \
__FILE__, __LINE__, ERR_QUOTE(err), ERR_getErrorName(err_code)); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return err_code; \
} \
} while(0);
#if defined (__cplusplus)
}
#endif
#endif /* ERROR_H_MODULE */

639
ext/zstd/lib/common/fse.h Normal file
View File

@@ -0,0 +1,639 @@
/* ******************************************************************
* FSE : Finite State Entropy codec
* Public Prototypes declaration
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#if defined (__cplusplus)
extern "C" {
#endif
#ifndef FSE_H
#define FSE_H
/*-*****************************************
* Dependencies
******************************************/
#include "zstd_deps.h" /* size_t, ptrdiff_t */
/*-*****************************************
* FSE_PUBLIC_API : control library symbols visibility
******************************************/
#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
# define FSE_PUBLIC_API __attribute__ ((visibility ("default")))
#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */
# define FSE_PUBLIC_API __declspec(dllexport)
#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
# define FSE_PUBLIC_API __declspec(dllimport) /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
#else
# define FSE_PUBLIC_API
#endif
/*------ Version ------*/
#define FSE_VERSION_MAJOR 0
#define FSE_VERSION_MINOR 9
#define FSE_VERSION_RELEASE 0
#define FSE_LIB_VERSION FSE_VERSION_MAJOR.FSE_VERSION_MINOR.FSE_VERSION_RELEASE
#define FSE_QUOTE(str) #str
#define FSE_EXPAND_AND_QUOTE(str) FSE_QUOTE(str)
#define FSE_VERSION_STRING FSE_EXPAND_AND_QUOTE(FSE_LIB_VERSION)
#define FSE_VERSION_NUMBER (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE)
FSE_PUBLIC_API unsigned FSE_versionNumber(void); /**< library version number; to be used when checking dll version */
/*-*****************************************
* Tool functions
******************************************/
FSE_PUBLIC_API size_t FSE_compressBound(size_t size); /* maximum compressed size */
/* Error Management */
FSE_PUBLIC_API unsigned FSE_isError(size_t code); /* tells if a return value is an error code */
FSE_PUBLIC_API const char* FSE_getErrorName(size_t code); /* provides error code string (useful for debugging) */
/*-*****************************************
* FSE detailed API
******************************************/
/*!
FSE_compress() does the following:
1. count symbol occurrence from source[] into table count[] (see hist.h)
2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog)
3. save normalized counters to memory buffer using writeNCount()
4. build encoding table 'CTable' from normalized counters
5. encode the data stream using encoding table 'CTable'
FSE_decompress() does the following:
1. read normalized counters with readNCount()
2. build decoding table 'DTable' from normalized counters
3. decode the data stream using decoding table 'DTable'
The following API allows targeting specific sub-functions for advanced tasks.
For example, it's possible to compress several blocks using the same 'CTable',
or to save and provide normalized distribution using external method.
*/
/* *** COMPRESSION *** */
/*! FSE_optimalTableLog():
dynamically downsize 'tableLog' when conditions are met.
It saves CPU time, by using smaller tables, while preserving or even improving compression ratio.
@return : recommended tableLog (necessarily <= 'maxTableLog') */
FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
/*! FSE_normalizeCount():
normalize counts so that sum(count[]) == Power_of_2 (2^tableLog)
'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1).
useLowProbCount is a boolean parameter which trades off compressed size for
faster header decoding. When it is set to 1, the compressed data will be slightly
smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be
faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0
is a good default, since header deserialization makes a big speed difference.
Otherwise, useLowProbCount=1 is a good default, since the speed difference is small.
@return : tableLog,
or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog,
const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount);
/*! FSE_NCountWriteBound():
Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'.
Typically useful for allocation purpose. */
FSE_PUBLIC_API size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog);
/*! FSE_writeNCount():
Compactly save 'normalizedCounter' into 'buffer'.
@return : size of the compressed table,
or an errorCode, which can be tested using FSE_isError(). */
FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
const short* normalizedCounter,
unsigned maxSymbolValue, unsigned tableLog);
/*! Constructor and Destructor of FSE_CTable.
Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
typedef unsigned FSE_CTable; /* don't allocate that. It's only meant to be more restrictive than void* */
/*! FSE_buildCTable():
Builds `ct`, which must be already allocated, using FSE_createCTable().
@return : 0, or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
/*! FSE_compress_usingCTable():
Compress `src` using `ct` into `dst` which must be already allocated.
@return : size of compressed data (<= `dstCapacity`),
or 0 if compressed data could not fit into `dst`,
or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct);
/*!
Tutorial :
----------
The first step is to count all symbols. FSE_count() does this job very fast.
Result will be saved into 'count', a table of unsigned int, which must be already allocated, and have 'maxSymbolValuePtr[0]+1' cells.
'src' is a table of bytes of size 'srcSize'. All values within 'src' MUST be <= maxSymbolValuePtr[0]
maxSymbolValuePtr[0] will be updated, with its real value (necessarily <= original value)
FSE_count() will return the number of occurrence of the most frequent symbol.
This can be used to know if there is a single symbol within 'src', and to quickly evaluate its compressibility.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
The next step is to normalize the frequencies.
FSE_normalizeCount() will ensure that sum of frequencies is == 2 ^'tableLog'.
It also guarantees a minimum of 1 to any Symbol with frequency >= 1.
You can use 'tableLog'==0 to mean "use default tableLog value".
If you are unsure of which tableLog value to use, you can ask FSE_optimalTableLog(),
which will provide the optimal valid tableLog given sourceSize, maxSymbolValue, and a user-defined maximum (0 means "default").
The result of FSE_normalizeCount() will be saved into a table,
called 'normalizedCounter', which is a table of signed short.
'normalizedCounter' must be already allocated, and have at least 'maxSymbolValue+1' cells.
The return value is tableLog if everything proceeded as expected.
It is 0 if there is a single symbol within distribution.
If there is an error (ex: invalid tableLog value), the function will return an ErrorCode (which can be tested using FSE_isError()).
'normalizedCounter' can be saved in a compact manner to a memory area using FSE_writeNCount().
'buffer' must be already allocated.
For guaranteed success, buffer size must be at least FSE_headerBound().
The result of the function is the number of bytes written into 'buffer'.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError(); ex : buffer size too small).
'normalizedCounter' can then be used to create the compression table 'CTable'.
The space required by 'CTable' must be already allocated, using FSE_createCTable().
You can then use FSE_buildCTable() to fill 'CTable'.
If there is an error, both functions will return an ErrorCode (which can be tested using FSE_isError()).
'CTable' can then be used to compress 'src', with FSE_compress_usingCTable().
Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize'
The function returns the size of compressed data (without header), necessarily <= `dstCapacity`.
If it returns '0', compressed data could not fit into 'dst'.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
*/
/* *** DECOMPRESSION *** */
/*! FSE_readNCount():
Read compactly saved 'normalizedCounter' from 'rBuffer'.
@return : size read from 'rBuffer',
or an errorCode, which can be tested using FSE_isError().
maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */
FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter,
unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
const void* rBuffer, size_t rBuffSize);
/*! FSE_readNCount_bmi2():
* Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise.
*/
FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter,
unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
const void* rBuffer, size_t rBuffSize, int bmi2);
typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */
/*!
Tutorial :
----------
(Note : these functions only decompress FSE-compressed blocks.
If block is uncompressed, use memcpy() instead
If block is a single repeated byte, use memset() instead )
The first step is to obtain the normalized frequencies of symbols.
This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount().
'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short.
In practice, that means it's necessary to know 'maxSymbolValue' beforehand,
or size the table to handle worst case situations (typically 256).
FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'.
The result of FSE_readNCount() is the number of bytes read from 'rBuffer'.
Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that.
If there is an error, the function will return an error code, which can be tested using FSE_isError().
The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'.
This is performed by the function FSE_buildDTable().
The space required by 'FSE_DTable' must be already allocated using FSE_createDTable().
If there is an error, the function will return an error code, which can be tested using FSE_isError().
`FSE_DTable` can then be used to decompress `cSrc`, with FSE_decompress_usingDTable().
`cSrcSize` must be strictly correct, otherwise decompression will fail.
FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=`dstCapacity`).
If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small)
*/
#endif /* FSE_H */
#if defined(FSE_STATIC_LINKING_ONLY) && !defined(FSE_H_FSE_STATIC_LINKING_ONLY)
#define FSE_H_FSE_STATIC_LINKING_ONLY
/* *** Dependency *** */
#include "bitstream.h"
/* *****************************************
* Static allocation
*******************************************/
/* FSE buffer bounds */
#define FSE_NCOUNTBOUND 512
#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
#define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
/* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */
#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2))
#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog)))
/* or use the size to malloc() space directly. Pay attention to alignment restrictions though */
#define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue) (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable))
#define FSE_DTABLE_SIZE(maxTableLog) (FSE_DTABLE_SIZE_U32(maxTableLog) * sizeof(FSE_DTable))
/* *****************************************
* FSE advanced API
***************************************** */
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
/**< same as FSE_optimalTableLog(), which used `minus==2` */
size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue);
/**< build a fake FSE_CTable, designed to compress always the same symbolValue */
/* FSE_buildCTable_wksp() :
* Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
* `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`.
* See FSE_buildCTable_wksp() for breakdown of workspace usage.
*/
#define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */)
#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog))
size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8)
#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned))
FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */
#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned))
size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2);
/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)`.
* Set bmi2 to 1 if your CPU supports BMI2 or 0 if it doesn't */
typedef enum {
FSE_repeat_none, /**< Cannot use the previous table */
FSE_repeat_check, /**< Can use the previous table but it must be checked */
FSE_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} FSE_repeat;
/* *****************************************
* FSE symbol compression API
*******************************************/
/*!
This API consists of small unitary functions, which highly benefit from being inlined.
Hence their body are included in next section.
*/
typedef struct {
ptrdiff_t value;
const void* stateTable;
const void* symbolTT;
unsigned stateLog;
} FSE_CState_t;
static void FSE_initCState(FSE_CState_t* CStatePtr, const FSE_CTable* ct);
static void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* CStatePtr, unsigned symbol);
static void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* CStatePtr);
/**<
These functions are inner components of FSE_compress_usingCTable().
They allow the creation of custom streams, mixing multiple tables and bit sources.
A key property to keep in mind is that encoding and decoding are done **in reverse direction**.
So the first symbol you will encode is the last you will decode, like a LIFO stack.
You will need a few variables to track your CStream. They are :
FSE_CTable ct; // Provided by FSE_buildCTable()
BIT_CStream_t bitStream; // bitStream tracking structure
FSE_CState_t state; // State tracking structure (can have several)
The first thing to do is to init bitStream and state.
size_t errorCode = BIT_initCStream(&bitStream, dstBuffer, maxDstSize);
FSE_initCState(&state, ct);
Note that BIT_initCStream() can produce an error code, so its result should be tested, using FSE_isError();
You can then encode your input data, byte after byte.
FSE_encodeSymbol() outputs a maximum of 'tableLog' bits at a time.
Remember decoding will be done in reverse direction.
FSE_encodeByte(&bitStream, &state, symbol);
At any time, you can also add any bit sequence.
Note : maximum allowed nbBits is 25, for compatibility with 32-bits decoders
BIT_addBits(&bitStream, bitField, nbBits);
The above methods don't commit data to memory, they just store it into local register, for speed.
Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
Writing data to memory is a manual operation, performed by the flushBits function.
BIT_flushBits(&bitStream);
Your last FSE encoding operation shall be to flush your last state value(s).
FSE_flushState(&bitStream, &state);
Finally, you must close the bitStream.
The function returns the size of CStream in bytes.
If data couldn't fit into dstBuffer, it will return a 0 ( == not compressible)
If there is an error, it returns an errorCode (which can be tested using FSE_isError()).
size_t size = BIT_closeCStream(&bitStream);
*/
/* *****************************************
* FSE symbol decompression API
*******************************************/
typedef struct {
size_t state;
const void* table; /* precise table may vary, depending on U16 */
} FSE_DState_t;
static void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt);
static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr);
/**<
Let's now decompose FSE_decompress_usingDTable() into its unitary components.
You will decode FSE-encoded symbols from the bitStream,
and also any other bitFields you put in, **in reverse order**.
You will need a few variables to track your bitStream. They are :
BIT_DStream_t DStream; // Stream context
FSE_DState_t DState; // State context. Multiple ones are possible
FSE_DTable* DTablePtr; // Decoding table, provided by FSE_buildDTable()
The first thing to do is to init the bitStream.
errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize);
You should then retrieve your initial state(s)
(in reverse flushing order if you have several ones) :
errorCode = FSE_initDState(&DState, &DStream, DTablePtr);
You can then decode your data, symbol after symbol.
For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'.
Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out).
unsigned char symbol = FSE_decodeSymbol(&DState, &DStream);
You can retrieve any bitfield you eventually stored into the bitStream (in reverse order)
Note : maximum allowed nbBits is 25, for 32-bits compatibility
size_t bitField = BIT_readBits(&DStream, nbBits);
All above operations only read from local register (which size depends on size_t).
Refueling the register from memory is manually performed by the reload method.
endSignal = FSE_reloadDStream(&DStream);
BIT_reloadDStream() result tells if there is still some more data to read from DStream.
BIT_DStream_unfinished : there is still some data left into the DStream.
BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled.
BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed.
BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted.
When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop,
to properly detect the exact end of stream.
After each decoded symbol, check if DStream is fully consumed using this simple test :
BIT_reloadDStream(&DStream) >= BIT_DStream_completed
When it's done, verify decompression is fully completed, by checking both DStream and the relevant states.
Checking if DStream has reached its end is performed by :
BIT_endOfDStream(&DStream);
Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible.
FSE_endOfDState(&DState);
*/
/* *****************************************
* FSE unsafe API
*******************************************/
static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
/* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */
/* *****************************************
* Implementation of inlined functions
*******************************************/
typedef struct {
int deltaFindState;
U32 deltaNbBits;
} FSE_symbolCompressionTransform; /* total 8 bytes */
MEM_STATIC void FSE_initCState(FSE_CState_t* statePtr, const FSE_CTable* ct)
{
const void* ptr = ct;
const U16* u16ptr = (const U16*) ptr;
const U32 tableLog = MEM_read16(ptr);
statePtr->value = (ptrdiff_t)1<<tableLog;
statePtr->stateTable = u16ptr+2;
statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1);
statePtr->stateLog = tableLog;
}
/*! FSE_initCState2() :
* Same as FSE_initCState(), but the first symbol to include (which will be the last to be read)
* uses the smallest state value possible, saving the cost of this symbol */
MEM_STATIC void FSE_initCState2(FSE_CState_t* statePtr, const FSE_CTable* ct, U32 symbol)
{
FSE_initCState(statePtr, ct);
{ const FSE_symbolCompressionTransform symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
const U16* stateTable = (const U16*)(statePtr->stateTable);
U32 nbBitsOut = (U32)((symbolTT.deltaNbBits + (1<<15)) >> 16);
statePtr->value = (nbBitsOut << 16) - symbolTT.deltaNbBits;
statePtr->value = stateTable[(statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
}
}
MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol)
{
FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
const U16* const stateTable = (const U16*)(statePtr->stateTable);
U32 const nbBitsOut = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16);
BIT_addBits(bitC, statePtr->value, nbBitsOut);
statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
}
MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr)
{
BIT_addBits(bitC, statePtr->value, statePtr->stateLog);
BIT_flushBits(bitC);
}
/* FSE_getMaxNbBits() :
* Approximate maximum cost of a symbol, in bits.
* Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
* note 1 : assume symbolValue is valid (<= maxSymbolValue)
* note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue)
{
const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16;
}
/* FSE_bitCost() :
* Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits)
* note 1 : assume symbolValue is valid (<= maxSymbolValue)
* note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
MEM_STATIC U32 FSE_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog)
{
const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16;
U32 const threshold = (minNbBits+1) << 16;
assert(tableLog < 16);
assert(accuracyLog < 31-tableLog); /* ensure enough room for renormalization double shift */
{ U32 const tableSize = 1 << tableLog;
U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize);
U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog; /* linear interpolation (very approximate) */
U32 const bitMultiplier = 1 << accuracyLog;
assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold);
assert(normalizedDeltaFromThreshold <= bitMultiplier);
return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold;
}
}
/* ====== Decompression ====== */
typedef struct {
U16 tableLog;
U16 fastMode;
} FSE_DTableHeader; /* sizeof U32 */
typedef struct
{
unsigned short newState;
unsigned char symbol;
unsigned char nbBits;
} FSE_decode_t; /* size == U32 */
MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt)
{
const void* ptr = dt;
const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr;
DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
BIT_reloadDStream(bitD);
DStatePtr->table = dt + 1;
}
MEM_STATIC BYTE FSE_peekSymbol(const FSE_DState_t* DStatePtr)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
return DInfo.symbol;
}
MEM_STATIC void FSE_updateState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
size_t const lowBits = BIT_readBits(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
}
MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
BYTE const symbol = DInfo.symbol;
size_t const lowBits = BIT_readBits(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
/*! FSE_decodeSymbolFast() :
unsafe, only works if no symbol has a probability > 50% */
MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
BYTE const symbol = DInfo.symbol;
size_t const lowBits = BIT_readBitsFast(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
{
return DStatePtr->state == 0;
}
#ifndef FSE_COMMONDEFS_ONLY
/* **************************************************************
* Tuning parameters
****************************************************************/
/*!MEMORY_USAGE :
* Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
* Increasing memory usage improves compression ratio
* Reduced memory usage can improve speed, due to cache effect
* Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */
#ifndef FSE_MAX_MEMORY_USAGE
# define FSE_MAX_MEMORY_USAGE 14
#endif
#ifndef FSE_DEFAULT_MEMORY_USAGE
# define FSE_DEFAULT_MEMORY_USAGE 13
#endif
#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE)
# error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE"
#endif
/*!FSE_MAX_SYMBOL_VALUE :
* Maximum symbol value authorized.
* Required for proper stack allocation */
#ifndef FSE_MAX_SYMBOL_VALUE
# define FSE_MAX_SYMBOL_VALUE 255
#endif
/* **************************************************************
* template functions type & suffix
****************************************************************/
#define FSE_FUNCTION_TYPE BYTE
#define FSE_FUNCTION_EXTENSION
#define FSE_DECODE_TYPE FSE_decode_t
#endif /* !FSE_COMMONDEFS_ONLY */
/* ***************************************************************
* Constants
*****************************************************************/
#define FSE_MAX_TABLELOG (FSE_MAX_MEMORY_USAGE-2)
#define FSE_MAX_TABLESIZE (1U<<FSE_MAX_TABLELOG)
#define FSE_MAXTABLESIZE_MASK (FSE_MAX_TABLESIZE-1)
#define FSE_DEFAULT_TABLELOG (FSE_DEFAULT_MEMORY_USAGE-2)
#define FSE_MIN_TABLELOG 5
#define FSE_TABLELOG_ABSOLUTE_MAX 15
#if FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX
# error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported"
#endif
#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3)
#endif /* FSE_STATIC_LINKING_ONLY */
#if defined (__cplusplus)
}
#endif

311
ext/zstd/lib/common/fse_decompress.c vendored Normal file
View File

@@ -0,0 +1,311 @@
/* ******************************************************************
* FSE : Finite State Entropy decoder
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* **************************************************************
* Includes
****************************************************************/
#include "debug.h" /* assert */
#include "bitstream.h"
#include "compiler.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#include "error_private.h"
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h"
#include "bits.h" /* ZSTD_highbit32 */
/* **************************************************************
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
#define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */
/* **************************************************************
* Templates
****************************************************************/
/*
designed to be included
for type-specific functions (template emulation in C)
Objective is to write these functions only once, for improved maintenance
*/
/* safety checks */
#ifndef FSE_FUNCTION_EXTENSION
# error "FSE_FUNCTION_EXTENSION must be defined"
#endif
#ifndef FSE_FUNCTION_TYPE
# error "FSE_FUNCTION_TYPE must be defined"
#endif
/* Function names */
#define FSE_CAT(X,Y) X##Y
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
{
void* const tdPtr = dt+1; /* because *dt is unsigned, 32-bits aligned on 32-bits */
FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr);
U16* symbolNext = (U16*)workSpace;
BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1);
U32 const maxSV1 = maxSymbolValue + 1;
U32 const tableSize = 1 << tableLog;
U32 highThreshold = tableSize-1;
/* Sanity Checks */
if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge);
if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge);
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
/* Init, lay down lowprob symbols */
{ FSE_DTableHeader DTableH;
DTableH.tableLog = (U16)tableLog;
DTableH.fastMode = 1;
{ S16 const largeLimit= (S16)(1 << (tableLog-1));
U32 s;
for (s=0; s<maxSV1; s++) {
if (normalizedCounter[s]==-1) {
tableDecode[highThreshold--].symbol = (FSE_FUNCTION_TYPE)s;
symbolNext[s] = 1;
} else {
if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
symbolNext[s] = normalizedCounter[s];
} } }
ZSTD_memcpy(dt, &DTableH, sizeof(DTableH));
}
/* Spread symbols */
if (highThreshold == tableSize - 1) {
size_t const tableMask = tableSize-1;
size_t const step = FSE_TABLESTEP(tableSize);
/* First lay down the symbols in order.
* We use a uint64_t to lay down 8 bytes at a time. This reduces branch
* misses since small blocks generally have small table logs, so nearly
* all symbols have counts <= 8. We ensure we have 8 bytes at the end of
* our buffer to handle the over-write.
*/
{
U64 const add = 0x0101010101010101ull;
size_t pos = 0;
U64 sv = 0;
U32 s;
for (s=0; s<maxSV1; ++s, sv += add) {
int i;
int const n = normalizedCounter[s];
MEM_write64(spread + pos, sv);
for (i = 8; i < n; i += 8) {
MEM_write64(spread + pos + i, sv);
}
pos += n;
}
}
/* Now we spread those positions across the table.
* The benefit of doing it in two stages is that we avoid the
* variable size inner loop, which caused lots of branch misses.
* Now we can run through all the positions without any branch misses.
* We unroll the loop twice, since that is what empirically worked best.
*/
{
size_t position = 0;
size_t s;
size_t const unroll = 2;
assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
for (s = 0; s < (size_t)tableSize; s += unroll) {
size_t u;
for (u = 0; u < unroll; ++u) {
size_t const uPosition = (position + (u * step)) & tableMask;
tableDecode[uPosition].symbol = spread[s + u];
}
position = (position + (unroll * step)) & tableMask;
}
assert(position == 0);
}
} else {
U32 const tableMask = tableSize-1;
U32 const step = FSE_TABLESTEP(tableSize);
U32 s, position = 0;
for (s=0; s<maxSV1; s++) {
int i;
for (i=0; i<normalizedCounter[s]; i++) {
tableDecode[position].symbol = (FSE_FUNCTION_TYPE)s;
position = (position + step) & tableMask;
while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */
} }
if (position!=0) return ERROR(GENERIC); /* position must reach all cells once, otherwise normalizedCounter is incorrect */
}
/* Build Decoding table */
{ U32 u;
for (u=0; u<tableSize; u++) {
FSE_FUNCTION_TYPE const symbol = (FSE_FUNCTION_TYPE)(tableDecode[u].symbol);
U32 const nextState = symbolNext[symbol]++;
tableDecode[u].nbBits = (BYTE) (tableLog - ZSTD_highbit32(nextState) );
tableDecode[u].newState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
} }
return 0;
}
size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
{
return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize);
}
#ifndef FSE_COMMONDEFS_ONLY
/*-*******************************************************
* Decompression (Byte symbols)
*********************************************************/
FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
void* dst, size_t maxDstSize,
const void* cSrc, size_t cSrcSize,
const FSE_DTable* dt, const unsigned fast)
{
BYTE* const ostart = (BYTE*) dst;
BYTE* op = ostart;
BYTE* const omax = op + maxDstSize;
BYTE* const olimit = omax-3;
BIT_DStream_t bitD;
FSE_DState_t state1;
FSE_DState_t state2;
/* Init */
CHECK_F(BIT_initDStream(&bitD, cSrc, cSrcSize));
FSE_initDState(&state1, &bitD, dt);
FSE_initDState(&state2, &bitD, dt);
#define FSE_GETSYMBOL(statePtr) fast ? FSE_decodeSymbolFast(statePtr, &bitD) : FSE_decodeSymbol(statePtr, &bitD)
/* 4 symbols per loop */
for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) & (op<olimit) ; op+=4) {
op[0] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[1] = FSE_GETSYMBOL(&state2);
if (FSE_MAX_TABLELOG*4+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
{ if (BIT_reloadDStream(&bitD) > BIT_DStream_unfinished) { op+=2; break; } }
op[2] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[3] = FSE_GETSYMBOL(&state2);
}
/* tail */
/* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */
while (1) {
if (op>(omax-2)) return ERROR(dstSize_tooSmall);
*op++ = FSE_GETSYMBOL(&state1);
if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
*op++ = FSE_GETSYMBOL(&state2);
break;
}
if (op>(omax-2)) return ERROR(dstSize_tooSmall);
*op++ = FSE_GETSYMBOL(&state2);
if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
*op++ = FSE_GETSYMBOL(&state1);
break;
} }
return op-ostart;
}
typedef struct {
short ncount[FSE_MAX_SYMBOL_VALUE + 1];
FSE_DTable dtable[1]; /* Dynamically sized */
} FSE_DecompressWksp;
FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body(
void* dst, size_t dstCapacity,
const void* cSrc, size_t cSrcSize,
unsigned maxLog, void* workSpace, size_t wkspSize,
int bmi2)
{
const BYTE* const istart = (const BYTE*)cSrc;
const BYTE* ip = istart;
unsigned tableLog;
unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace;
DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0);
if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC);
/* normal FSE decoding mode */
{
size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2);
if (FSE_isError(NCountLength)) return NCountLength;
if (tableLog > maxLog) return ERROR(tableLog_tooLarge);
assert(NCountLength <= cSrcSize);
ip += NCountLength;
cSrcSize -= NCountLength;
}
if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge);
assert(sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog) <= wkspSize);
workSpace = (BYTE*)workSpace + sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) );
{
const void* ptr = wksp->dtable;
const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
const U32 fastMode = DTableH->fastMode;
/* select fast mode (static) */
if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1);
return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0);
}
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
{
return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0);
}
#if DYNAMIC_BMI2
BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
{
return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1);
}
#endif
size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2)
{
#if DYNAMIC_BMI2
if (bmi2) {
return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
}
#endif
(void)bmi2;
return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
}
#endif /* FSE_COMMONDEFS_ONLY */

273
ext/zstd/lib/common/huf.h Normal file
View File

@@ -0,0 +1,273 @@
/* ******************************************************************
* huff0 huffman codec,
* part of Finite State Entropy library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#if defined (__cplusplus)
extern "C" {
#endif
#ifndef HUF_H_298734234
#define HUF_H_298734234
/* *** Dependencies *** */
#include "zstd_deps.h" /* size_t */
#include "mem.h" /* U32 */
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
/* *** Tool functions *** */
#define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
/* Error Management */
unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */
#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
/* *** Constants *** */
#define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
#define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */
#define HUF_SYMBOLVALUE_MAX 255
#define HUF_TABLELOG_ABSOLUTEMAX 12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
#if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
# error "HUF_TABLELOG_MAX is too large !"
#endif
/* ****************************************
* Static allocation
******************************************/
/* HUF buffer bounds */
#define HUF_CTABLEBOUND 129
#define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */
#define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
/* static allocation of HUF's Compression Table */
/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */
typedef size_t HUF_CElt; /* consider it an incomplete type */
#define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */
#define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))
#define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */
/* static allocation of HUF's DTable */
typedef U32 HUF_DTable;
#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog)))
#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
/* ****************************************
* Advanced decompression functions
******************************************/
/**
* Huffman flags bitset.
* For all flags, 0 is the default value.
*/
typedef enum {
/**
* If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime.
* Otherwise: Ignored.
*/
HUF_flags_bmi2 = (1 << 0),
/**
* If set: Test possible table depths to find the one that produces the smallest header + encoded size.
* If unset: Use heuristic to find the table depth.
*/
HUF_flags_optimalDepth = (1 << 1),
/**
* If set: If the previous table can encode the input, always reuse the previous table.
* If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
*/
HUF_flags_preferRepeat = (1 << 2),
/**
* If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress.
* If unset: Always histogram the entire input.
*/
HUF_flags_suspectUncompressible = (1 << 3),
/**
* If set: Don't use assembly implementations
* If unset: Allow using assembly implementations
*/
HUF_flags_disableAsm = (1 << 4),
/**
* If set: Don't use the fast decoding loop, always use the fallback decoding loop.
* If unset: Use the fast decoding loop when possible.
*/
HUF_flags_disableFast = (1 << 5)
} HUF_flags_e;
/* ****************************************
* HUF detailed API
* ****************************************/
#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
/*! HUF_compress() does the following:
* 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
* 2. (optional) refine tableLog using HUF_optimalTableLog()
* 3. build Huffman table from count using HUF_buildCTable()
* 4. save Huffman table to memory buffer using HUF_writeCTable()
* 5. encode the data stream using HUF_compress4X_usingCTable()
*
* The following API allows targeting specific sub-functions for advanced tasks.
* For example, it's possible to compress several blocks using the same 'CTable',
* or to save and regenerate 'CTable' using external methods.
*/
unsigned HUF_minTableLog(unsigned symbolCardinality);
unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
size_t wkspSize, HUF_CElt* table, const unsigned* count, int flags); /* table is used as scratch space for building and testing tables, not a return value */
size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
typedef enum {
HUF_repeat_none, /**< Cannot use the previous table */
HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
HUF_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} HUF_repeat;
/** HUF_compress4X_repeat() :
* Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
* If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
* If preferRepeat then the old table will always be used if valid.
* If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
/** HUF_buildCTable_wksp() :
* Same as HUF_buildCTable(), but using externally allocated scratch buffer.
* `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
*/
#define HUF_CTABLE_WORKSPACE_SIZE_U32 ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
size_t HUF_buildCTable_wksp (HUF_CElt* tree,
const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
void* workSpace, size_t wkspSize);
/*! HUF_readStats() :
* Read compact Huffman tree, saved by HUF_writeCTable().
* `huffWeight` is destination buffer.
* @return : size read from `src` , or an error Code .
* Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize);
/*! HUF_readStats_wksp() :
* Same as HUF_readStats() but takes an external workspace which must be
* 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE.
* If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
*/
#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)
#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))
size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workspace, size_t wkspSize,
int flags);
/** HUF_readCTable() :
* Loading a CTable saved with HUF_writeCTable() */
size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
/** HUF_getNbBitsFromCTable() :
* Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
* Note 1 : is not inlined, as HUF_CElt definition is private */
U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue);
/*
* HUF_decompress() does the following:
* 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
* 2. build Huffman table from save, using HUF_readDTableX?()
* 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
*/
/** HUF_selectDecoder() :
* Tells which decoder is likely to decode faster,
* based on a set of pre-computed metrics.
* @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
* Assumption : 0 < dstSize <= 128 KB */
U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
/**
* The minimum workspace size for the `workSpace` used in
* HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
*
* The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
* HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
* Buffer overflow errors may potentially occur if code modifications result in
* a required workspace size greater than that specified in the following
* macro.
*/
#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
/* ====================== */
/* single stream variants */
/* ====================== */
size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
/** HUF_compress1X_repeat() :
* Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
* If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
* If preferRepeat then the old table will always be used if valid.
* If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags); /**< double-symbols decoder */
#endif
/* BMI2 variants.
* If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
*/
size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#endif
size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#endif /* HUF_H_298734234 */
#if defined (__cplusplus)
}
#endif

435
ext/zstd/lib/common/mem.h Normal file
View File

@@ -0,0 +1,435 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef MEM_H_MODULE
#define MEM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/*-****************************************
* Dependencies
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
#include "compiler.h" /* __has_builtin */
#include "debug.h" /* DEBUG_STATIC_ASSERT */
#include "zstd_deps.h" /* ZSTD_memcpy */
/*-****************************************
* Compiler specifics
******************************************/
#if defined(_MSC_VER) /* Visual Studio */
# include <stdlib.h> /* _byteswap_ulong */
# include <intrin.h> /* _byteswap_* */
#endif
#if defined(__GNUC__)
# define MEM_STATIC static __inline __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define MEM_STATIC static inline
#elif defined(_MSC_VER)
# define MEM_STATIC static __inline
#else
# define MEM_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/*-**************************************************************
* Basic Types
*****************************************************************/
#if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# if defined(_AIX)
# include <inttypes.h>
# else
# include <stdint.h> /* intptr_t */
# endif
typedef uint8_t BYTE;
typedef uint8_t U8;
typedef int8_t S8;
typedef uint16_t U16;
typedef int16_t S16;
typedef uint32_t U32;
typedef int32_t S32;
typedef uint64_t U64;
typedef int64_t S64;
#else
# include <limits.h>
#if CHAR_BIT != 8
# error "this implementation requires char to be exactly 8-bit type"
#endif
typedef unsigned char BYTE;
typedef unsigned char U8;
typedef signed char S8;
#if USHRT_MAX != 65535
# error "this implementation requires short to be exactly 16-bit type"
#endif
typedef unsigned short U16;
typedef signed short S16;
#if UINT_MAX != 4294967295
# error "this implementation requires int to be exactly 32-bit type"
#endif
typedef unsigned int U32;
typedef signed int S32;
/* note : there are no limits defined for long long type in C90.
* limits exist in C99, however, in such case, <stdint.h> is preferred */
typedef unsigned long long U64;
typedef signed long long S64;
#endif
/*-**************************************************************
* Memory I/O API
*****************************************************************/
/*=== Static platform detection ===*/
MEM_STATIC unsigned MEM_32bits(void);
MEM_STATIC unsigned MEM_64bits(void);
MEM_STATIC unsigned MEM_isLittleEndian(void);
/*=== Native unaligned read/write ===*/
MEM_STATIC U16 MEM_read16(const void* memPtr);
MEM_STATIC U32 MEM_read32(const void* memPtr);
MEM_STATIC U64 MEM_read64(const void* memPtr);
MEM_STATIC size_t MEM_readST(const void* memPtr);
MEM_STATIC void MEM_write16(void* memPtr, U16 value);
MEM_STATIC void MEM_write32(void* memPtr, U32 value);
MEM_STATIC void MEM_write64(void* memPtr, U64 value);
/*=== Little endian unaligned read/write ===*/
MEM_STATIC U16 MEM_readLE16(const void* memPtr);
MEM_STATIC U32 MEM_readLE24(const void* memPtr);
MEM_STATIC U32 MEM_readLE32(const void* memPtr);
MEM_STATIC U64 MEM_readLE64(const void* memPtr);
MEM_STATIC size_t MEM_readLEST(const void* memPtr);
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val);
MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val);
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32);
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64);
MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val);
/*=== Big endian unaligned read/write ===*/
MEM_STATIC U32 MEM_readBE32(const void* memPtr);
MEM_STATIC U64 MEM_readBE64(const void* memPtr);
MEM_STATIC size_t MEM_readBEST(const void* memPtr);
MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32);
MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64);
MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val);
/*=== Byteswap ===*/
MEM_STATIC U32 MEM_swap32(U32 in);
MEM_STATIC U64 MEM_swap64(U64 in);
MEM_STATIC size_t MEM_swapST(size_t in);
/*-**************************************************************
* Memory I/O Implementation
*****************************************************************/
/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
* Method 0 : always use `memcpy()`. Safe and portable.
* Method 1 : Use compiler extension to set unaligned access.
* Method 2 : direct access. This method is portable but violate C standard.
* It can generate buggy code on targets depending on alignment.
* Default : method 1 if supported, else method 0
*/
#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# ifdef __GNUC__
# define MEM_FORCE_MEMORY_ACCESS 1
# endif
#endif
MEM_STATIC unsigned MEM_32bits(void) { return sizeof(size_t)==4; }
MEM_STATIC unsigned MEM_64bits(void) { return sizeof(size_t)==8; }
MEM_STATIC unsigned MEM_isLittleEndian(void)
{
#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
return 1;
#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
return 0;
#elif defined(__clang__) && __LITTLE_ENDIAN__
return 1;
#elif defined(__clang__) && __BIG_ENDIAN__
return 0;
#elif defined(_MSC_VER) && (_M_AMD64 || _M_IX86)
return 1;
#elif defined(__DMC__) && defined(_M_IX86)
return 1;
#else
const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
return one.c[0];
#endif
}
#if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2)
/* violates C standard, by lying on structure alignment.
Only use if no other choice to achieve best performance on target platform */
MEM_STATIC U16 MEM_read16(const void* memPtr) { return *(const U16*) memPtr; }
MEM_STATIC U32 MEM_read32(const void* memPtr) { return *(const U32*) memPtr; }
MEM_STATIC U64 MEM_read64(const void* memPtr) { return *(const U64*) memPtr; }
MEM_STATIC size_t MEM_readST(const void* memPtr) { return *(const size_t*) memPtr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
typedef __attribute__((aligned(1))) U16 unalign16;
typedef __attribute__((aligned(1))) U32 unalign32;
typedef __attribute__((aligned(1))) U64 unalign64;
typedef __attribute__((aligned(1))) size_t unalignArch;
MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
#else
/* default method, safe and standard.
can sometimes prove slower */
MEM_STATIC U16 MEM_read16(const void* memPtr)
{
U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U32 MEM_read32(const void* memPtr)
{
U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U64 MEM_read64(const void* memPtr)
{
U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC size_t MEM_readST(const void* memPtr)
{
size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC void MEM_write16(void* memPtr, U16 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write32(void* memPtr, U32 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write64(void* memPtr, U64 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
#endif /* MEM_FORCE_MEMORY_ACCESS */
MEM_STATIC U32 MEM_swap32_fallback(U32 in)
{
return ((in << 24) & 0xff000000 ) |
((in << 8) & 0x00ff0000 ) |
((in >> 8) & 0x0000ff00 ) |
((in >> 24) & 0x000000ff );
}
MEM_STATIC U32 MEM_swap32(U32 in)
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_ulong(in);
#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
|| (defined(__clang__) && __has_builtin(__builtin_bswap32))
return __builtin_bswap32(in);
#else
return MEM_swap32_fallback(in);
#endif
}
MEM_STATIC U64 MEM_swap64_fallback(U64 in)
{
return ((in << 56) & 0xff00000000000000ULL) |
((in << 40) & 0x00ff000000000000ULL) |
((in << 24) & 0x0000ff0000000000ULL) |
((in << 8) & 0x000000ff00000000ULL) |
((in >> 8) & 0x00000000ff000000ULL) |
((in >> 24) & 0x0000000000ff0000ULL) |
((in >> 40) & 0x000000000000ff00ULL) |
((in >> 56) & 0x00000000000000ffULL);
}
MEM_STATIC U64 MEM_swap64(U64 in)
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_uint64(in);
#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
|| (defined(__clang__) && __has_builtin(__builtin_bswap64))
return __builtin_bswap64(in);
#else
return MEM_swap64_fallback(in);
#endif
}
MEM_STATIC size_t MEM_swapST(size_t in)
{
if (MEM_32bits())
return (size_t)MEM_swap32((U32)in);
else
return (size_t)MEM_swap64((U64)in);
}
/*=== Little endian r/w ===*/
MEM_STATIC U16 MEM_readLE16(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read16(memPtr);
else {
const BYTE* p = (const BYTE*)memPtr;
return (U16)(p[0] + (p[1]<<8));
}
}
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
{
if (MEM_isLittleEndian()) {
MEM_write16(memPtr, val);
} else {
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val;
p[1] = (BYTE)(val>>8);
}
}
MEM_STATIC U32 MEM_readLE24(const void* memPtr)
{
return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16);
}
MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val)
{
MEM_writeLE16(memPtr, (U16)val);
((BYTE*)memPtr)[2] = (BYTE)(val>>16);
}
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read32(memPtr);
else
return MEM_swap32(MEM_read32(memPtr));
}
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
MEM_write32(memPtr, val32);
else
MEM_write32(memPtr, MEM_swap32(val32));
}
MEM_STATIC U64 MEM_readLE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read64(memPtr);
else
return MEM_swap64(MEM_read64(memPtr));
}
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
MEM_write64(memPtr, val64);
else
MEM_write64(memPtr, MEM_swap64(val64));
}
MEM_STATIC size_t MEM_readLEST(const void* memPtr)
{
if (MEM_32bits())
return (size_t)MEM_readLE32(memPtr);
else
return (size_t)MEM_readLE64(memPtr);
}
MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val)
{
if (MEM_32bits())
MEM_writeLE32(memPtr, (U32)val);
else
MEM_writeLE64(memPtr, (U64)val);
}
/*=== Big endian r/w ===*/
MEM_STATIC U32 MEM_readBE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_swap32(MEM_read32(memPtr));
else
return MEM_read32(memPtr);
}
MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
MEM_write32(memPtr, MEM_swap32(val32));
else
MEM_write32(memPtr, val32);
}
MEM_STATIC U64 MEM_readBE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_swap64(MEM_read64(memPtr));
else
return MEM_read64(memPtr);
}
MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
MEM_write64(memPtr, MEM_swap64(val64));
else
MEM_write64(memPtr, val64);
}
MEM_STATIC size_t MEM_readBEST(const void* memPtr)
{
if (MEM_32bits())
return (size_t)MEM_readBE32(memPtr);
else
return (size_t)MEM_readBE64(memPtr);
}
MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val)
{
if (MEM_32bits())
MEM_writeBE32(memPtr, (U32)val);
else
MEM_writeBE64(memPtr, (U64)val);
}
/* code only tested on 32 and 64 bits systems */
MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
#if defined (__cplusplus)
}
#endif
#endif /* MEM_H_MODULE */

371
ext/zstd/lib/common/pool.c vendored Normal file
View File

@@ -0,0 +1,371 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* ====== Dependencies ======= */
#include "../common/allocations.h" /* ZSTD_customCalloc, ZSTD_customFree */
#include "zstd_deps.h" /* size_t */
#include "debug.h" /* assert */
#include "pool.h"
/* ====== Compiler specifics ====== */
#if defined(_MSC_VER)
# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */
#endif
#ifdef ZSTD_MULTITHREAD
#include "threading.h" /* pthread adaptation */
/* A job is a function and an opaque argument */
typedef struct POOL_job_s {
POOL_function function;
void *opaque;
} POOL_job;
struct POOL_ctx_s {
ZSTD_customMem customMem;
/* Keep track of the threads */
ZSTD_pthread_t* threads;
size_t threadCapacity;
size_t threadLimit;
/* The queue is a circular buffer */
POOL_job *queue;
size_t queueHead;
size_t queueTail;
size_t queueSize;
/* The number of threads working on jobs */
size_t numThreadsBusy;
/* Indicates if the queue is empty */
int queueEmpty;
/* The mutex protects the queue */
ZSTD_pthread_mutex_t queueMutex;
/* Condition variable for pushers to wait on when the queue is full */
ZSTD_pthread_cond_t queuePushCond;
/* Condition variables for poppers to wait on when the queue is empty */
ZSTD_pthread_cond_t queuePopCond;
/* Indicates if the queue is shutting down */
int shutdown;
};
/* POOL_thread() :
* Work thread for the thread pool.
* Waits for jobs and executes them.
* @returns : NULL on failure else non-null.
*/
static void* POOL_thread(void* opaque) {
POOL_ctx* const ctx = (POOL_ctx*)opaque;
if (!ctx) { return NULL; }
for (;;) {
/* Lock the mutex and wait for a non-empty queue or until shutdown */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
while ( ctx->queueEmpty
|| (ctx->numThreadsBusy >= ctx->threadLimit) ) {
if (ctx->shutdown) {
/* even if !queueEmpty, (possible if numThreadsBusy >= threadLimit),
* a few threads will be shutdown while !queueEmpty,
* but enough threads will remain active to finish the queue */
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return opaque;
}
ZSTD_pthread_cond_wait(&ctx->queuePopCond, &ctx->queueMutex);
}
/* Pop a job off the queue */
{ POOL_job const job = ctx->queue[ctx->queueHead];
ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
ctx->numThreadsBusy++;
ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
/* Unlock the mutex, signal a pusher, and run the job */
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
job.function(job.opaque);
/* If the intended queue size was 0, signal after finishing job */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
ctx->numThreadsBusy--;
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
} /* for (;;) */
assert(0); /* Unreachable */
}
/* ZSTD_createThreadPool() : public access point */
POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
return POOL_create (numThreads, 0);
}
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
ZSTD_customMem customMem)
{
POOL_ctx* ctx;
/* Check parameters */
if (!numThreads) { return NULL; }
/* Allocate the context and zero initialize */
ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem);
if (!ctx) { return NULL; }
/* Initialize the job queue.
* It needs one extra space since one space is wasted to differentiate
* empty and full queues.
*/
ctx->queueSize = queueSize + 1;
ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
ctx->queueHead = 0;
ctx->queueTail = 0;
ctx->numThreadsBusy = 0;
ctx->queueEmpty = 1;
{
int error = 0;
error |= ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL);
error |= ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL);
error |= ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL);
if (error) { POOL_free(ctx); return NULL; }
}
ctx->shutdown = 0;
/* Allocate space for the thread handles */
ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
ctx->threadCapacity = 0;
ctx->customMem = customMem;
/* Check for errors */
if (!ctx->threads || !ctx->queue) { POOL_free(ctx); return NULL; }
/* Initialize the threads */
{ size_t i;
for (i = 0; i < numThreads; ++i) {
if (ZSTD_pthread_create(&ctx->threads[i], NULL, &POOL_thread, ctx)) {
ctx->threadCapacity = i;
POOL_free(ctx);
return NULL;
} }
ctx->threadCapacity = numThreads;
ctx->threadLimit = numThreads;
}
return ctx;
}
/*! POOL_join() :
Shutdown the queue, wake any sleeping threads, and join all of the threads.
*/
static void POOL_join(POOL_ctx* ctx) {
/* Shut down the queue */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
ctx->shutdown = 1;
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
/* Wake up sleeping threads */
ZSTD_pthread_cond_broadcast(&ctx->queuePushCond);
ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
/* Join all of the threads */
{ size_t i;
for (i = 0; i < ctx->threadCapacity; ++i) {
ZSTD_pthread_join(ctx->threads[i]); /* note : could fail */
} }
}
void POOL_free(POOL_ctx *ctx) {
if (!ctx) { return; }
POOL_join(ctx);
ZSTD_pthread_mutex_destroy(&ctx->queueMutex);
ZSTD_pthread_cond_destroy(&ctx->queuePushCond);
ZSTD_pthread_cond_destroy(&ctx->queuePopCond);
ZSTD_customFree(ctx->queue, ctx->customMem);
ZSTD_customFree(ctx->threads, ctx->customMem);
ZSTD_customFree(ctx, ctx->customMem);
}
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx) {
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
}
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
POOL_free (pool);
}
size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
return sizeof(*ctx)
+ ctx->queueSize * sizeof(POOL_job)
+ ctx->threadCapacity * sizeof(ZSTD_pthread_t);
}
/* @return : 0 on success, 1 on error */
static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
{
if (numThreads <= ctx->threadCapacity) {
if (!numThreads) return 1;
ctx->threadLimit = numThreads;
return 0;
}
/* numThreads > threadCapacity */
{ ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
if (!threadPool) return 1;
/* replace existing thread pool */
ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
ZSTD_customFree(ctx->threads, ctx->customMem);
ctx->threads = threadPool;
/* Initialize additional threads */
{ size_t threadId;
for (threadId = ctx->threadCapacity; threadId < numThreads; ++threadId) {
if (ZSTD_pthread_create(&threadPool[threadId], NULL, &POOL_thread, ctx)) {
ctx->threadCapacity = threadId;
return 1;
} }
} }
/* successfully expanded */
ctx->threadCapacity = numThreads;
ctx->threadLimit = numThreads;
return 0;
}
/* @return : 0 on success, 1 on error */
int POOL_resize(POOL_ctx* ctx, size_t numThreads)
{
int result;
if (ctx==NULL) return 1;
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
result = POOL_resize_internal(ctx, numThreads);
ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return result;
}
/**
* Returns 1 if the queue is full and 0 otherwise.
*
* When queueSize is 1 (pool was created with an intended queueSize of 0),
* then a queue is empty if there is a thread free _and_ no job is waiting.
*/
static int isQueueFull(POOL_ctx const* ctx) {
if (ctx->queueSize > 1) {
return ctx->queueHead == ((ctx->queueTail + 1) % ctx->queueSize);
} else {
return (ctx->numThreadsBusy == ctx->threadLimit) ||
!ctx->queueEmpty;
}
}
static void
POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
{
POOL_job job;
job.function = function;
job.opaque = opaque;
assert(ctx != NULL);
if (ctx->shutdown) return;
ctx->queueEmpty = 0;
ctx->queue[ctx->queueTail] = job;
ctx->queueTail = (ctx->queueTail + 1) % ctx->queueSize;
ZSTD_pthread_cond_signal(&ctx->queuePopCond);
}
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque)
{
assert(ctx != NULL);
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
/* Wait until there is space in the queue for the new job */
while (isQueueFull(ctx) && (!ctx->shutdown)) {
ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
}
POOL_add_internal(ctx, function, opaque);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque)
{
assert(ctx != NULL);
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
if (isQueueFull(ctx)) {
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return 0;
}
POOL_add_internal(ctx, function, opaque);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return 1;
}
#else /* ZSTD_MULTITHREAD not defined */
/* ========================== */
/* No multi-threading support */
/* ========================== */
/* We don't need any data, but if it is empty, malloc() might return NULL. */
struct POOL_ctx_s {
int dummy;
};
static POOL_ctx g_poolCtx;
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
POOL_ctx*
POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem)
{
(void)numThreads;
(void)queueSize;
(void)customMem;
return &g_poolCtx;
}
void POOL_free(POOL_ctx* ctx) {
assert(!ctx || ctx == &g_poolCtx);
(void)ctx;
}
void POOL_joinJobs(POOL_ctx* ctx){
assert(!ctx || ctx == &g_poolCtx);
(void)ctx;
}
int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
(void)ctx; (void)numThreads;
return 0;
}
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) {
(void)ctx;
function(opaque);
}
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
(void)ctx;
function(opaque);
return 1;
}
size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
assert(ctx == &g_poolCtx);
return sizeof(*ctx);
}
#endif /* ZSTD_MULTITHREAD */

90
ext/zstd/lib/common/pool.h Normal file
View File

@@ -0,0 +1,90 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef POOL_H
#define POOL_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_deps.h"
#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_customMem */
#include "../zstd.h"
typedef struct POOL_ctx_s POOL_ctx;
/*! POOL_create() :
* Create a thread pool with at most `numThreads` threads.
* `numThreads` must be at least 1.
* The maximum number of queued jobs before blocking is `queueSize`.
* @return : POOL_ctx pointer on success, else NULL.
*/
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize);
POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
ZSTD_customMem customMem);
/*! POOL_free() :
* Free a thread pool returned by POOL_create().
*/
void POOL_free(POOL_ctx* ctx);
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx);
/*! POOL_resize() :
* Expands or shrinks pool's number of threads.
* This is more efficient than releasing + creating a new context,
* since it tries to preserve and re-use existing threads.
* `numThreads` must be at least 1.
* @return : 0 when resize was successful,
* !0 (typically 1) if there is an error.
* note : only numThreads can be resized, queueSize remains unchanged.
*/
int POOL_resize(POOL_ctx* ctx, size_t numThreads);
/*! POOL_sizeof() :
* @return threadpool memory usage
* note : compatible with NULL (returns 0 in this case)
*/
size_t POOL_sizeof(const POOL_ctx* ctx);
/*! POOL_function :
* The function type that can be added to a thread pool.
*/
typedef void (*POOL_function)(void*);
/*! POOL_add() :
* Add the job `function(opaque)` to the thread pool. `ctx` must be valid.
* Possibly blocks until there is room in the queue.
* Note : The function may be executed asynchronously,
* therefore, `opaque` must live until function has been completed.
*/
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
/*! POOL_tryAdd() :
* Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
* Returns immediately even if not (does not block).
* @return : 1 if successful, 0 if not.
*/
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque);
#if defined (__cplusplus)
}
#endif
#endif

156
ext/zstd/lib/common/portability_macros.h Normal file
View File

@@ -0,0 +1,156 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_PORTABILITY_MACROS_H
#define ZSTD_PORTABILITY_MACROS_H
/**
* This header file contains macro definitions to support portability.
* This header is shared between C and ASM code, so it MUST only
* contain macro definitions. It MUST not contain any C code.
*
* This header ONLY defines macros to detect platforms/feature support.
*
*/
/* compat. with non-clang compilers */
#ifndef __has_attribute
#define __has_attribute(x) 0
#endif
/* compat. with non-clang compilers */
#ifndef __has_builtin
# define __has_builtin(x) 0
#endif
/* compat. with non-clang compilers */
#ifndef __has_feature
# define __has_feature(x) 0
#endif
/* detects whether we are being compiled under msan */
#ifndef ZSTD_MEMORY_SANITIZER
# if __has_feature(memory_sanitizer)
# define ZSTD_MEMORY_SANITIZER 1
# else
# define ZSTD_MEMORY_SANITIZER 0
# endif
#endif
/* detects whether we are being compiled under asan */
#ifndef ZSTD_ADDRESS_SANITIZER
# if __has_feature(address_sanitizer)
# define ZSTD_ADDRESS_SANITIZER 1
# elif defined(__SANITIZE_ADDRESS__)
# define ZSTD_ADDRESS_SANITIZER 1
# else
# define ZSTD_ADDRESS_SANITIZER 0
# endif
#endif
/* detects whether we are being compiled under dfsan */
#ifndef ZSTD_DATAFLOW_SANITIZER
# if __has_feature(dataflow_sanitizer)
# define ZSTD_DATAFLOW_SANITIZER 1
# else
# define ZSTD_DATAFLOW_SANITIZER 0
# endif
#endif
/* Mark the internal assembly functions as hidden */
#ifdef __ELF__
# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func
#else
# define ZSTD_HIDE_ASM_FUNCTION(func)
#endif
/* Enable runtime BMI2 dispatch based on the CPU.
* Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
*/
#ifndef DYNAMIC_BMI2
#if ((defined(__clang__) && __has_attribute(__target__)) \
|| (defined(__GNUC__) \
&& (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
&& (defined(__x86_64__) || defined(_M_X64)) \
&& !defined(__BMI2__)
# define DYNAMIC_BMI2 1
#else
# define DYNAMIC_BMI2 0
#endif
#endif
/**
* Only enable assembly for GNUC compatible compilers,
* because other platforms may not support GAS assembly syntax.
*
* Only enable assembly for Linux / MacOS, other platforms may
* work, but they haven't been tested. This could likely be
* extended to BSD systems.
*
* Disable assembly when MSAN is enabled, because MSAN requires
* 100% of code to be instrumented to work.
*/
#if defined(__GNUC__)
# if defined(__linux__) || defined(__linux) || defined(__APPLE__)
# if ZSTD_MEMORY_SANITIZER
# define ZSTD_ASM_SUPPORTED 0
# elif ZSTD_DATAFLOW_SANITIZER
# define ZSTD_ASM_SUPPORTED 0
# else
# define ZSTD_ASM_SUPPORTED 1
# endif
# else
# define ZSTD_ASM_SUPPORTED 0
# endif
#else
# define ZSTD_ASM_SUPPORTED 0
#endif
/**
* Determines whether we should enable assembly for x86-64
* with BMI2.
*
* Enable if all of the following conditions hold:
* - ASM hasn't been explicitly disabled by defining ZSTD_DISABLE_ASM
* - Assembly is supported
* - We are compiling for x86-64 and either:
* - DYNAMIC_BMI2 is enabled
* - BMI2 is supported at compile time
*/
#if !defined(ZSTD_DISABLE_ASM) && \
ZSTD_ASM_SUPPORTED && \
defined(__x86_64__) && \
(DYNAMIC_BMI2 || defined(__BMI2__))
# define ZSTD_ENABLE_ASM_X86_64_BMI2 1
#else
# define ZSTD_ENABLE_ASM_X86_64_BMI2 0
#endif
/*
* For x86 ELF targets, add .note.gnu.property section for Intel CET in
* assembly sources when CET is enabled.
*
* Additionally, any function that may be called indirectly must begin
* with ZSTD_CET_ENDBRANCH.
*/
#if defined(__ELF__) && (defined(__x86_64__) || defined(__i386__)) \
&& defined(__has_include)
# if __has_include(<cet.h>)
# include <cet.h>
# define ZSTD_CET_ENDBRANCH _CET_ENDBR
# endif
#endif
#ifndef ZSTD_CET_ENDBRANCH
# define ZSTD_CET_ENDBRANCH
#endif
#endif /* ZSTD_PORTABILITY_MACROS_H */

176
ext/zstd/lib/common/threading.c vendored Normal file
View File

@@ -0,0 +1,176 @@
/**
* Copyright (c) 2016 Tino Reichardt
* All rights reserved.
*
* You can contact the author at:
* - zstdmt source repository: https://github.com/mcmilk/zstdmt
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/**
* This file will hold wrapper for systems, which do not support pthreads
*/
#include "threading.h"
/* create fake symbol to avoid empty translation unit warning */
int g_ZSTD_threading_useless_symbol;
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper
*/
/* === Dependencies === */
#include <process.h>
#include <errno.h>
/* === Implementation === */
typedef struct {
void* (*start_routine)(void*);
void* arg;
int initialized;
ZSTD_pthread_cond_t initialized_cond;
ZSTD_pthread_mutex_t initialized_mutex;
} ZSTD_thread_params_t;
static unsigned __stdcall worker(void *arg)
{
void* (*start_routine)(void*);
void* thread_arg;
/* Initialized thread_arg and start_routine and signal main thread that we don't need it
* to wait any longer.
*/
{
ZSTD_thread_params_t* thread_param = (ZSTD_thread_params_t*)arg;
thread_arg = thread_param->arg;
start_routine = thread_param->start_routine;
/* Signal main thread that we are running and do not depend on its memory anymore */
ZSTD_pthread_mutex_lock(&thread_param->initialized_mutex);
thread_param->initialized = 1;
ZSTD_pthread_cond_signal(&thread_param->initialized_cond);
ZSTD_pthread_mutex_unlock(&thread_param->initialized_mutex);
}
start_routine(thread_arg);
return 0;
}
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg)
{
ZSTD_thread_params_t thread_param;
(void)unused;
thread_param.start_routine = start_routine;
thread_param.arg = arg;
thread_param.initialized = 0;
*thread = NULL;
/* Setup thread initialization synchronization */
if(ZSTD_pthread_cond_init(&thread_param.initialized_cond, NULL)) {
/* Should never happen on Windows */
return -1;
}
if(ZSTD_pthread_mutex_init(&thread_param.initialized_mutex, NULL)) {
/* Should never happen on Windows */
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return -1;
}
/* Spawn thread */
*thread = (HANDLE)_beginthreadex(NULL, 0, worker, &thread_param, 0, NULL);
if (!thread) {
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return errno;
}
/* Wait for thread to be initialized */
ZSTD_pthread_mutex_lock(&thread_param.initialized_mutex);
while(!thread_param.initialized) {
ZSTD_pthread_cond_wait(&thread_param.initialized_cond, &thread_param.initialized_mutex);
}
ZSTD_pthread_mutex_unlock(&thread_param.initialized_mutex);
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return 0;
}
int ZSTD_pthread_join(ZSTD_pthread_t thread)
{
DWORD result;
if (!thread) return 0;
result = WaitForSingleObject(thread, INFINITE);
CloseHandle(thread);
switch (result) {
case WAIT_OBJECT_0:
return 0;
case WAIT_ABANDONED:
return EINVAL;
default:
return GetLastError();
}
}
#endif /* ZSTD_MULTITHREAD */
#if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32)
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h"
int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr)
{
*mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t));
if (!*mutex)
return 1;
return pthread_mutex_init(*mutex, attr);
}
int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex)
{
if (!*mutex)
return 0;
{
int const ret = pthread_mutex_destroy(*mutex);
ZSTD_free(*mutex);
return ret;
}
}
int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr)
{
*cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t));
if (!*cond)
return 1;
return pthread_cond_init(*cond, attr);
}
int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond)
{
if (!*cond)
return 0;
{
int const ret = pthread_cond_destroy(*cond);
ZSTD_free(*cond);
return ret;
}
}
#endif

150
ext/zstd/lib/common/threading.h Normal file
View File

@@ -0,0 +1,150 @@
/**
* Copyright (c) 2016 Tino Reichardt
* All rights reserved.
*
* You can contact the author at:
* - zstdmt source repository: https://github.com/mcmilk/zstdmt
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef THREADING_H_938743
#define THREADING_H_938743
#include "debug.h"
#if defined (__cplusplus)
extern "C" {
#endif
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper
*/
#ifdef WINVER
# undef WINVER
#endif
#define WINVER 0x0600
#ifdef _WIN32_WINNT
# undef _WIN32_WINNT
#endif
#define _WIN32_WINNT 0x0600
#ifndef WIN32_LEAN_AND_MEAN
# define WIN32_LEAN_AND_MEAN
#endif
#undef ERROR /* reported already defined on VS 2015 (Rich Geldreich) */
#include <windows.h>
#undef ERROR
#define ERROR(name) ZSTD_ERROR(name)
/* mutex */
#define ZSTD_pthread_mutex_t CRITICAL_SECTION
#define ZSTD_pthread_mutex_init(a, b) ((void)(b), InitializeCriticalSection((a)), 0)
#define ZSTD_pthread_mutex_destroy(a) DeleteCriticalSection((a))
#define ZSTD_pthread_mutex_lock(a) EnterCriticalSection((a))
#define ZSTD_pthread_mutex_unlock(a) LeaveCriticalSection((a))
/* condition variable */
#define ZSTD_pthread_cond_t CONDITION_VARIABLE
#define ZSTD_pthread_cond_init(a, b) ((void)(b), InitializeConditionVariable((a)), 0)
#define ZSTD_pthread_cond_destroy(a) ((void)(a))
#define ZSTD_pthread_cond_wait(a, b) SleepConditionVariableCS((a), (b), INFINITE)
#define ZSTD_pthread_cond_signal(a) WakeConditionVariable((a))
#define ZSTD_pthread_cond_broadcast(a) WakeAllConditionVariable((a))
/* ZSTD_pthread_create() and ZSTD_pthread_join() */
typedef HANDLE ZSTD_pthread_t;
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg);
int ZSTD_pthread_join(ZSTD_pthread_t thread);
/**
* add here more wrappers as required
*/
#elif defined(ZSTD_MULTITHREAD) /* posix assumed ; need a better detection method */
/* === POSIX Systems === */
# include <pthread.h>
#if DEBUGLEVEL < 1
#define ZSTD_pthread_mutex_t pthread_mutex_t
#define ZSTD_pthread_mutex_init(a, b) pthread_mutex_init((a), (b))
#define ZSTD_pthread_mutex_destroy(a) pthread_mutex_destroy((a))
#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock((a))
#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock((a))
#define ZSTD_pthread_cond_t pthread_cond_t
#define ZSTD_pthread_cond_init(a, b) pthread_cond_init((a), (b))
#define ZSTD_pthread_cond_destroy(a) pthread_cond_destroy((a))
#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait((a), (b))
#define ZSTD_pthread_cond_signal(a) pthread_cond_signal((a))
#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast((a))
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#else /* DEBUGLEVEL >= 1 */
/* Debug implementation of threading.
* In this implementation we use pointers for mutexes and condition variables.
* This way, if we forget to init/destroy them the program will crash or ASAN
* will report leaks.
*/
#define ZSTD_pthread_mutex_t pthread_mutex_t*
int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr);
int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex);
#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock(*(a))
#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock(*(a))
#define ZSTD_pthread_cond_t pthread_cond_t*
int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr);
int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait(*(a), *(b))
#define ZSTD_pthread_cond_signal(a) pthread_cond_signal(*(a))
#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast(*(a))
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#endif
#else /* ZSTD_MULTITHREAD not defined */
/* No multithreading support */
typedef int ZSTD_pthread_mutex_t;
#define ZSTD_pthread_mutex_init(a, b) ((void)(a), (void)(b), 0)
#define ZSTD_pthread_mutex_destroy(a) ((void)(a))
#define ZSTD_pthread_mutex_lock(a) ((void)(a))
#define ZSTD_pthread_mutex_unlock(a) ((void)(a))
typedef int ZSTD_pthread_cond_t;
#define ZSTD_pthread_cond_init(a, b) ((void)(a), (void)(b), 0)
#define ZSTD_pthread_cond_destroy(a) ((void)(a))
#define ZSTD_pthread_cond_wait(a, b) ((void)(a), (void)(b))
#define ZSTD_pthread_cond_signal(a) ((void)(a))
#define ZSTD_pthread_cond_broadcast(a) ((void)(a))
/* do not use ZSTD_pthread_t */
#endif /* ZSTD_MULTITHREAD */
#if defined (__cplusplus)
}
#endif
#endif /* THREADING_H_938743 */

24
ext/zstd/lib/common/xxhash.c vendored Normal file
View File

@@ -0,0 +1,24 @@
/*
* xxHash - Fast Hash algorithm
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - xxHash homepage: https://cyan4973.github.io/xxHash/
* - xxHash source repository : https://github.com/Cyan4973/xxHash
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*
* xxhash.c instantiates functions defined in xxhash.h
*/
#define XXH_STATIC_LINKING_ONLY /* access advanced declarations */
#define XXH_IMPLEMENTATION /* access definitions */
#include "xxhash.h"

5686
ext/zstd/lib/common/xxhash.h Normal file
View File

File diff suppressed because it is too large Load Diff

48
ext/zstd/lib/common/zstd_common.c vendored Normal file
View File

@@ -0,0 +1,48 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#define ZSTD_DEPS_NEED_MALLOC
#include "error_private.h"
#include "zstd_internal.h"
/*-****************************************
* Version
******************************************/
unsigned ZSTD_versionNumber(void) { return ZSTD_VERSION_NUMBER; }
const char* ZSTD_versionString(void) { return ZSTD_VERSION_STRING; }
/*-****************************************
* ZSTD Error Management
******************************************/
#undef ZSTD_isError /* defined within zstd_internal.h */
/*! ZSTD_isError() :
* tells if a return value is an error code
* symbol is required for external callers */
unsigned ZSTD_isError(size_t code) { return ERR_isError(code); }
/*! ZSTD_getErrorName() :
* provides error code string from function result (useful for debugging) */
const char* ZSTD_getErrorName(size_t code) { return ERR_getErrorName(code); }
/*! ZSTD_getError() :
* convert a `size_t` function result into a proper ZSTD_errorCode enum */
ZSTD_ErrorCode ZSTD_getErrorCode(size_t code) { return ERR_getErrorCode(code); }
/*! ZSTD_getErrorString() :
* provides error code string from enum */
const char* ZSTD_getErrorString(ZSTD_ErrorCode code) { return ERR_getErrorString(code); }

111
ext/zstd/lib/common/zstd_deps.h Normal file
View File

@@ -0,0 +1,111 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* This file provides common libc dependencies that zstd requires.
* The purpose is to allow replacing this file with a custom implementation
* to compile zstd without libc support.
*/
/* Need:
* NULL
* INT_MAX
* UINT_MAX
* ZSTD_memcpy()
* ZSTD_memset()
* ZSTD_memmove()
*/
#ifndef ZSTD_DEPS_COMMON
#define ZSTD_DEPS_COMMON
#include <limits.h>
#include <stddef.h>
#include <string.h>
#if defined(__GNUC__) && __GNUC__ >= 4
# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l))
# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l))
# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l))
#else
# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l))
# define ZSTD_memmove(d,s,l) memmove((d),(s),(l))
# define ZSTD_memset(p,v,l) memset((p),(v),(l))
#endif
#endif /* ZSTD_DEPS_COMMON */
/* Need:
* ZSTD_malloc()
* ZSTD_free()
* ZSTD_calloc()
*/
#ifdef ZSTD_DEPS_NEED_MALLOC
#ifndef ZSTD_DEPS_MALLOC
#define ZSTD_DEPS_MALLOC
#include <stdlib.h>
#define ZSTD_malloc(s) malloc(s)
#define ZSTD_calloc(n,s) calloc((n), (s))
#define ZSTD_free(p) free((p))
#endif /* ZSTD_DEPS_MALLOC */
#endif /* ZSTD_DEPS_NEED_MALLOC */
/*
* Provides 64-bit math support.
* Need:
* U64 ZSTD_div64(U64 dividend, U32 divisor)
*/
#ifdef ZSTD_DEPS_NEED_MATH64
#ifndef ZSTD_DEPS_MATH64
#define ZSTD_DEPS_MATH64
#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor))
#endif /* ZSTD_DEPS_MATH64 */
#endif /* ZSTD_DEPS_NEED_MATH64 */
/* Need:
* assert()
*/
#ifdef ZSTD_DEPS_NEED_ASSERT
#ifndef ZSTD_DEPS_ASSERT
#define ZSTD_DEPS_ASSERT
#include <assert.h>
#endif /* ZSTD_DEPS_ASSERT */
#endif /* ZSTD_DEPS_NEED_ASSERT */
/* Need:
* ZSTD_DEBUG_PRINT()
*/
#ifdef ZSTD_DEPS_NEED_IO
#ifndef ZSTD_DEPS_IO
#define ZSTD_DEPS_IO
#include <stdio.h>
#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__)
#endif /* ZSTD_DEPS_IO */
#endif /* ZSTD_DEPS_NEED_IO */
/* Only requested when <stdint.h> is known to be present.
* Need:
* intptr_t
*/
#ifdef ZSTD_DEPS_NEED_STDINT
#ifndef ZSTD_DEPS_STDINT
#define ZSTD_DEPS_STDINT
#include <stdint.h>
#endif /* ZSTD_DEPS_STDINT */
#endif /* ZSTD_DEPS_NEED_STDINT */

392
ext/zstd/lib/common/zstd_internal.h Normal file
View File

@@ -0,0 +1,392 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CCOMMON_H_MODULE
#define ZSTD_CCOMMON_H_MODULE
/* this module contains definitions which must be identical
* across compression, decompression and dictBuilder.
* It also contains a few functions useful to at least 2 of them
* and which benefit from being inlined */
/*-*************************************
* Dependencies
***************************************/
#include "compiler.h"
#include "cpu.h"
#include "mem.h"
#include "debug.h" /* assert, DEBUGLOG, RAWLOG, g_debuglevel */
#include "error_private.h"
#define ZSTD_STATIC_LINKING_ONLY
#include "../zstd.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#include "huf.h"
#ifndef XXH_STATIC_LINKING_ONLY
# define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */
#endif
#include "xxhash.h" /* XXH_reset, update, digest */
#ifndef ZSTD_NO_TRACE
# include "zstd_trace.h"
#else
# define ZSTD_TRACE 0
#endif
#if defined (__cplusplus)
extern "C" {
#endif
/* ---- static assert (debug) --- */
#define ZSTD_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)
#define ZSTD_isError ERR_isError /* for inlining */
#define FSE_isError ERR_isError
#define HUF_isError ERR_isError
/*-*************************************
* shared macros
***************************************/
#undef MIN
#undef MAX
#define MIN(a,b) ((a)<(b) ? (a) : (b))
#define MAX(a,b) ((a)>(b) ? (a) : (b))
#define BOUNDED(min,val,max) (MAX(min,MIN(val,max)))
/*-*************************************
* Common constants
***************************************/
#define ZSTD_OPT_NUM (1<<12)
#define ZSTD_REP_NUM 3 /* number of repcodes */
static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
#define KB *(1 <<10)
#define MB *(1 <<20)
#define GB *(1U<<30)
#define BIT7 128
#define BIT6 64
#define BIT5 32
#define BIT4 16
#define BIT1 2
#define BIT0 1
#define ZSTD_WINDOWLOG_ABSOLUTEMIN 10
static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 };
static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 };
#define ZSTD_FRAMEIDSIZE 4 /* magic number size */
#define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */
static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE;
typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e;
#define ZSTD_FRAMECHECKSUMSIZE 4
#define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */) /* for a non-null block */
#define MIN_LITERALS_FOR_4_STREAMS 6
typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e;
#define LONGNBSEQ 0x7F00
#define MINMATCH 3
#define Litbits 8
#define LitHufLog 11
#define MaxLit ((1<<Litbits) - 1)
#define MaxML 52
#define MaxLL 35
#define DefaultMaxOff 28
#define MaxOff 31
#define MaxSeq MAX(MaxLL, MaxML) /* Assumption : MaxOff < MaxLL,MaxML */
#define MLFSELog 9
#define LLFSELog 9
#define OffFSELog 8
#define MaxFSELog MAX(MAX(MLFSELog, LLFSELog), OffFSELog)
#define MaxMLBits 16
#define MaxLLBits 16
#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */
/* Each table cannot take more than #symbols * FSELog bits */
#define ZSTD_MAX_FSE_HEADERS_SIZE (((MaxML + 1) * MLFSELog + (MaxLL + 1) * LLFSELog + (MaxOff + 1) * OffFSELog + 7) / 8)
static UNUSED_ATTR const U8 LL_bits[MaxLL+1] = {
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3,
4, 6, 7, 8, 9,10,11,12,
13,14,15,16
};
static UNUSED_ATTR const S16 LL_defaultNorm[MaxLL+1] = {
4, 3, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 1, 1, 1,
2, 2, 2, 2, 2, 2, 2, 2,
2, 3, 2, 1, 1, 1, 1, 1,
-1,-1,-1,-1
};
#define LL_DEFAULTNORMLOG 6 /* for static allocation */
static UNUSED_ATTR const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG;
static UNUSED_ATTR const U8 ML_bits[MaxML+1] = {
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3,
4, 4, 5, 7, 8, 9,10,11,
12,13,14,15,16
};
static UNUSED_ATTR const S16 ML_defaultNorm[MaxML+1] = {
1, 4, 3, 2, 2, 2, 2, 2,
2, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1,-1,-1,
-1,-1,-1,-1,-1
};
#define ML_DEFAULTNORMLOG 6 /* for static allocation */
static UNUSED_ATTR const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG;
static UNUSED_ATTR const S16 OF_defaultNorm[DefaultMaxOff+1] = {
1, 1, 1, 1, 1, 1, 2, 2,
2, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
-1,-1,-1,-1,-1
};
#define OF_DEFAULTNORMLOG 5 /* for static allocation */
static UNUSED_ATTR const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG;
/*-*******************************************
* Shared functions to include for inlining
*********************************************/
static void ZSTD_copy8(void* dst, const void* src) {
#if defined(ZSTD_ARCH_ARM_NEON)
vst1_u8((uint8_t*)dst, vld1_u8((const uint8_t*)src));
#else
ZSTD_memcpy(dst, src, 8);
#endif
}
#define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; }
/* Need to use memmove here since the literal buffer can now be located within
the dst buffer. In circumstances where the op "catches up" to where the
literal buffer is, there can be partial overlaps in this call on the final
copy if the literal is being shifted by less than 16 bytes. */
static void ZSTD_copy16(void* dst, const void* src) {
#if defined(ZSTD_ARCH_ARM_NEON)
vst1q_u8((uint8_t*)dst, vld1q_u8((const uint8_t*)src));
#elif defined(ZSTD_ARCH_X86_SSE2)
_mm_storeu_si128((__m128i*)dst, _mm_loadu_si128((const __m128i*)src));
#elif defined(__clang__)
ZSTD_memmove(dst, src, 16);
#else
/* ZSTD_memmove is not inlined properly by gcc */
BYTE copy16_buf[16];
ZSTD_memcpy(copy16_buf, src, 16);
ZSTD_memcpy(dst, copy16_buf, 16);
#endif
}
#define COPY16(d,s) { ZSTD_copy16(d,s); d+=16; s+=16; }
#define WILDCOPY_OVERLENGTH 32
#define WILDCOPY_VECLEN 16
typedef enum {
ZSTD_no_overlap,
ZSTD_overlap_src_before_dst
/* ZSTD_overlap_dst_before_src, */
} ZSTD_overlap_e;
/*! ZSTD_wildcopy() :
* Custom version of ZSTD_memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0)
* @param ovtype controls the overlap detection
* - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart.
* - ZSTD_overlap_src_before_dst: The src and dst may overlap, but they MUST be at least 8 bytes apart.
* The src buffer must be before the dst buffer.
*/
MEM_STATIC FORCE_INLINE_ATTR
void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e const ovtype)
{
ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
BYTE* const oend = op + length;
if (ovtype == ZSTD_overlap_src_before_dst && diff < WILDCOPY_VECLEN) {
/* Handle short offset copies. */
do {
COPY8(op, ip)
} while (op < oend);
} else {
assert(diff >= WILDCOPY_VECLEN || diff <= -WILDCOPY_VECLEN);
/* Separate out the first COPY16() call because the copy length is
* almost certain to be short, so the branches have different
* probabilities. Since it is almost certain to be short, only do
* one COPY16() in the first call. Then, do two calls per loop since
* at that point it is more likely to have a high trip count.
*/
ZSTD_copy16(op, ip);
if (16 >= length) return;
op += 16;
ip += 16;
do {
COPY16(op, ip);
COPY16(op, ip);
}
while (op < oend);
}
}
MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
size_t const length = MIN(dstCapacity, srcSize);
if (length > 0) {
ZSTD_memcpy(dst, src, length);
}
return length;
}
/* define "workspace is too large" as this number of times larger than needed */
#define ZSTD_WORKSPACETOOLARGE_FACTOR 3
/* when workspace is continuously too large
* during at least this number of times,
* context's memory usage is considered wasteful,
* because it's sized to handle a worst case scenario which rarely happens.
* In which case, resize it down to free some memory */
#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128
/* Controls whether the input/output buffer is buffered or stable. */
typedef enum {
ZSTD_bm_buffered = 0, /* Buffer the input/output */
ZSTD_bm_stable = 1 /* ZSTD_inBuffer/ZSTD_outBuffer is stable */
} ZSTD_bufferMode_e;
/*-*******************************************
* Private declarations
*********************************************/
typedef struct seqDef_s {
U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
U16 litLength;
U16 mlBase; /* mlBase == matchLength - MINMATCH */
} seqDef;
/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */
typedef enum {
ZSTD_llt_none = 0, /* no longLengthType */
ZSTD_llt_literalLength = 1, /* represents a long literal */
ZSTD_llt_matchLength = 2 /* represents a long match */
} ZSTD_longLengthType_e;
typedef struct {
seqDef* sequencesStart;
seqDef* sequences; /* ptr to end of sequences */
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
size_t maxNbSeq;
size_t maxNbLit;
/* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength
* in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
* the existing value of the litLength or matchLength by 0x10000.
*/
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
} seqStore_t;
typedef struct {
U32 litLength;
U32 matchLength;
} ZSTD_sequenceLength;
/**
* Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences
* indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength.
*/
MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq)
{
ZSTD_sequenceLength seqLen;
seqLen.litLength = seq->litLength;
seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0x10000;
}
if (seqStore->longLengthType == ZSTD_llt_matchLength) {
seqLen.matchLength += 0x10000;
}
}
return seqLen;
}
/**
* Contains the compressed frame size and an upper-bound for the decompressed frame size.
* Note: before using `compressedSize`, check for errors using ZSTD_isError().
* similarly, before using `decompressedBound`, check for errors using:
* `decompressedBound != ZSTD_CONTENTSIZE_ERROR`
*/
typedef struct {
size_t nbBlocks;
size_t compressedSize;
unsigned long long decompressedBound;
} ZSTD_frameSizeInfo; /* decompress & legacy */
const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */
int ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
/* ZSTD_invalidateRepCodes() :
* ensures next compression will not use repcodes from previous block.
* Note : only works with regular variant;
* do not use with extDict variant ! */
void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx); /* zstdmt, adaptive_compression (shouldn't get this definition from here) */
typedef struct {
blockType_e blockType;
U32 lastBlock;
U32 origSize;
} blockProperties_t; /* declared here for decompress and fullbench */
/*! ZSTD_getcBlockSize() :
* Provides the size of compressed block from block header `src` */
/* Used by: decompress, fullbench (does not get its definition from here) */
size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
blockProperties_t* bpPtr);
/*! ZSTD_decodeSeqHeaders() :
* decode sequence header from src */
/* Used by: decompress, fullbench (does not get its definition from here) */
size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
const void* src, size_t srcSize);
/**
* @returns true iff the CPU supports dynamic BMI2 dispatch.
*/
MEM_STATIC int ZSTD_cpuSupportsBmi2(void)
{
ZSTD_cpuid_t cpuid = ZSTD_cpuid();
return ZSTD_cpuid_bmi1(cpuid) && ZSTD_cpuid_bmi2(cpuid);
}
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_CCOMMON_H_MODULE */

163
ext/zstd/lib/common/zstd_trace.h Normal file
View File

@@ -0,0 +1,163 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_TRACE_H
#define ZSTD_TRACE_H
#if defined (__cplusplus)
extern "C" {
#endif
#include <stddef.h>
/* weak symbol support
* For now, enable conservatively:
* - Only GNUC
* - Only ELF
* - Only x86-64, i386 and aarch64
* Also, explicitly disable on platforms known not to work so they aren't
* forgotten in the future.
*/
#if !defined(ZSTD_HAVE_WEAK_SYMBOLS) && \
defined(__GNUC__) && defined(__ELF__) && \
(defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86) || defined(__aarch64__)) && \
!defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \
!defined(__CYGWIN__) && !defined(_AIX)
# define ZSTD_HAVE_WEAK_SYMBOLS 1
#else
# define ZSTD_HAVE_WEAK_SYMBOLS 0
#endif
#if ZSTD_HAVE_WEAK_SYMBOLS
# define ZSTD_WEAK_ATTR __attribute__((__weak__))
#else
# define ZSTD_WEAK_ATTR
#endif
/* Only enable tracing when weak symbols are available. */
#ifndef ZSTD_TRACE
# define ZSTD_TRACE ZSTD_HAVE_WEAK_SYMBOLS
#endif
#if ZSTD_TRACE
struct ZSTD_CCtx_s;
struct ZSTD_DCtx_s;
struct ZSTD_CCtx_params_s;
typedef struct {
/**
* ZSTD_VERSION_NUMBER
*
* This is guaranteed to be the first member of ZSTD_trace.
* Otherwise, this struct is not stable between versions. If
* the version number does not match your expectation, you
* should not interpret the rest of the struct.
*/
unsigned version;
/**
* Non-zero if streaming (de)compression is used.
*/
unsigned streaming;
/**
* The dictionary ID.
*/
unsigned dictionaryID;
/**
* Is the dictionary cold?
* Only set on decompression.
*/
unsigned dictionaryIsCold;
/**
* The dictionary size or zero if no dictionary.
*/
size_t dictionarySize;
/**
* The uncompressed size of the data.
*/
size_t uncompressedSize;
/**
* The compressed size of the data.
*/
size_t compressedSize;
/**
* The fully resolved CCtx parameters (NULL on decompression).
*/
struct ZSTD_CCtx_params_s const* params;
/**
* The ZSTD_CCtx pointer (NULL on decompression).
*/
struct ZSTD_CCtx_s const* cctx;
/**
* The ZSTD_DCtx pointer (NULL on compression).
*/
struct ZSTD_DCtx_s const* dctx;
} ZSTD_Trace;
/**
* A tracing context. It must be 0 when tracing is disabled.
* Otherwise, any non-zero value returned by a tracing begin()
* function is presented to any subsequent calls to end().
*
* Any non-zero value is treated as tracing is enabled and not
* interpreted by the library.
*
* Two possible uses are:
* * A timestamp for when the begin() function was called.
* * A unique key identifying the (de)compression, like the
* address of the [dc]ctx pointer if you need to track
* more information than just a timestamp.
*/
typedef unsigned long long ZSTD_TraceCtx;
/**
* Trace the beginning of a compression call.
* @param cctx The dctx pointer for the compression.
* It can be used as a key to map begin() to end().
* @returns Non-zero if tracing is enabled. The return value is
* passed to ZSTD_trace_compress_end().
*/
ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_compress_begin(
struct ZSTD_CCtx_s const* cctx);
/**
* Trace the end of a compression call.
* @param ctx The return value of ZSTD_trace_compress_begin().
* @param trace The zstd tracing info.
*/
ZSTD_WEAK_ATTR void ZSTD_trace_compress_end(
ZSTD_TraceCtx ctx,
ZSTD_Trace const* trace);
/**
* Trace the beginning of a decompression call.
* @param dctx The dctx pointer for the decompression.
* It can be used as a key to map begin() to end().
* @returns Non-zero if tracing is enabled. The return value is
* passed to ZSTD_trace_compress_end().
*/
ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_decompress_begin(
struct ZSTD_DCtx_s const* dctx);
/**
* Trace the end of a decompression call.
* @param ctx The return value of ZSTD_trace_decompress_begin().
* @param trace The zstd tracing info.
*/
ZSTD_WEAK_ATTR void ZSTD_trace_decompress_end(
ZSTD_TraceCtx ctx,
ZSTD_Trace const* trace);
#endif /* ZSTD_TRACE */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_TRACE_H */

134
ext/zstd/lib/compress/clevels.h Normal file
View File

@@ -0,0 +1,134 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CLEVELS_H
#define ZSTD_CLEVELS_H
#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */
#include "../zstd.h"
/*-===== Pre-defined compression levels =====-*/
#define ZSTD_MAX_CLEVEL 22
#ifdef __GNUC__
__attribute__((__unused__))
#endif
static const ZSTD_compressionParameters ZSTD_defaultCParameters[4][ZSTD_MAX_CLEVEL+1] = {
{ /* "default" - for any srcSize > 256 KB */
/* W, C, H, S, L, TL, strat */
{ 19, 12, 13, 1, 6, 1, ZSTD_fast }, /* base for negative levels */
{ 19, 13, 14, 1, 7, 0, ZSTD_fast }, /* level 1 */
{ 20, 15, 16, 1, 6, 0, ZSTD_fast }, /* level 2 */
{ 21, 16, 17, 1, 5, 0, ZSTD_dfast }, /* level 3 */
{ 21, 18, 18, 1, 5, 0, ZSTD_dfast }, /* level 4 */
{ 21, 18, 19, 3, 5, 2, ZSTD_greedy }, /* level 5 */
{ 21, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6 */
{ 21, 19, 20, 4, 5, 8, ZSTD_lazy }, /* level 7 */
{ 21, 19, 20, 4, 5, 16, ZSTD_lazy2 }, /* level 8 */
{ 22, 20, 21, 4, 5, 16, ZSTD_lazy2 }, /* level 9 */
{ 22, 21, 22, 5, 5, 16, ZSTD_lazy2 }, /* level 10 */
{ 22, 21, 22, 6, 5, 16, ZSTD_lazy2 }, /* level 11 */
{ 22, 22, 23, 6, 5, 32, ZSTD_lazy2 }, /* level 12 */
{ 22, 22, 22, 4, 5, 32, ZSTD_btlazy2 }, /* level 13 */
{ 22, 22, 23, 5, 5, 32, ZSTD_btlazy2 }, /* level 14 */
{ 22, 23, 23, 6, 5, 32, ZSTD_btlazy2 }, /* level 15 */
{ 22, 22, 22, 5, 5, 48, ZSTD_btopt }, /* level 16 */
{ 23, 23, 22, 5, 4, 64, ZSTD_btopt }, /* level 17 */
{ 23, 23, 22, 6, 3, 64, ZSTD_btultra }, /* level 18 */
{ 23, 24, 22, 7, 3,256, ZSTD_btultra2}, /* level 19 */
{ 25, 25, 23, 7, 3,256, ZSTD_btultra2}, /* level 20 */
{ 26, 26, 24, 7, 3,512, ZSTD_btultra2}, /* level 21 */
{ 27, 27, 25, 9, 3,999, ZSTD_btultra2}, /* level 22 */
},
{ /* for srcSize <= 256 KB */
/* W, C, H, S, L, T, strat */
{ 18, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 18, 13, 14, 1, 6, 0, ZSTD_fast }, /* level 1 */
{ 18, 14, 14, 1, 5, 0, ZSTD_dfast }, /* level 2 */
{ 18, 16, 16, 1, 4, 0, ZSTD_dfast }, /* level 3 */
{ 18, 16, 17, 3, 5, 2, ZSTD_greedy }, /* level 4.*/
{ 18, 17, 18, 5, 5, 2, ZSTD_greedy }, /* level 5.*/
{ 18, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6.*/
{ 18, 18, 19, 4, 4, 4, ZSTD_lazy }, /* level 7 */
{ 18, 18, 19, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
{ 18, 18, 19, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
{ 18, 18, 19, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
{ 18, 18, 19, 5, 4, 12, ZSTD_btlazy2 }, /* level 11.*/
{ 18, 19, 19, 7, 4, 12, ZSTD_btlazy2 }, /* level 12.*/
{ 18, 18, 19, 4, 4, 16, ZSTD_btopt }, /* level 13 */
{ 18, 18, 19, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
{ 18, 18, 19, 6, 3,128, ZSTD_btopt }, /* level 15.*/
{ 18, 19, 19, 6, 3,128, ZSTD_btultra }, /* level 16.*/
{ 18, 19, 19, 8, 3,256, ZSTD_btultra }, /* level 17.*/
{ 18, 19, 19, 6, 3,128, ZSTD_btultra2}, /* level 18.*/
{ 18, 19, 19, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 18, 19, 19, 10, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 18, 19, 19, 12, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 18, 19, 19, 13, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 128 KB */
/* W, C, H, S, L, T, strat */
{ 17, 12, 12, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 17, 12, 13, 1, 6, 0, ZSTD_fast }, /* level 1 */
{ 17, 13, 15, 1, 5, 0, ZSTD_fast }, /* level 2 */
{ 17, 15, 16, 2, 5, 0, ZSTD_dfast }, /* level 3 */
{ 17, 17, 17, 2, 4, 0, ZSTD_dfast }, /* level 4 */
{ 17, 16, 17, 3, 4, 2, ZSTD_greedy }, /* level 5 */
{ 17, 16, 17, 3, 4, 4, ZSTD_lazy }, /* level 6 */
{ 17, 16, 17, 3, 4, 8, ZSTD_lazy2 }, /* level 7 */
{ 17, 16, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
{ 17, 16, 17, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
{ 17, 16, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
{ 17, 17, 17, 5, 4, 8, ZSTD_btlazy2 }, /* level 11 */
{ 17, 18, 17, 7, 4, 12, ZSTD_btlazy2 }, /* level 12 */
{ 17, 18, 17, 3, 4, 12, ZSTD_btopt }, /* level 13.*/
{ 17, 18, 17, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
{ 17, 18, 17, 6, 3,256, ZSTD_btopt }, /* level 15.*/
{ 17, 18, 17, 6, 3,128, ZSTD_btultra }, /* level 16.*/
{ 17, 18, 17, 8, 3,256, ZSTD_btultra }, /* level 17.*/
{ 17, 18, 17, 10, 3,512, ZSTD_btultra }, /* level 18.*/
{ 17, 18, 17, 5, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 17, 18, 17, 7, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 17, 18, 17, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 17, 18, 17, 11, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 16 KB */
/* W, C, H, S, L, T, strat */
{ 14, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 14, 14, 15, 1, 5, 0, ZSTD_fast }, /* level 1 */
{ 14, 14, 15, 1, 4, 0, ZSTD_fast }, /* level 2 */
{ 14, 14, 15, 2, 4, 0, ZSTD_dfast }, /* level 3 */
{ 14, 14, 14, 4, 4, 2, ZSTD_greedy }, /* level 4 */
{ 14, 14, 14, 3, 4, 4, ZSTD_lazy }, /* level 5.*/
{ 14, 14, 14, 4, 4, 8, ZSTD_lazy2 }, /* level 6 */
{ 14, 14, 14, 6, 4, 8, ZSTD_lazy2 }, /* level 7 */
{ 14, 14, 14, 8, 4, 8, ZSTD_lazy2 }, /* level 8.*/
{ 14, 15, 14, 5, 4, 8, ZSTD_btlazy2 }, /* level 9.*/
{ 14, 15, 14, 9, 4, 8, ZSTD_btlazy2 }, /* level 10.*/
{ 14, 15, 14, 3, 4, 12, ZSTD_btopt }, /* level 11.*/
{ 14, 15, 14, 4, 3, 24, ZSTD_btopt }, /* level 12.*/
{ 14, 15, 14, 5, 3, 32, ZSTD_btultra }, /* level 13.*/
{ 14, 15, 15, 6, 3, 64, ZSTD_btultra }, /* level 14.*/
{ 14, 15, 15, 7, 3,256, ZSTD_btultra }, /* level 15.*/
{ 14, 15, 15, 5, 3, 48, ZSTD_btultra2}, /* level 16.*/
{ 14, 15, 15, 6, 3,128, ZSTD_btultra2}, /* level 17.*/
{ 14, 15, 15, 7, 3,256, ZSTD_btultra2}, /* level 18.*/
{ 14, 15, 15, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 14, 15, 15, 8, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 14, 15, 15, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 14, 15, 15, 10, 3,999, ZSTD_btultra2}, /* level 22.*/
},
};
#endif /* ZSTD_CLEVELS_H */

624
ext/zstd/lib/compress/fse_compress.c vendored Normal file
View File

@@ -0,0 +1,624 @@
/* ******************************************************************
* FSE : Finite State Entropy encoder
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* **************************************************************
* Includes
****************************************************************/
#include "../common/compiler.h"
#include "../common/mem.h" /* U32, U16, etc. */
#include "../common/debug.h" /* assert, DEBUGLOG */
#include "hist.h" /* HIST_count_wksp */
#include "../common/bitstream.h"
#define FSE_STATIC_LINKING_ONLY
#include "../common/fse.h"
#include "../common/error_private.h"
#define ZSTD_DEPS_NEED_MALLOC
#define ZSTD_DEPS_NEED_MATH64
#include "../common/zstd_deps.h" /* ZSTD_malloc, ZSTD_free, ZSTD_memcpy, ZSTD_memset */
#include "../common/bits.h" /* ZSTD_highbit32 */
/* **************************************************************
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
/* **************************************************************
* Templates
****************************************************************/
/*
designed to be included
for type-specific functions (template emulation in C)
Objective is to write these functions only once, for improved maintenance
*/
/* safety checks */
#ifndef FSE_FUNCTION_EXTENSION
# error "FSE_FUNCTION_EXTENSION must be defined"
#endif
#ifndef FSE_FUNCTION_TYPE
# error "FSE_FUNCTION_TYPE must be defined"
#endif
/* Function names */
#define FSE_CAT(X,Y) X##Y
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
/* Function templates */
/* FSE_buildCTable_wksp() :
* Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
* wkspSize should be sized to handle worst case situation, which is `1<<max_tableLog * sizeof(FSE_FUNCTION_TYPE)`
* workSpace must also be properly aligned with FSE_FUNCTION_TYPE requirements
*/
size_t FSE_buildCTable_wksp(FSE_CTable* ct,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize)
{
U32 const tableSize = 1 << tableLog;
U32 const tableMask = tableSize - 1;
void* const ptr = ct;
U16* const tableU16 = ( (U16*) ptr) + 2;
void* const FSCT = ((U32*)ptr) + 1 /* header */ + (tableLog ? tableSize>>1 : 1) ;
FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT);
U32 const step = FSE_TABLESTEP(tableSize);
U32 const maxSV1 = maxSymbolValue+1;
U16* cumul = (U16*)workSpace; /* size = maxSV1 */
FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)(cumul + (maxSV1+1)); /* size = tableSize */
U32 highThreshold = tableSize-1;
assert(((size_t)workSpace & 1) == 0); /* Must be 2 bytes-aligned */
if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge);
/* CTable header */
tableU16[-2] = (U16) tableLog;
tableU16[-1] = (U16) maxSymbolValue;
assert(tableLog < 16); /* required for threshold strategy to work */
/* For explanations on how to distribute symbol values over the table :
* https://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
#ifdef __clang_analyzer__
ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */
#endif
/* symbol start positions */
{ U32 u;
cumul[0] = 0;
for (u=1; u <= maxSV1; u++) {
if (normalizedCounter[u-1]==-1) { /* Low proba symbol */
cumul[u] = cumul[u-1] + 1;
tableSymbol[highThreshold--] = (FSE_FUNCTION_TYPE)(u-1);
} else {
assert(normalizedCounter[u-1] >= 0);
cumul[u] = cumul[u-1] + (U16)normalizedCounter[u-1];
assert(cumul[u] >= cumul[u-1]); /* no overflow */
} }
cumul[maxSV1] = (U16)(tableSize+1);
}
/* Spread symbols */
if (highThreshold == tableSize - 1) {
/* Case for no low prob count symbols. Lay down 8 bytes at a time
* to reduce branch misses since we are operating on a small block
*/
BYTE* const spread = tableSymbol + tableSize; /* size = tableSize + 8 (may write beyond tableSize) */
{ U64 const add = 0x0101010101010101ull;
size_t pos = 0;
U64 sv = 0;
U32 s;
for (s=0; s<maxSV1; ++s, sv += add) {
int i;
int const n = normalizedCounter[s];
MEM_write64(spread + pos, sv);
for (i = 8; i < n; i += 8) {
MEM_write64(spread + pos + i, sv);
}
assert(n>=0);
pos += (size_t)n;
}
}
/* Spread symbols across the table. Lack of lowprob symbols means that
* we don't need variable sized inner loop, so we can unroll the loop and
* reduce branch misses.
*/
{ size_t position = 0;
size_t s;
size_t const unroll = 2; /* Experimentally determined optimal unroll */
assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
for (s = 0; s < (size_t)tableSize; s += unroll) {
size_t u;
for (u = 0; u < unroll; ++u) {
size_t const uPosition = (position + (u * step)) & tableMask;
tableSymbol[uPosition] = spread[s + u];
}
position = (position + (unroll * step)) & tableMask;
}
assert(position == 0); /* Must have initialized all positions */
}
} else {
U32 position = 0;
U32 symbol;
for (symbol=0; symbol<maxSV1; symbol++) {
int nbOccurrences;
int const freq = normalizedCounter[symbol];
for (nbOccurrences=0; nbOccurrences<freq; nbOccurrences++) {
tableSymbol[position] = (FSE_FUNCTION_TYPE)symbol;
position = (position + step) & tableMask;
while (position > highThreshold)
position = (position + step) & tableMask; /* Low proba area */
} }
assert(position==0); /* Must have initialized all positions */
}
/* Build table */
{ U32 u; for (u=0; u<tableSize; u++) {
FSE_FUNCTION_TYPE s = tableSymbol[u]; /* note : static analyzer may not understand tableSymbol is properly initialized */
tableU16[cumul[s]++] = (U16) (tableSize+u); /* TableU16 : sorted by symbol order; gives next state value */
} }
/* Build Symbol Transformation Table */
{ unsigned total = 0;
unsigned s;
for (s=0; s<=maxSymbolValue; s++) {
switch (normalizedCounter[s])
{
case 0:
/* filling nonetheless, for compatibility with FSE_getMaxNbBits() */
symbolTT[s].deltaNbBits = ((tableLog+1) << 16) - (1<<tableLog);
break;
case -1:
case 1:
symbolTT[s].deltaNbBits = (tableLog << 16) - (1<<tableLog);
assert(total <= INT_MAX);
symbolTT[s].deltaFindState = (int)(total - 1);
total ++;
break;
default :
assert(normalizedCounter[s] > 1);
{ U32 const maxBitsOut = tableLog - ZSTD_highbit32 ((U32)normalizedCounter[s]-1);
U32 const minStatePlus = (U32)normalizedCounter[s] << maxBitsOut;
symbolTT[s].deltaNbBits = (maxBitsOut << 16) - minStatePlus;
symbolTT[s].deltaFindState = (int)(total - (unsigned)normalizedCounter[s]);
total += (unsigned)normalizedCounter[s];
} } } }
#if 0 /* debug : symbol costs */
DEBUGLOG(5, "\n --- table statistics : ");
{ U32 symbol;
for (symbol=0; symbol<=maxSymbolValue; symbol++) {
DEBUGLOG(5, "%3u: w=%3i, maxBits=%u, fracBits=%.2f",
symbol, normalizedCounter[symbol],
FSE_getMaxNbBits(symbolTT, symbol),
(double)FSE_bitCost(symbolTT, tableLog, symbol, 8) / 256);
} }
#endif
return 0;
}
#ifndef FSE_COMMONDEFS_ONLY
/*-**************************************************************
* FSE NCount encoding
****************************************************************/
size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog)
{
size_t const maxHeaderSize = (((maxSymbolValue+1) * tableLog
+ 4 /* bitCount initialized at 4 */
+ 2 /* first two symbols may use one additional bit each */) / 8)
+ 1 /* round up to whole nb bytes */
+ 2 /* additional two bytes for bitstream flush */;
return maxSymbolValue ? maxHeaderSize : FSE_NCOUNTBOUND; /* maxSymbolValue==0 ? use default */
}
static size_t
FSE_writeNCount_generic (void* header, size_t headerBufferSize,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
unsigned writeIsSafe)
{
BYTE* const ostart = (BYTE*) header;
BYTE* out = ostart;
BYTE* const oend = ostart + headerBufferSize;
int nbBits;
const int tableSize = 1 << tableLog;
int remaining;
int threshold;
U32 bitStream = 0;
int bitCount = 0;
unsigned symbol = 0;
unsigned const alphabetSize = maxSymbolValue + 1;
int previousIs0 = 0;
/* Table Size */
bitStream += (tableLog-FSE_MIN_TABLELOG) << bitCount;
bitCount += 4;
/* Init */
remaining = tableSize+1; /* +1 for extra accuracy */
threshold = tableSize;
nbBits = tableLog+1;
while ((symbol < alphabetSize) && (remaining>1)) { /* stops at 1 */
if (previousIs0) {
unsigned start = symbol;
while ((symbol < alphabetSize) && !normalizedCounter[symbol]) symbol++;
if (symbol == alphabetSize) break; /* incorrect distribution */
while (symbol >= start+24) {
start+=24;
bitStream += 0xFFFFU << bitCount;
if ((!writeIsSafe) && (out > oend-2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE) bitStream;
out[1] = (BYTE)(bitStream>>8);
out+=2;
bitStream>>=16;
}
while (symbol >= start+3) {
start+=3;
bitStream += 3 << bitCount;
bitCount += 2;
}
bitStream += (symbol-start) << bitCount;
bitCount += 2;
if (bitCount>16) {
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
bitStream >>= 16;
bitCount -= 16;
} }
{ int count = normalizedCounter[symbol++];
int const max = (2*threshold-1) - remaining;
remaining -= count < 0 ? -count : count;
count++; /* +1 for extra accuracy */
if (count>=threshold)
count += max; /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */
bitStream += count << bitCount;
bitCount += nbBits;
bitCount -= (count<max);
previousIs0 = (count==1);
if (remaining<1) return ERROR(GENERIC);
while (remaining<threshold) { nbBits--; threshold>>=1; }
}
if (bitCount>16) {
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
bitStream >>= 16;
bitCount -= 16;
} }
if (remaining != 1)
return ERROR(GENERIC); /* incorrect normalized distribution */
assert(symbol <= alphabetSize);
/* flush remaining bitStream */
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out+= (bitCount+7) /8;
return (out-ostart);
}
size_t FSE_writeNCount (void* buffer, size_t bufferSize,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
{
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported */
if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported */
if (bufferSize < FSE_NCountWriteBound(maxSymbolValue, tableLog))
return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 0);
return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 1 /* write in buffer is safe */);
}
/*-**************************************************************
* FSE Compression Code
****************************************************************/
/* provides the minimum logSize to safely represent a distribution */
static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue)
{
U32 minBitsSrc = ZSTD_highbit32((U32)(srcSize)) + 1;
U32 minBitsSymbols = ZSTD_highbit32(maxSymbolValue) + 2;
U32 minBits = minBitsSrc < minBitsSymbols ? minBitsSrc : minBitsSymbols;
assert(srcSize > 1); /* Not supported, RLE should be used instead */
return minBits;
}
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus)
{
U32 maxBitsSrc = ZSTD_highbit32((U32)(srcSize - 1)) - minus;
U32 tableLog = maxTableLog;
U32 minBits = FSE_minTableLog(srcSize, maxSymbolValue);
assert(srcSize > 1); /* Not supported, RLE should be used instead */
if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG;
if (maxBitsSrc < tableLog) tableLog = maxBitsSrc; /* Accuracy can be reduced */
if (minBits > tableLog) tableLog = minBits; /* Need a minimum to safely represent all symbol values */
if (tableLog < FSE_MIN_TABLELOG) tableLog = FSE_MIN_TABLELOG;
if (tableLog > FSE_MAX_TABLELOG) tableLog = FSE_MAX_TABLELOG;
return tableLog;
}
unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue)
{
return FSE_optimalTableLog_internal(maxTableLog, srcSize, maxSymbolValue, 2);
}
/* Secondary normalization method.
To be used when primary method fails. */
static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount)
{
short const NOT_YET_ASSIGNED = -2;
U32 s;
U32 distributed = 0;
U32 ToDistribute;
/* Init */
U32 const lowThreshold = (U32)(total >> tableLog);
U32 lowOne = (U32)((total * 3) >> (tableLog + 1));
for (s=0; s<=maxSymbolValue; s++) {
if (count[s] == 0) {
norm[s]=0;
continue;
}
if (count[s] <= lowThreshold) {
norm[s] = lowProbCount;
distributed++;
total -= count[s];
continue;
}
if (count[s] <= lowOne) {
norm[s] = 1;
distributed++;
total -= count[s];
continue;
}
norm[s]=NOT_YET_ASSIGNED;
}
ToDistribute = (1 << tableLog) - distributed;
if (ToDistribute == 0)
return 0;
if ((total / ToDistribute) > lowOne) {
/* risk of rounding to zero */
lowOne = (U32)((total * 3) / (ToDistribute * 2));
for (s=0; s<=maxSymbolValue; s++) {
if ((norm[s] == NOT_YET_ASSIGNED) && (count[s] <= lowOne)) {
norm[s] = 1;
distributed++;
total -= count[s];
continue;
} }
ToDistribute = (1 << tableLog) - distributed;
}
if (distributed == maxSymbolValue+1) {
/* all values are pretty poor;
probably incompressible data (should have already been detected);
find max, then give all remaining points to max */
U32 maxV = 0, maxC = 0;
for (s=0; s<=maxSymbolValue; s++)
if (count[s] > maxC) { maxV=s; maxC=count[s]; }
norm[maxV] += (short)ToDistribute;
return 0;
}
if (total == 0) {
/* all of the symbols were low enough for the lowOne or lowThreshold */
for (s=0; ToDistribute > 0; s = (s+1)%(maxSymbolValue+1))
if (norm[s] > 0) { ToDistribute--; norm[s]++; }
return 0;
}
{ U64 const vStepLog = 62 - tableLog;
U64 const mid = (1ULL << (vStepLog-1)) - 1;
U64 const rStep = ZSTD_div64((((U64)1<<vStepLog) * ToDistribute) + mid, (U32)total); /* scale on remaining */
U64 tmpTotal = mid;
for (s=0; s<=maxSymbolValue; s++) {
if (norm[s]==NOT_YET_ASSIGNED) {
U64 const end = tmpTotal + (count[s] * rStep);
U32 const sStart = (U32)(tmpTotal >> vStepLog);
U32 const sEnd = (U32)(end >> vStepLog);
U32 const weight = sEnd - sStart;
if (weight < 1)
return ERROR(GENERIC);
norm[s] = (short)weight;
tmpTotal = end;
} } }
return 0;
}
size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
const unsigned* count, size_t total,
unsigned maxSymbolValue, unsigned useLowProbCount)
{
/* Sanity checks */
if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG;
if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported size */
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported size */
if (tableLog < FSE_minTableLog(total, maxSymbolValue)) return ERROR(GENERIC); /* Too small tableLog, compression potentially impossible */
{ static U32 const rtbTable[] = { 0, 473195, 504333, 520860, 550000, 700000, 750000, 830000 };
short const lowProbCount = useLowProbCount ? -1 : 1;
U64 const scale = 62 - tableLog;
U64 const step = ZSTD_div64((U64)1<<62, (U32)total); /* <== here, one division ! */
U64 const vStep = 1ULL<<(scale-20);
int stillToDistribute = 1<<tableLog;
unsigned s;
unsigned largest=0;
short largestP=0;
U32 lowThreshold = (U32)(total >> tableLog);
for (s=0; s<=maxSymbolValue; s++) {
if (count[s] == total) return 0; /* rle special case */
if (count[s] == 0) { normalizedCounter[s]=0; continue; }
if (count[s] <= lowThreshold) {
normalizedCounter[s] = lowProbCount;
stillToDistribute--;
} else {
short proba = (short)((count[s]*step) >> scale);
if (proba<8) {
U64 restToBeat = vStep * rtbTable[proba];
proba += (count[s]*step) - ((U64)proba<<scale) > restToBeat;
}
if (proba > largestP) { largestP=proba; largest=s; }
normalizedCounter[s] = proba;
stillToDistribute -= proba;
} }
if (-stillToDistribute >= (normalizedCounter[largest] >> 1)) {
/* corner case, need another normalization method */
size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount);
if (FSE_isError(errorCode)) return errorCode;
}
else normalizedCounter[largest] += (short)stillToDistribute;
}
#if 0
{ /* Print Table (debug) */
U32 s;
U32 nTotal = 0;
for (s=0; s<=maxSymbolValue; s++)
RAWLOG(2, "%3i: %4i \n", s, normalizedCounter[s]);
for (s=0; s<=maxSymbolValue; s++)
nTotal += abs(normalizedCounter[s]);
if (nTotal != (1U<<tableLog))
RAWLOG(2, "Warning !!! Total == %u != %u !!!", nTotal, 1U<<tableLog);
getchar();
}
#endif
return tableLog;
}
/* fake FSE_CTable, for rle input (always same symbol) */
size_t FSE_buildCTable_rle (FSE_CTable* ct, BYTE symbolValue)
{
void* ptr = ct;
U16* tableU16 = ( (U16*) ptr) + 2;
void* FSCTptr = (U32*)ptr + 2;
FSE_symbolCompressionTransform* symbolTT = (FSE_symbolCompressionTransform*) FSCTptr;
/* header */
tableU16[-2] = (U16) 0;
tableU16[-1] = (U16) symbolValue;
/* Build table */
tableU16[0] = 0;
tableU16[1] = 0; /* just in case */
/* Build Symbol Transformation Table */
symbolTT[symbolValue].deltaNbBits = 0;
symbolTT[symbolValue].deltaFindState = 0;
return 0;
}
static size_t FSE_compress_usingCTable_generic (void* dst, size_t dstSize,
const void* src, size_t srcSize,
const FSE_CTable* ct, const unsigned fast)
{
const BYTE* const istart = (const BYTE*) src;
const BYTE* const iend = istart + srcSize;
const BYTE* ip=iend;
BIT_CStream_t bitC;
FSE_CState_t CState1, CState2;
/* init */
if (srcSize <= 2) return 0;
{ size_t const initError = BIT_initCStream(&bitC, dst, dstSize);
if (FSE_isError(initError)) return 0; /* not enough space available to write a bitstream */ }
#define FSE_FLUSHBITS(s) (fast ? BIT_flushBitsFast(s) : BIT_flushBits(s))
if (srcSize & 1) {
FSE_initCState2(&CState1, ct, *--ip);
FSE_initCState2(&CState2, ct, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
FSE_FLUSHBITS(&bitC);
} else {
FSE_initCState2(&CState2, ct, *--ip);
FSE_initCState2(&CState1, ct, *--ip);
}
/* join to mod 4 */
srcSize -= 2;
if ((sizeof(bitC.bitContainer)*8 > FSE_MAX_TABLELOG*4+7 ) && (srcSize & 2)) { /* test bit 2 */
FSE_encodeSymbol(&bitC, &CState2, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
FSE_FLUSHBITS(&bitC);
}
/* 2 or 4 encoding per loop */
while ( ip>istart ) {
FSE_encodeSymbol(&bitC, &CState2, *--ip);
if (sizeof(bitC.bitContainer)*8 < FSE_MAX_TABLELOG*2+7 ) /* this test must be static */
FSE_FLUSHBITS(&bitC);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
if (sizeof(bitC.bitContainer)*8 > FSE_MAX_TABLELOG*4+7 ) { /* this test must be static */
FSE_encodeSymbol(&bitC, &CState2, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
}
FSE_FLUSHBITS(&bitC);
}
FSE_flushCState(&bitC, &CState2);
FSE_flushCState(&bitC, &CState1);
return BIT_closeCStream(&bitC);
}
size_t FSE_compress_usingCTable (void* dst, size_t dstSize,
const void* src, size_t srcSize,
const FSE_CTable* ct)
{
unsigned const fast = (dstSize >= FSE_BLOCKBOUND(srcSize));
if (fast)
return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 1);
else
return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 0);
}
size_t FSE_compressBound(size_t size) { return FSE_COMPRESSBOUND(size); }
#endif /* FSE_COMMONDEFS_ONLY */

181
ext/zstd/lib/compress/hist.c vendored Normal file
View File

@@ -0,0 +1,181 @@
/* ******************************************************************
* hist : Histogram functions
* part of Finite State Entropy project
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* --- dependencies --- */
#include "../common/mem.h" /* U32, BYTE, etc. */
#include "../common/debug.h" /* assert, DEBUGLOG */
#include "../common/error_private.h" /* ERROR */
#include "hist.h"
/* --- Error management --- */
unsigned HIST_isError(size_t code) { return ERR_isError(code); }
/*-**************************************************************
* Histogram functions
****************************************************************/
unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize)
{
const BYTE* ip = (const BYTE*)src;
const BYTE* const end = ip + srcSize;
unsigned maxSymbolValue = *maxSymbolValuePtr;
unsigned largestCount=0;
ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
while (ip<end) {
assert(*ip <= maxSymbolValue);
count[*ip++]++;
}
while (!count[maxSymbolValue]) maxSymbolValue--;
*maxSymbolValuePtr = maxSymbolValue;
{ U32 s;
for (s=0; s<=maxSymbolValue; s++)
if (count[s] > largestCount) largestCount = count[s];
}
return largestCount;
}
typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e;
/* HIST_count_parallel_wksp() :
* store histogram into 4 intermediate tables, recombined at the end.
* this design makes better use of OoO cpus,
* and is noticeably faster when some values are heavily repeated.
* But it needs some additional workspace for intermediate tables.
* `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32.
* @return : largest histogram frequency,
* or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */
static size_t HIST_count_parallel_wksp(
unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
HIST_checkInput_e check,
U32* const workSpace)
{
const BYTE* ip = (const BYTE*)source;
const BYTE* const iend = ip+sourceSize;
size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count);
unsigned max=0;
U32* const Counting1 = workSpace;
U32* const Counting2 = Counting1 + 256;
U32* const Counting3 = Counting2 + 256;
U32* const Counting4 = Counting3 + 256;
/* safety checks */
assert(*maxSymbolValuePtr <= 255);
if (!sourceSize) {
ZSTD_memset(count, 0, countSize);
*maxSymbolValuePtr = 0;
return 0;
}
ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned));
/* by stripes of 16 bytes */
{ U32 cached = MEM_read32(ip); ip += 4;
while (ip < iend-15) {
U32 c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
}
ip-=4;
}
/* finish last symbols */
while (ip<iend) Counting1[*ip++]++;
{ U32 s;
for (s=0; s<256; s++) {
Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
if (Counting1[s] > max) max = Counting1[s];
} }
{ unsigned maxSymbolValue = 255;
while (!Counting1[maxSymbolValue]) maxSymbolValue--;
if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall);
*maxSymbolValuePtr = maxSymbolValue;
ZSTD_memmove(count, Counting1, countSize); /* in case count & Counting1 are overlapping */
}
return (size_t)max;
}
/* HIST_countFast_wksp() :
* Same as HIST_countFast(), but using an externally provided scratch buffer.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
void* workSpace, size_t workSpaceSize)
{
if (sourceSize < 1500) /* heuristic threshold */
return HIST_count_simple(count, maxSymbolValuePtr, source, sourceSize);
if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace);
}
/* HIST_count_wksp() :
* Same as HIST_count(), but using an externally provided scratch buffer.
* `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */
size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
void* workSpace, size_t workSpaceSize)
{
if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
if (*maxSymbolValuePtr < 255)
return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, checkMaxSymbolValue, (U32*)workSpace);
*maxSymbolValuePtr = 255;
return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize);
}
#ifndef ZSTD_NO_UNUSED_FUNCTIONS
/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize)
{
unsigned tmpCounters[HIST_WKSP_SIZE_U32];
return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters));
}
size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize)
{
unsigned tmpCounters[HIST_WKSP_SIZE_U32];
return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters));
}
#endif

75
ext/zstd/lib/compress/hist.h Normal file
View File

@@ -0,0 +1,75 @@
/* ******************************************************************
* hist : Histogram functions
* part of Finite State Entropy project
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* --- dependencies --- */
#include "../common/zstd_deps.h" /* size_t */
/* --- simple histogram functions --- */
/*! HIST_count():
* Provides the precise count of each byte within a table 'count'.
* 'count' is a table of unsigned int, of minimum size (*maxSymbolValuePtr+1).
* Updates *maxSymbolValuePtr with actual largest symbol value detected.
* @return : count of the most frequent symbol (which isn't identified).
* or an error code, which can be tested using HIST_isError().
* note : if return == srcSize, there is only one symbol.
*/
size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);
unsigned HIST_isError(size_t code); /**< tells if a return value is an error code */
/* --- advanced histogram functions --- */
#define HIST_WKSP_SIZE_U32 1024
#define HIST_WKSP_SIZE (HIST_WKSP_SIZE_U32 * sizeof(unsigned))
/** HIST_count_wksp() :
* Same as HIST_count(), but using an externally provided scratch buffer.
* Benefit is this function will use very little stack space.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize,
void* workSpace, size_t workSpaceSize);
/** HIST_countFast() :
* same as HIST_count(), but blindly trusts that all byte values within src are <= *maxSymbolValuePtr.
* This function is unsafe, and will segfault if any value within `src` is `> *maxSymbolValuePtr`
*/
size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);
/** HIST_countFast_wksp() :
* Same as HIST_countFast(), but using an externally provided scratch buffer.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize,
void* workSpace, size_t workSpaceSize);
/*! HIST_count_simple() :
* Same as HIST_countFast(), this function is unsafe,
* and will segfault if any value within `src` is `> *maxSymbolValuePtr`.
* It is also a bit slower for large inputs.
* However, it does not need any additional memory (not even on stack).
* @return : count of the most frequent symbol.
* Note this function doesn't produce any error (i.e. it must succeed).
*/
unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);

1435
ext/zstd/lib/compress/huf_compress.c vendored Normal file
View File

File diff suppressed because it is too large Load Diff

7032
ext/zstd/lib/compress/zstd_compress.c vendored Normal file
View File

File diff suppressed because it is too large Load Diff

1532
ext/zstd/lib/compress/zstd_compress_internal.h Normal file
View File

File diff suppressed because it is too large Load Diff

235
ext/zstd/lib/compress/zstd_compress_literals.c vendored Normal file
View File

@@ -0,0 +1,235 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_literals.h"
/* **************************************************************
* Debug Traces
****************************************************************/
#if DEBUGLEVEL >= 2
static size_t showHexa(const void* src, size_t srcSize)
{
const BYTE* const ip = (const BYTE*)src;
size_t u;
for (u=0; u<srcSize; u++) {
RAWLOG(5, " %02X", ip[u]); (void)ip;
}
RAWLOG(5, " \n");
return srcSize;
}
#endif
/* **************************************************************
* Literals compression - special cases
****************************************************************/
size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
BYTE* const ostart = (BYTE*)dst;
U32 const flSize = 1 + (srcSize>31) + (srcSize>4095);
DEBUGLOG(5, "ZSTD_noCompressLiterals: srcSize=%zu, dstCapacity=%zu", srcSize, dstCapacity);
RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall, "");
switch(flSize)
{
case 1: /* 2 - 1 - 5 */
ostart[0] = (BYTE)((U32)set_basic + (srcSize<<3));
break;
case 2: /* 2 - 2 - 12 */
MEM_writeLE16(ostart, (U16)((U32)set_basic + (1<<2) + (srcSize<<4)));
break;
case 3: /* 2 - 2 - 20 */
MEM_writeLE32(ostart, (U32)((U32)set_basic + (3<<2) + (srcSize<<4)));
break;
default: /* not necessary : flSize is {1,2,3} */
assert(0);
}
ZSTD_memcpy(ostart + flSize, src, srcSize);
DEBUGLOG(5, "Raw (uncompressed) literals: %u -> %u", (U32)srcSize, (U32)(srcSize + flSize));
return srcSize + flSize;
}
static int allBytesIdentical(const void* src, size_t srcSize)
{
assert(srcSize >= 1);
assert(src != NULL);
{ const BYTE b = ((const BYTE*)src)[0];
size_t p;
for (p=1; p<srcSize; p++) {
if (((const BYTE*)src)[p] != b) return 0;
}
return 1;
}
}
size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
BYTE* const ostart = (BYTE*)dst;
U32 const flSize = 1 + (srcSize>31) + (srcSize>4095);
assert(dstCapacity >= 4); (void)dstCapacity;
assert(allBytesIdentical(src, srcSize));
switch(flSize)
{
case 1: /* 2 - 1 - 5 */
ostart[0] = (BYTE)((U32)set_rle + (srcSize<<3));
break;
case 2: /* 2 - 2 - 12 */
MEM_writeLE16(ostart, (U16)((U32)set_rle + (1<<2) + (srcSize<<4)));
break;
case 3: /* 2 - 2 - 20 */
MEM_writeLE32(ostart, (U32)((U32)set_rle + (3<<2) + (srcSize<<4)));
break;
default: /* not necessary : flSize is {1,2,3} */
assert(0);
}
ostart[flSize] = *(const BYTE*)src;
DEBUGLOG(5, "RLE : Repeated Literal (%02X: %u times) -> %u bytes encoded", ((const BYTE*)src)[0], (U32)srcSize, (U32)flSize + 1);
return flSize+1;
}
/* ZSTD_minLiteralsToCompress() :
* returns minimal amount of literals
* for literal compression to even be attempted.
* Minimum is made tighter as compression strategy increases.
*/
static size_t
ZSTD_minLiteralsToCompress(ZSTD_strategy strategy, HUF_repeat huf_repeat)
{
assert((int)strategy >= 0);
assert((int)strategy <= 9);
/* btultra2 : min 8 bytes;
* then 2x larger for each successive compression strategy
* max threshold 64 bytes */
{ int const shift = MIN(9-(int)strategy, 3);
size_t const mintc = (huf_repeat == HUF_repeat_valid) ? 6 : (size_t)8 << shift;
DEBUGLOG(7, "minLiteralsToCompress = %zu", mintc);
return mintc;
}
}
size_t ZSTD_compressLiterals (
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
void* entropyWorkspace, size_t entropyWorkspaceSize,
const ZSTD_hufCTables_t* prevHuf,
ZSTD_hufCTables_t* nextHuf,
ZSTD_strategy strategy,
int disableLiteralCompression,
int suspectUncompressible,
int bmi2)
{
size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB);
BYTE* const ostart = (BYTE*)dst;
U32 singleStream = srcSize < 256;
symbolEncodingType_e hType = set_compressed;
size_t cLitSize;
DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i, srcSize=%u, dstCapacity=%zu)",
disableLiteralCompression, (U32)srcSize, dstCapacity);
DEBUGLOG(6, "Completed literals listing (%zu bytes)", showHexa(src, srcSize));
/* Prepare nextEntropy assuming reusing the existing table */
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
if (disableLiteralCompression)
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
/* if too small, don't even attempt compression (speed opt) */
if (srcSize < ZSTD_minLiteralsToCompress(strategy, prevHuf->repeatMode))
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
RETURN_ERROR_IF(dstCapacity < lhSize+1, dstSize_tooSmall, "not enough space for compression");
{ HUF_repeat repeat = prevHuf->repeatMode;
int const flags = 0
| (bmi2 ? HUF_flags_bmi2 : 0)
| (strategy < ZSTD_lazy && srcSize <= 1024 ? HUF_flags_preferRepeat : 0)
| (strategy >= HUF_OPTIMAL_DEPTH_THRESHOLD ? HUF_flags_optimalDepth : 0)
| (suspectUncompressible ? HUF_flags_suspectUncompressible : 0);
typedef size_t (*huf_compress_f)(void*, size_t, const void*, size_t, unsigned, unsigned, void*, size_t, HUF_CElt*, HUF_repeat*, int);
huf_compress_f huf_compress;
if (repeat == HUF_repeat_valid && lhSize == 3) singleStream = 1;
huf_compress = singleStream ? HUF_compress1X_repeat : HUF_compress4X_repeat;
cLitSize = huf_compress(ostart+lhSize, dstCapacity-lhSize,
src, srcSize,
HUF_SYMBOLVALUE_MAX, LitHufLog,
entropyWorkspace, entropyWorkspaceSize,
(HUF_CElt*)nextHuf->CTable,
&repeat, flags);
DEBUGLOG(5, "%zu literals compressed into %zu bytes (before header)", srcSize, cLitSize);
if (repeat != HUF_repeat_none) {
/* reused the existing table */
DEBUGLOG(5, "reusing statistics from previous huffman block");
hType = set_repeat;
}
}
{ size_t const minGain = ZSTD_minGain(srcSize, strategy);
if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) {
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
} }
if (cLitSize==1) {
/* A return value of 1 signals that the alphabet consists of a single symbol.
* However, in some rare circumstances, it could be the compressed size (a single byte).
* For that outcome to have a chance to happen, it's necessary that `srcSize < 8`.
* (it's also necessary to not generate statistics).
* Therefore, in such a case, actively check that all bytes are identical. */
if ((srcSize >= 8) || allBytesIdentical(src, srcSize)) {
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize);
} }
if (hType == set_compressed) {
/* using a newly constructed table */
nextHuf->repeatMode = HUF_repeat_check;
}
/* Build header */
switch(lhSize)
{
case 3: /* 2 - 2 - 10 - 10 */
if (!singleStream) assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + ((U32)(!singleStream) << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<14);
MEM_writeLE24(ostart, lhc);
break;
}
case 4: /* 2 - 2 - 14 - 14 */
assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + (2 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<18);
MEM_writeLE32(ostart, lhc);
break;
}
case 5: /* 2 - 2 - 18 - 18 */
assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + (3 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<22);
MEM_writeLE32(ostart, lhc);
ostart[4] = (BYTE)(cLitSize >> 10);
break;
}
default: /* not possible : lhSize is {3,4,5} */
assert(0);
}
DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)srcSize, (U32)(lhSize+cLitSize));
return lhSize+cLitSize;
}

39
ext/zstd/lib/compress/zstd_compress_literals.h Normal file
View File

@@ -0,0 +1,39 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_LITERALS_H
#define ZSTD_COMPRESS_LITERALS_H
#include "zstd_compress_internal.h" /* ZSTD_hufCTables_t, ZSTD_minGain() */
size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* ZSTD_compressRleLiteralsBlock() :
* Conditions :
* - All bytes in @src are identical
* - dstCapacity >= 4 */
size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* ZSTD_compressLiterals():
* @entropyWorkspace: must be aligned on 4-bytes boundaries
* @entropyWorkspaceSize : must be >= HUF_WORKSPACE_SIZE
* @suspectUncompressible: sampling checks, to potentially skip huffman coding
*/
size_t ZSTD_compressLiterals (void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
void* entropyWorkspace, size_t entropyWorkspaceSize,
const ZSTD_hufCTables_t* prevHuf,
ZSTD_hufCTables_t* nextHuf,
ZSTD_strategy strategy, int disableLiteralCompression,
int suspectUncompressible,
int bmi2);
#endif /* ZSTD_COMPRESS_LITERALS_H */

442
ext/zstd/lib/compress/zstd_compress_sequences.c vendored Normal file
View File

@@ -0,0 +1,442 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_sequences.h"
/**
* -log2(x / 256) lookup table for x in [0, 256).
* If x == 0: Return 0
* Else: Return floor(-log2(x / 256) * 256)
*/
static unsigned const kInverseProbabilityLog256[256] = {
0, 2048, 1792, 1642, 1536, 1453, 1386, 1329, 1280, 1236, 1197, 1162,
1130, 1100, 1073, 1047, 1024, 1001, 980, 960, 941, 923, 906, 889,
874, 859, 844, 830, 817, 804, 791, 779, 768, 756, 745, 734,
724, 714, 704, 694, 685, 676, 667, 658, 650, 642, 633, 626,
618, 610, 603, 595, 588, 581, 574, 567, 561, 554, 548, 542,
535, 529, 523, 517, 512, 506, 500, 495, 489, 484, 478, 473,
468, 463, 458, 453, 448, 443, 438, 434, 429, 424, 420, 415,
411, 407, 402, 398, 394, 390, 386, 382, 377, 373, 370, 366,
362, 358, 354, 350, 347, 343, 339, 336, 332, 329, 325, 322,
318, 315, 311, 308, 305, 302, 298, 295, 292, 289, 286, 282,
279, 276, 273, 270, 267, 264, 261, 258, 256, 253, 250, 247,
244, 241, 239, 236, 233, 230, 228, 225, 222, 220, 217, 215,
212, 209, 207, 204, 202, 199, 197, 194, 192, 190, 187, 185,
182, 180, 178, 175, 173, 171, 168, 166, 164, 162, 159, 157,
155, 153, 151, 149, 146, 144, 142, 140, 138, 136, 134, 132,
130, 128, 126, 123, 121, 119, 117, 115, 114, 112, 110, 108,
106, 104, 102, 100, 98, 96, 94, 93, 91, 89, 87, 85,
83, 82, 80, 78, 76, 74, 73, 71, 69, 67, 66, 64,
62, 61, 59, 57, 55, 54, 52, 50, 49, 47, 46, 44,
42, 41, 39, 37, 36, 34, 33, 31, 30, 28, 26, 25,
23, 22, 20, 19, 17, 16, 14, 13, 11, 10, 8, 7,
5, 4, 2, 1,
};
static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) {
void const* ptr = ctable;
U16 const* u16ptr = (U16 const*)ptr;
U32 const maxSymbolValue = MEM_read16(u16ptr + 1);
return maxSymbolValue;
}
/**
* Returns true if we should use ncount=-1 else we should
* use ncount=1 for low probability symbols instead.
*/
static unsigned ZSTD_useLowProbCount(size_t const nbSeq)
{
/* Heuristic: This should cover most blocks <= 16K and
* start to fade out after 16K to about 32K depending on
* compressibility.
*/
return nbSeq >= 2048;
}
/**
* Returns the cost in bytes of encoding the normalized count header.
* Returns an error if any of the helper functions return an error.
*/
static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max,
size_t const nbSeq, unsigned const FSELog)
{
BYTE wksp[FSE_NCOUNTBOUND];
S16 norm[MaxSeq + 1];
const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), "");
return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog);
}
/**
* Returns the cost in bits of encoding the distribution described by count
* using the entropy bound.
*/
static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t const total)
{
unsigned cost = 0;
unsigned s;
assert(total > 0);
for (s = 0; s <= max; ++s) {
unsigned norm = (unsigned)((256 * count[s]) / total);
if (count[s] != 0 && norm == 0)
norm = 1;
assert(count[s] < total);
cost += count[s] * kInverseProbabilityLog256[norm];
}
return cost >> 8;
}
/**
* Returns the cost in bits of encoding the distribution in count using ctable.
* Returns an error if ctable cannot represent all the symbols in count.
*/
size_t ZSTD_fseBitCost(
FSE_CTable const* ctable,
unsigned const* count,
unsigned const max)
{
unsigned const kAccuracyLog = 8;
size_t cost = 0;
unsigned s;
FSE_CState_t cstate;
FSE_initCState(&cstate, ctable);
if (ZSTD_getFSEMaxSymbolValue(ctable) < max) {
DEBUGLOG(5, "Repeat FSE_CTable has maxSymbolValue %u < %u",
ZSTD_getFSEMaxSymbolValue(ctable), max);
return ERROR(GENERIC);
}
for (s = 0; s <= max; ++s) {
unsigned const tableLog = cstate.stateLog;
unsigned const badCost = (tableLog + 1) << kAccuracyLog;
unsigned const bitCost = FSE_bitCost(cstate.symbolTT, tableLog, s, kAccuracyLog);
if (count[s] == 0)
continue;
if (bitCost >= badCost) {
DEBUGLOG(5, "Repeat FSE_CTable has Prob[%u] == 0", s);
return ERROR(GENERIC);
}
cost += (size_t)count[s] * bitCost;
}
return cost >> kAccuracyLog;
}
/**
* Returns the cost in bits of encoding the distribution in count using the
* table described by norm. The max symbol support by norm is assumed >= max.
* norm must be valid for every symbol with non-zero probability in count.
*/
size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
unsigned const* count, unsigned const max)
{
unsigned const shift = 8 - accuracyLog;
size_t cost = 0;
unsigned s;
assert(accuracyLog <= 8);
for (s = 0; s <= max; ++s) {
unsigned const normAcc = (norm[s] != -1) ? (unsigned)norm[s] : 1;
unsigned const norm256 = normAcc << shift;
assert(norm256 > 0);
assert(norm256 < 256);
cost += count[s] * kInverseProbabilityLog256[norm256];
}
return cost >> 8;
}
symbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy)
{
ZSTD_STATIC_ASSERT(ZSTD_defaultDisallowed == 0 && ZSTD_defaultAllowed != 0);
if (mostFrequent == nbSeq) {
*repeatMode = FSE_repeat_none;
if (isDefaultAllowed && nbSeq <= 2) {
/* Prefer set_basic over set_rle when there are 2 or fewer symbols,
* since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol.
* If basic encoding isn't possible, always choose RLE.
*/
DEBUGLOG(5, "Selected set_basic");
return set_basic;
}
DEBUGLOG(5, "Selected set_rle");
return set_rle;
}
if (strategy < ZSTD_lazy) {
if (isDefaultAllowed) {
size_t const staticFse_nbSeq_max = 1000;
size_t const mult = 10 - strategy;
size_t const baseLog = 3;
size_t const dynamicFse_nbSeq_min = (((size_t)1 << defaultNormLog) * mult) >> baseLog; /* 28-36 for offset, 56-72 for lengths */
assert(defaultNormLog >= 5 && defaultNormLog <= 6); /* xx_DEFAULTNORMLOG */
assert(mult <= 9 && mult >= 7);
if ( (*repeatMode == FSE_repeat_valid)
&& (nbSeq < staticFse_nbSeq_max) ) {
DEBUGLOG(5, "Selected set_repeat");
return set_repeat;
}
if ( (nbSeq < dynamicFse_nbSeq_min)
|| (mostFrequent < (nbSeq >> (defaultNormLog-1))) ) {
DEBUGLOG(5, "Selected set_basic");
/* The format allows default tables to be repeated, but it isn't useful.
* When using simple heuristics to select encoding type, we don't want
* to confuse these tables with dictionaries. When running more careful
* analysis, we don't need to waste time checking both repeating tables
* and default tables.
*/
*repeatMode = FSE_repeat_none;
return set_basic;
}
}
} else {
size_t const basicCost = isDefaultAllowed ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, count, max) : ERROR(GENERIC);
size_t const repeatCost = *repeatMode != FSE_repeat_none ? ZSTD_fseBitCost(prevCTable, count, max) : ERROR(GENERIC);
size_t const NCountCost = ZSTD_NCountCost(count, max, nbSeq, FSELog);
size_t const compressedCost = (NCountCost << 3) + ZSTD_entropyCost(count, max, nbSeq);
if (isDefaultAllowed) {
assert(!ZSTD_isError(basicCost));
assert(!(*repeatMode == FSE_repeat_valid && ZSTD_isError(repeatCost)));
}
assert(!ZSTD_isError(NCountCost));
assert(compressedCost < ERROR(maxCode));
DEBUGLOG(5, "Estimated bit costs: basic=%u\trepeat=%u\tcompressed=%u",
(unsigned)basicCost, (unsigned)repeatCost, (unsigned)compressedCost);
if (basicCost <= repeatCost && basicCost <= compressedCost) {
DEBUGLOG(5, "Selected set_basic");
assert(isDefaultAllowed);
*repeatMode = FSE_repeat_none;
return set_basic;
}
if (repeatCost <= compressedCost) {
DEBUGLOG(5, "Selected set_repeat");
assert(!ZSTD_isError(repeatCost));
return set_repeat;
}
assert(compressedCost < basicCost && compressedCost < repeatCost);
}
DEBUGLOG(5, "Selected set_compressed");
*repeatMode = FSE_repeat_check;
return set_compressed;
}
typedef struct {
S16 norm[MaxSeq + 1];
U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)];
} ZSTD_BuildCTableWksp;
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
const FSE_CTable* prevCTable, size_t prevCTableSize,
void* entropyWorkspace, size_t entropyWorkspaceSize)
{
BYTE* op = (BYTE*)dst;
const BYTE* const oend = op + dstCapacity;
DEBUGLOG(6, "ZSTD_buildCTable (dstCapacity=%u)", (unsigned)dstCapacity);
switch (type) {
case set_rle:
FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max), "");
RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall, "not enough space");
*op = codeTable[0];
return 1;
case set_repeat:
ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize);
return 0;
case set_basic:
FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize), ""); /* note : could be pre-calculated */
return 0;
case set_compressed: {
ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace;
size_t nbSeq_1 = nbSeq;
const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
if (count[codeTable[nbSeq-1]] > 1) {
count[codeTable[nbSeq-1]]--;
nbSeq_1--;
}
assert(nbSeq_1 > 1);
assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp));
(void)entropyWorkspaceSize;
FORWARD_IF_ERROR(FSE_normalizeCount(wksp->norm, tableLog, count, nbSeq_1, max, ZSTD_useLowProbCount(nbSeq_1)), "FSE_normalizeCount failed");
assert(oend >= op);
{ size_t const NCountSize = FSE_writeNCount(op, (size_t)(oend - op), wksp->norm, max, tableLog); /* overflow protected */
FORWARD_IF_ERROR(NCountSize, "FSE_writeNCount failed");
FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, wksp->norm, max, tableLog, wksp->wksp, sizeof(wksp->wksp)), "FSE_buildCTable_wksp failed");
return NCountSize;
}
}
default: assert(0); RETURN_ERROR(GENERIC, "impossible to reach");
}
}
FORCE_INLINE_TEMPLATE size_t
ZSTD_encodeSequences_body(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
BIT_CStream_t blockStream;
FSE_CState_t stateMatchLength;
FSE_CState_t stateOffsetBits;
FSE_CState_t stateLitLength;
RETURN_ERROR_IF(
ERR_isError(BIT_initCStream(&blockStream, dst, dstCapacity)),
dstSize_tooSmall, "not enough space remaining");
DEBUGLOG(6, "available space for bitstream : %i (dstCapacity=%u)",
(int)(blockStream.endPtr - blockStream.startPtr),
(unsigned)dstCapacity);
/* first symbols */
FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]);
FSE_initCState2(&stateOffsetBits, CTable_OffsetBits, ofCodeTable[nbSeq-1]);
FSE_initCState2(&stateLitLength, CTable_LitLength, llCodeTable[nbSeq-1]);
BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
if (longOffsets) {
U32 const ofBits = ofCodeTable[nbSeq-1];
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, extraBits);
BIT_flushBits(&blockStream);
}
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase >> extraBits,
ofBits - extraBits);
} else {
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]);
}
BIT_flushBits(&blockStream);
{ size_t n;
for (n=nbSeq-2 ; n<nbSeq ; n--) { /* intentional underflow */
BYTE const llCode = llCodeTable[n];
BYTE const ofCode = ofCodeTable[n];
BYTE const mlCode = mlCodeTable[n];
U32 const llBits = LL_bits[llCode];
U32 const ofBits = ofCode;
U32 const mlBits = ML_bits[mlCode];
DEBUGLOG(6, "encoding: litlen:%2u - matchlen:%2u - offCode:%7u",
(unsigned)sequences[n].litLength,
(unsigned)sequences[n].mlBase + MINMATCH,
(unsigned)sequences[n].offBase);
/* 32b*/ /* 64b*/
/* (7)*/ /* (7)*/
FSE_encodeSymbol(&blockStream, &stateOffsetBits, ofCode); /* 15 */ /* 15 */
FSE_encodeSymbol(&blockStream, &stateMatchLength, mlCode); /* 24 */ /* 24 */
if (MEM_32bits()) BIT_flushBits(&blockStream); /* (7)*/
FSE_encodeSymbol(&blockStream, &stateLitLength, llCode); /* 16 */ /* 33 */
if (MEM_32bits() || (ofBits+mlBits+llBits >= 64-7-(LLFSELog+MLFSELog+OffFSELog)))
BIT_flushBits(&blockStream); /* (7)*/
BIT_addBits(&blockStream, sequences[n].litLength, llBits);
if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream);
BIT_addBits(&blockStream, sequences[n].mlBase, mlBits);
if (MEM_32bits() || (ofBits+mlBits+llBits > 56)) BIT_flushBits(&blockStream);
if (longOffsets) {
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
BIT_addBits(&blockStream, sequences[n].offBase, extraBits);
BIT_flushBits(&blockStream); /* (7)*/
}
BIT_addBits(&blockStream, sequences[n].offBase >> extraBits,
ofBits - extraBits); /* 31 */
} else {
BIT_addBits(&blockStream, sequences[n].offBase, ofBits); /* 31 */
}
BIT_flushBits(&blockStream); /* (7)*/
DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr));
} }
DEBUGLOG(6, "ZSTD_encodeSequences: flushing ML state with %u bits", stateMatchLength.stateLog);
FSE_flushCState(&blockStream, &stateMatchLength);
DEBUGLOG(6, "ZSTD_encodeSequences: flushing Off state with %u bits", stateOffsetBits.stateLog);
FSE_flushCState(&blockStream, &stateOffsetBits);
DEBUGLOG(6, "ZSTD_encodeSequences: flushing LL state with %u bits", stateLitLength.stateLog);
FSE_flushCState(&blockStream, &stateLitLength);
{ size_t const streamSize = BIT_closeCStream(&blockStream);
RETURN_ERROR_IF(streamSize==0, dstSize_tooSmall, "not enough space");
return streamSize;
}
}
static size_t
ZSTD_encodeSequences_default(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#if DYNAMIC_BMI2
static BMI2_TARGET_ATTRIBUTE size_t
ZSTD_encodeSequences_bmi2(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#endif
size_t ZSTD_encodeSequences(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2)
{
DEBUGLOG(5, "ZSTD_encodeSequences: dstCapacity = %u", (unsigned)dstCapacity);
#if DYNAMIC_BMI2
if (bmi2) {
return ZSTD_encodeSequences_bmi2(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#endif
(void)bmi2;
return ZSTD_encodeSequences_default(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}

54
ext/zstd/lib/compress/zstd_compress_sequences.h Normal file
View File

@@ -0,0 +1,54 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_SEQUENCES_H
#define ZSTD_COMPRESS_SEQUENCES_H
#include "../common/fse.h" /* FSE_repeat, FSE_CTable */
#include "../common/zstd_internal.h" /* symbolEncodingType_e, ZSTD_strategy */
typedef enum {
ZSTD_defaultDisallowed = 0,
ZSTD_defaultAllowed = 1
} ZSTD_defaultPolicy_e;
symbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy);
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
const FSE_CTable* prevCTable, size_t prevCTableSize,
void* entropyWorkspace, size_t entropyWorkspaceSize);
size_t ZSTD_encodeSequences(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2);
size_t ZSTD_fseBitCost(
FSE_CTable const* ctable,
unsigned const* count,
unsigned const max);
size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
unsigned const* count, unsigned const max);
#endif /* ZSTD_COMPRESS_SEQUENCES_H */

577
ext/zstd/lib/compress/zstd_compress_superblock.c vendored Normal file
View File

@@ -0,0 +1,577 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_superblock.h"
#include "../common/zstd_internal.h" /* ZSTD_getSequenceLength */
#include "hist.h" /* HIST_countFast_wksp */
#include "zstd_compress_internal.h" /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */
#include "zstd_compress_sequences.h"
#include "zstd_compress_literals.h"
/** ZSTD_compressSubBlock_literal() :
* Compresses literals section for a sub-block.
* When we have to write the Huffman table we will sometimes choose a header
* size larger than necessary. This is because we have to pick the header size
* before we know the table size + compressed size, so we have a bound on the
* table size. If we guessed incorrectly, we fall back to uncompressed literals.
*
* We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded
* in writing the header, otherwise it is set to 0.
*
* hufMetadata->hType has literals block type info.
* If it is set_basic, all sub-blocks literals section will be Raw_Literals_Block.
* If it is set_rle, all sub-blocks literals section will be RLE_Literals_Block.
* If it is set_compressed, first sub-block's literals section will be Compressed_Literals_Block
* If it is set_compressed, first sub-block's literals section will be Treeless_Literals_Block
* and the following sub-blocks' literals sections will be Treeless_Literals_Block.
* @return : compressed size of literals section of a sub-block
* Or 0 if unable to compress.
* Or error code */
static size_t
ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable,
const ZSTD_hufCTablesMetadata_t* hufMetadata,
const BYTE* literals, size_t litSize,
void* dst, size_t dstSize,
const int bmi2, int writeEntropy, int* entropyWritten)
{
size_t const header = writeEntropy ? 200 : 0;
size_t const lhSize = 3 + (litSize >= (1 KB - header)) + (litSize >= (16 KB - header));
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstSize;
BYTE* op = ostart + lhSize;
U32 const singleStream = lhSize == 3;
symbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat;
size_t cLitSize = 0;
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (litSize=%zu, lhSize=%zu, writeEntropy=%d)", litSize, lhSize, writeEntropy);
*entropyWritten = 0;
if (litSize == 0 || hufMetadata->hType == set_basic) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
} else if (hufMetadata->hType == set_rle) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using rle literal");
return ZSTD_compressRleLiteralsBlock(dst, dstSize, literals, litSize);
}
assert(litSize > 0);
assert(hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat);
if (writeEntropy && hufMetadata->hType == set_compressed) {
ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize);
op += hufMetadata->hufDesSize;
cLitSize += hufMetadata->hufDesSize;
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (hSize=%zu)", hufMetadata->hufDesSize);
}
{ int const flags = bmi2 ? HUF_flags_bmi2 : 0;
const size_t cSize = singleStream ? HUF_compress1X_usingCTable(op, oend-op, literals, litSize, hufTable, flags)
: HUF_compress4X_usingCTable(op, oend-op, literals, litSize, hufTable, flags);
op += cSize;
cLitSize += cSize;
if (cSize == 0 || ERR_isError(cSize)) {
DEBUGLOG(5, "Failed to write entropy tables %s", ZSTD_getErrorName(cSize));
return 0;
}
/* If we expand and we aren't writing a header then emit uncompressed */
if (!writeEntropy && cLitSize >= litSize) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal because uncompressible");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
}
/* If we are writing headers then allow expansion that doesn't change our header size. */
if (lhSize < (size_t)(3 + (cLitSize >= 1 KB) + (cLitSize >= 16 KB))) {
assert(cLitSize > litSize);
DEBUGLOG(5, "Literals expanded beyond allowed header size");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
}
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (cSize=%zu)", cSize);
}
/* Build header */
switch(lhSize)
{
case 3: /* 2 - 2 - 10 - 10 */
{ U32 const lhc = hType + ((!singleStream) << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<14);
MEM_writeLE24(ostart, lhc);
break;
}
case 4: /* 2 - 2 - 14 - 14 */
{ U32 const lhc = hType + (2 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<18);
MEM_writeLE32(ostart, lhc);
break;
}
case 5: /* 2 - 2 - 18 - 18 */
{ U32 const lhc = hType + (3 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<22);
MEM_writeLE32(ostart, lhc);
ostart[4] = (BYTE)(cLitSize >> 10);
break;
}
default: /* not possible : lhSize is {3,4,5} */
assert(0);
}
*entropyWritten = 1;
DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)litSize, (U32)(op-ostart));
return op-ostart;
}
static size_t
ZSTD_seqDecompressedSize(seqStore_t const* seqStore,
const seqDef* sequences, size_t nbSeq,
size_t litSize, int lastSequence)
{
const seqDef* const sstart = sequences;
const seqDef* const send = sequences + nbSeq;
const seqDef* sp = sstart;
size_t matchLengthSum = 0;
size_t litLengthSum = 0;
(void)(litLengthSum); /* suppress unused variable warning on some environments */
while (send-sp > 0) {
ZSTD_sequenceLength const seqLen = ZSTD_getSequenceLength(seqStore, sp);
litLengthSum += seqLen.litLength;
matchLengthSum += seqLen.matchLength;
sp++;
}
assert(litLengthSum <= litSize);
if (!lastSequence) {
assert(litLengthSum == litSize);
}
return matchLengthSum + litSize;
}
/** ZSTD_compressSubBlock_sequences() :
* Compresses sequences section for a sub-block.
* fseMetadata->llType, fseMetadata->ofType, and fseMetadata->mlType have
* symbol compression modes for the super-block.
* The first successfully compressed block will have these in its header.
* We set entropyWritten=1 when we succeed in compressing the sequences.
* The following sub-blocks will always have repeat mode.
* @return : compressed size of sequences section of a sub-block
* Or 0 if it is unable to compress
* Or error code. */
static size_t
ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables,
const ZSTD_fseCTablesMetadata_t* fseMetadata,
const seqDef* sequences, size_t nbSeq,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const int bmi2, int writeEntropy, int* entropyWritten)
{
const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
BYTE* seqHead;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (nbSeq=%zu, writeEntropy=%d, longOffsets=%d)", nbSeq, writeEntropy, longOffsets);
*entropyWritten = 0;
/* Sequences Header */
RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/,
dstSize_tooSmall, "");
if (nbSeq < 0x7F)
*op++ = (BYTE)nbSeq;
else if (nbSeq < LONGNBSEQ)
op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
else
op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
if (nbSeq==0) {
return op - ostart;
}
/* seqHead : flags for FSE encoding type */
seqHead = op++;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (seqHeadSize=%u)", (unsigned)(op-ostart));
if (writeEntropy) {
const U32 LLtype = fseMetadata->llType;
const U32 Offtype = fseMetadata->ofType;
const U32 MLtype = fseMetadata->mlType;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (fseTablesSize=%zu)", fseMetadata->fseTablesSize);
*seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize);
op += fseMetadata->fseTablesSize;
} else {
const U32 repeat = set_repeat;
*seqHead = (BYTE)((repeat<<6) + (repeat<<4) + (repeat<<2));
}
{ size_t const bitstreamSize = ZSTD_encodeSequences(
op, oend - op,
fseTables->matchlengthCTable, mlCode,
fseTables->offcodeCTable, ofCode,
fseTables->litlengthCTable, llCode,
sequences, nbSeq,
longOffsets, bmi2);
FORWARD_IF_ERROR(bitstreamSize, "ZSTD_encodeSequences failed");
op += bitstreamSize;
/* zstd versions <= 1.3.4 mistakenly report corruption when
* FSE_readNCount() receives a buffer < 4 bytes.
* Fixed by https://github.com/facebook/zstd/pull/1146.
* This can happen when the last set_compressed table present is 2
* bytes and the bitstream is only one byte.
* In this exceedingly rare case, we will simply emit an uncompressed
* block, since it isn't worth optimizing.
*/
#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (writeEntropy && fseMetadata->lastCountSize && fseMetadata->lastCountSize + bitstreamSize < 4) {
/* NCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */
assert(fseMetadata->lastCountSize + bitstreamSize == 3);
DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by "
"emitting an uncompressed block.");
return 0;
}
#endif
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (bitstreamSize=%zu)", bitstreamSize);
}
/* zstd versions <= 1.4.0 mistakenly report error when
* sequences section body size is less than 3 bytes.
* Fixed by https://github.com/facebook/zstd/pull/1664.
* This can happen when the previous sequences section block is compressed
* with rle mode and the current block's sequences section is compressed
* with repeat mode where sequences section body size can be 1 byte.
*/
#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (op-seqHead < 4) {
DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.4.0 by emitting "
"an uncompressed block when sequences are < 4 bytes");
return 0;
}
#endif
*entropyWritten = 1;
return op - ostart;
}
/** ZSTD_compressSubBlock() :
* Compresses a single sub-block.
* @return : compressed size of the sub-block
* Or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock(const ZSTD_entropyCTables_t* entropy,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
const seqDef* sequences, size_t nbSeq,
const BYTE* literals, size_t litSize,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const int bmi2,
int writeLitEntropy, int writeSeqEntropy,
int* litEntropyWritten, int* seqEntropyWritten,
U32 lastBlock)
{
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart + ZSTD_blockHeaderSize;
DEBUGLOG(5, "ZSTD_compressSubBlock (litSize=%zu, nbSeq=%zu, writeLitEntropy=%d, writeSeqEntropy=%d, lastBlock=%d)",
litSize, nbSeq, writeLitEntropy, writeSeqEntropy, lastBlock);
{ size_t cLitSize = ZSTD_compressSubBlock_literal((const HUF_CElt*)entropy->huf.CTable,
&entropyMetadata->hufMetadata, literals, litSize,
op, oend-op, bmi2, writeLitEntropy, litEntropyWritten);
FORWARD_IF_ERROR(cLitSize, "ZSTD_compressSubBlock_literal failed");
if (cLitSize == 0) return 0;
op += cLitSize;
}
{ size_t cSeqSize = ZSTD_compressSubBlock_sequences(&entropy->fse,
&entropyMetadata->fseMetadata,
sequences, nbSeq,
llCode, mlCode, ofCode,
cctxParams,
op, oend-op,
bmi2, writeSeqEntropy, seqEntropyWritten);
FORWARD_IF_ERROR(cSeqSize, "ZSTD_compressSubBlock_sequences failed");
if (cSeqSize == 0) return 0;
op += cSeqSize;
}
/* Write block header */
{ size_t cSize = (op-ostart)-ZSTD_blockHeaderSize;
U32 const cBlockHeader24 = lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3);
MEM_writeLE24(ostart, cBlockHeader24);
}
return op-ostart;
}
static size_t ZSTD_estimateSubBlockSize_literal(const BYTE* literals, size_t litSize,
const ZSTD_hufCTables_t* huf,
const ZSTD_hufCTablesMetadata_t* hufMetadata,
void* workspace, size_t wkspSize,
int writeEntropy)
{
unsigned* const countWksp = (unsigned*)workspace;
unsigned maxSymbolValue = 255;
size_t literalSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
if (hufMetadata->hType == set_basic) return litSize;
else if (hufMetadata->hType == set_rle) return 1;
else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) {
size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize);
if (ZSTD_isError(largest)) return litSize;
{ size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue);
if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize;
return cLitSizeEstimate + literalSectionHeaderSize;
} }
assert(0); /* impossible */
return 0;
}
static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type,
const BYTE* codeTable, unsigned maxCode,
size_t nbSeq, const FSE_CTable* fseCTable,
const U8* additionalBits,
short const* defaultNorm, U32 defaultNormLog, U32 defaultMax,
void* workspace, size_t wkspSize)
{
unsigned* const countWksp = (unsigned*)workspace;
const BYTE* ctp = codeTable;
const BYTE* const ctStart = ctp;
const BYTE* const ctEnd = ctStart + nbSeq;
size_t cSymbolTypeSizeEstimateInBits = 0;
unsigned max = maxCode;
HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */
if (type == set_basic) {
/* We selected this encoding type, so it must be valid. */
assert(max <= defaultMax);
cSymbolTypeSizeEstimateInBits = max <= defaultMax
? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max)
: ERROR(GENERIC);
} else if (type == set_rle) {
cSymbolTypeSizeEstimateInBits = 0;
} else if (type == set_compressed || type == set_repeat) {
cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max);
}
if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) return nbSeq * 10;
while (ctp < ctEnd) {
if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp];
else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */
ctp++;
}
return cSymbolTypeSizeEstimateInBits / 8;
}
static size_t ZSTD_estimateSubBlockSize_sequences(const BYTE* ofCodeTable,
const BYTE* llCodeTable,
const BYTE* mlCodeTable,
size_t nbSeq,
const ZSTD_fseCTables_t* fseTables,
const ZSTD_fseCTablesMetadata_t* fseMetadata,
void* workspace, size_t wkspSize,
int writeEntropy)
{
size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
size_t cSeqSizeEstimate = 0;
if (nbSeq == 0) return sequencesSectionHeaderSize;
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, MaxOff,
nbSeq, fseTables->offcodeCTable, NULL,
OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff,
workspace, wkspSize);
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->llType, llCodeTable, MaxLL,
nbSeq, fseTables->litlengthCTable, LL_bits,
LL_defaultNorm, LL_defaultNormLog, MaxLL,
workspace, wkspSize);
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, MaxML,
nbSeq, fseTables->matchlengthCTable, ML_bits,
ML_defaultNorm, ML_defaultNormLog, MaxML,
workspace, wkspSize);
if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize;
return cSeqSizeEstimate + sequencesSectionHeaderSize;
}
static size_t ZSTD_estimateSubBlockSize(const BYTE* literals, size_t litSize,
const BYTE* ofCodeTable,
const BYTE* llCodeTable,
const BYTE* mlCodeTable,
size_t nbSeq,
const ZSTD_entropyCTables_t* entropy,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
void* workspace, size_t wkspSize,
int writeLitEntropy, int writeSeqEntropy) {
size_t cSizeEstimate = 0;
cSizeEstimate += ZSTD_estimateSubBlockSize_literal(literals, litSize,
&entropy->huf, &entropyMetadata->hufMetadata,
workspace, wkspSize, writeLitEntropy);
cSizeEstimate += ZSTD_estimateSubBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable,
nbSeq, &entropy->fse, &entropyMetadata->fseMetadata,
workspace, wkspSize, writeSeqEntropy);
return cSizeEstimate + ZSTD_blockHeaderSize;
}
static int ZSTD_needSequenceEntropyTables(ZSTD_fseCTablesMetadata_t const* fseMetadata)
{
if (fseMetadata->llType == set_compressed || fseMetadata->llType == set_rle)
return 1;
if (fseMetadata->mlType == set_compressed || fseMetadata->mlType == set_rle)
return 1;
if (fseMetadata->ofType == set_compressed || fseMetadata->ofType == set_rle)
return 1;
return 0;
}
/** ZSTD_compressSubBlock_multi() :
* Breaks super-block into multiple sub-blocks and compresses them.
* Entropy will be written to the first block.
* The following blocks will use repeat mode to compress.
* All sub-blocks are compressed blocks (no raw or rle blocks).
* @return : compressed size of the super block (which is multiple ZSTD blocks)
* Or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
const ZSTD_compressedBlockState_t* prevCBlock,
ZSTD_compressedBlockState_t* nextCBlock,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const int bmi2, U32 lastBlock,
void* workspace, size_t wkspSize)
{
const seqDef* const sstart = seqStorePtr->sequencesStart;
const seqDef* const send = seqStorePtr->sequences;
const seqDef* sp = sstart;
const BYTE* const lstart = seqStorePtr->litStart;
const BYTE* const lend = seqStorePtr->lit;
const BYTE* lp = lstart;
BYTE const* ip = (BYTE const*)src;
BYTE const* const iend = ip + srcSize;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
const BYTE* llCodePtr = seqStorePtr->llCode;
const BYTE* mlCodePtr = seqStorePtr->mlCode;
const BYTE* ofCodePtr = seqStorePtr->ofCode;
size_t targetCBlockSize = cctxParams->targetCBlockSize;
size_t litSize, seqCount;
int writeLitEntropy = entropyMetadata->hufMetadata.hType == set_compressed;
int writeSeqEntropy = 1;
int lastSequence = 0;
DEBUGLOG(5, "ZSTD_compressSubBlock_multi (litSize=%u, nbSeq=%u)",
(unsigned)(lend-lp), (unsigned)(send-sstart));
litSize = 0;
seqCount = 0;
do {
size_t cBlockSizeEstimate = 0;
if (sstart == send) {
lastSequence = 1;
} else {
const seqDef* const sequence = sp + seqCount;
lastSequence = sequence == send - 1;
litSize += ZSTD_getSequenceLength(seqStorePtr, sequence).litLength;
seqCount++;
}
if (lastSequence) {
assert(lp <= lend);
assert(litSize <= (size_t)(lend - lp));
litSize = (size_t)(lend - lp);
}
/* I think there is an optimization opportunity here.
* Calling ZSTD_estimateSubBlockSize for every sequence can be wasteful
* since it recalculates estimate from scratch.
* For example, it would recount literal distribution and symbol codes every time.
*/
cBlockSizeEstimate = ZSTD_estimateSubBlockSize(lp, litSize, ofCodePtr, llCodePtr, mlCodePtr, seqCount,
&nextCBlock->entropy, entropyMetadata,
workspace, wkspSize, writeLitEntropy, writeSeqEntropy);
if (cBlockSizeEstimate > targetCBlockSize || lastSequence) {
int litEntropyWritten = 0;
int seqEntropyWritten = 0;
const size_t decompressedSize = ZSTD_seqDecompressedSize(seqStorePtr, sp, seqCount, litSize, lastSequence);
const size_t cSize = ZSTD_compressSubBlock(&nextCBlock->entropy, entropyMetadata,
sp, seqCount,
lp, litSize,
llCodePtr, mlCodePtr, ofCodePtr,
cctxParams,
op, oend-op,
bmi2, writeLitEntropy, writeSeqEntropy,
&litEntropyWritten, &seqEntropyWritten,
lastBlock && lastSequence);
FORWARD_IF_ERROR(cSize, "ZSTD_compressSubBlock failed");
if (cSize > 0 && cSize < decompressedSize) {
DEBUGLOG(5, "Committed the sub-block");
assert(ip + decompressedSize <= iend);
ip += decompressedSize;
sp += seqCount;
lp += litSize;
op += cSize;
llCodePtr += seqCount;
mlCodePtr += seqCount;
ofCodePtr += seqCount;
litSize = 0;
seqCount = 0;
/* Entropy only needs to be written once */
if (litEntropyWritten) {
writeLitEntropy = 0;
}
if (seqEntropyWritten) {
writeSeqEntropy = 0;
}
}
}
} while (!lastSequence);
if (writeLitEntropy) {
DEBUGLOG(5, "ZSTD_compressSubBlock_multi has literal entropy tables unwritten");
ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf));
}
if (writeSeqEntropy && ZSTD_needSequenceEntropyTables(&entropyMetadata->fseMetadata)) {
/* If we haven't written our entropy tables, then we've violated our contract and
* must emit an uncompressed block.
*/
DEBUGLOG(5, "ZSTD_compressSubBlock_multi has sequence entropy tables unwritten");
return 0;
}
if (ip < iend) {
size_t const cSize = ZSTD_noCompressBlock(op, oend - op, ip, iend - ip, lastBlock);
DEBUGLOG(5, "ZSTD_compressSubBlock_multi last sub-block uncompressed, %zu bytes", (size_t)(iend - ip));
FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed");
assert(cSize != 0);
op += cSize;
/* We have to regenerate the repcodes because we've skipped some sequences */
if (sp < send) {
seqDef const* seq;
repcodes_t rep;
ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep));
for (seq = sstart; seq < sp; ++seq) {
ZSTD_updateRep(rep.rep, seq->offBase, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);
}
ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep));
}
}
DEBUGLOG(5, "ZSTD_compressSubBlock_multi compressed");
return op-ostart;
}
size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
void* dst, size_t dstCapacity,
void const* src, size_t srcSize,
unsigned lastBlock) {
ZSTD_entropyCTablesMetadata_t entropyMetadata;
FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore,
&zc->blockState.prevCBlock->entropy,
&zc->blockState.nextCBlock->entropy,
&zc->appliedParams,
&entropyMetadata,
zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), "");
return ZSTD_compressSubBlock_multi(&zc->seqStore,
zc->blockState.prevCBlock,
zc->blockState.nextCBlock,
&entropyMetadata,
&zc->appliedParams,
dst, dstCapacity,
src, srcSize,
zc->bmi2, lastBlock,
zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */);
}

32
ext/zstd/lib/compress/zstd_compress_superblock.h Normal file
View File

@@ -0,0 +1,32 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_ADVANCED_H
#define ZSTD_COMPRESS_ADVANCED_H
/*-*************************************
* Dependencies
***************************************/
#include "../zstd.h" /* ZSTD_CCtx */
/*-*************************************
* Target Compressed Block Size
***************************************/
/* ZSTD_compressSuperBlock() :
* Used to compress a super block when targetCBlockSize is being used.
* The given block will be compressed into multiple sub blocks that are around targetCBlockSize. */
size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
void* dst, size_t dstCapacity,
void const* src, size_t srcSize,
unsigned lastBlock);
#endif /* ZSTD_COMPRESS_ADVANCED_H */

742
ext/zstd/lib/compress/zstd_cwksp.h Normal file
View File

@@ -0,0 +1,742 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CWKSP_H
#define ZSTD_CWKSP_H
/*-*************************************
* Dependencies
***************************************/
#include "../common/allocations.h" /* ZSTD_customMalloc, ZSTD_customFree */
#include "../common/zstd_internal.h"
#include "../common/portability_macros.h"
#if defined (__cplusplus)
extern "C" {
#endif
/*-*************************************
* Constants
***************************************/
/* Since the workspace is effectively its own little malloc implementation /
* arena, when we run under ASAN, we should similarly insert redzones between
* each internal element of the workspace, so ASAN will catch overruns that
* reach outside an object but that stay inside the workspace.
*
* This defines the size of that redzone.
*/
#ifndef ZSTD_CWKSP_ASAN_REDZONE_SIZE
#define ZSTD_CWKSP_ASAN_REDZONE_SIZE 128
#endif
/* Set our tables and aligneds to align by 64 bytes */
#define ZSTD_CWKSP_ALIGNMENT_BYTES 64
/*-*************************************
* Structures
***************************************/
typedef enum {
ZSTD_cwksp_alloc_objects,
ZSTD_cwksp_alloc_aligned_init_once,
ZSTD_cwksp_alloc_aligned,
ZSTD_cwksp_alloc_buffers
} ZSTD_cwksp_alloc_phase_e;
/**
* Used to describe whether the workspace is statically allocated (and will not
* necessarily ever be freed), or if it's dynamically allocated and we can
* expect a well-formed caller to free this.
*/
typedef enum {
ZSTD_cwksp_dynamic_alloc,
ZSTD_cwksp_static_alloc
} ZSTD_cwksp_static_alloc_e;
/**
* Zstd fits all its internal datastructures into a single continuous buffer,
* so that it only needs to perform a single OS allocation (or so that a buffer
* can be provided to it and it can perform no allocations at all). This buffer
* is called the workspace.
*
* Several optimizations complicate that process of allocating memory ranges
* from this workspace for each internal datastructure:
*
* - These different internal datastructures have different setup requirements:
*
* - The static objects need to be cleared once and can then be trivially
* reused for each compression.
*
* - Various buffers don't need to be initialized at all--they are always
* written into before they're read.
*
* - The matchstate tables have a unique requirement that they don't need
* their memory to be totally cleared, but they do need the memory to have
* some bound, i.e., a guarantee that all values in the memory they've been
* allocated is less than some maximum value (which is the starting value
* for the indices that they will then use for compression). When this
* guarantee is provided to them, they can use the memory without any setup
* work. When it can't, they have to clear the area.
*
* - These buffers also have different alignment requirements.
*
* - We would like to reuse the objects in the workspace for multiple
* compressions without having to perform any expensive reallocation or
* reinitialization work.
*
* - We would like to be able to efficiently reuse the workspace across
* multiple compressions **even when the compression parameters change** and
* we need to resize some of the objects (where possible).
*
* To attempt to manage this buffer, given these constraints, the ZSTD_cwksp
* abstraction was created. It works as follows:
*
* Workspace Layout:
*
* [ ... workspace ... ]
* [objects][tables ->] free space [<- buffers][<- aligned][<- init once]
*
* The various objects that live in the workspace are divided into the
* following categories, and are allocated separately:
*
* - Static objects: this is optionally the enclosing ZSTD_CCtx or ZSTD_CDict,
* so that literally everything fits in a single buffer. Note: if present,
* this must be the first object in the workspace, since ZSTD_customFree{CCtx,
* CDict}() rely on a pointer comparison to see whether one or two frees are
* required.
*
* - Fixed size objects: these are fixed-size, fixed-count objects that are
* nonetheless "dynamically" allocated in the workspace so that we can
* control how they're initialized separately from the broader ZSTD_CCtx.
* Examples:
* - Entropy Workspace
* - 2 x ZSTD_compressedBlockState_t
* - CDict dictionary contents
*
* - Tables: these are any of several different datastructures (hash tables,
* chain tables, binary trees) that all respect a common format: they are
* uint32_t arrays, all of whose values are between 0 and (nextSrc - base).
* Their sizes depend on the cparams. These tables are 64-byte aligned.
*
* - Init once: these buffers require to be initialized at least once before
* use. They should be used when we want to skip memory initialization
* while not triggering memory checkers (like Valgrind) when reading from
* from this memory without writing to it first.
* These buffers should be used carefully as they might contain data
* from previous compressions.
* Buffers are aligned to 64 bytes.
*
* - Aligned: these buffers don't require any initialization before they're
* used. The user of the buffer should make sure they write into a buffer
* location before reading from it.
* Buffers are aligned to 64 bytes.
*
* - Buffers: these buffers are used for various purposes that don't require
* any alignment or initialization before they're used. This means they can
* be moved around at no cost for a new compression.
*
* Allocating Memory:
*
* The various types of objects must be allocated in order, so they can be
* correctly packed into the workspace buffer. That order is:
*
* 1. Objects
* 2. Init once / Tables
* 3. Aligned / Tables
* 4. Buffers / Tables
*
* Attempts to reserve objects of different types out of order will fail.
*/
typedef struct {
void* workspace;
void* workspaceEnd;
void* objectEnd;
void* tableEnd;
void* tableValidEnd;
void* allocStart;
void* initOnceStart;
BYTE allocFailed;
int workspaceOversizedDuration;
ZSTD_cwksp_alloc_phase_e phase;
ZSTD_cwksp_static_alloc_e isStatic;
} ZSTD_cwksp;
/*-*************************************
* Functions
***************************************/
MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws);
MEM_STATIC void* ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws);
MEM_STATIC void ZSTD_cwksp_assert_internal_consistency(ZSTD_cwksp* ws) {
(void)ws;
assert(ws->workspace <= ws->objectEnd);
assert(ws->objectEnd <= ws->tableEnd);
assert(ws->objectEnd <= ws->tableValidEnd);
assert(ws->tableEnd <= ws->allocStart);
assert(ws->tableValidEnd <= ws->allocStart);
assert(ws->allocStart <= ws->workspaceEnd);
assert(ws->initOnceStart <= ZSTD_cwksp_initialAllocStart(ws));
assert(ws->workspace <= ws->initOnceStart);
#if ZSTD_MEMORY_SANITIZER
{
intptr_t const offset = __msan_test_shadow(ws->initOnceStart,
(U8*)ZSTD_cwksp_initialAllocStart(ws) - (U8*)ws->initOnceStart);
#if defined(ZSTD_MSAN_PRINT)
if(offset!=-1) {
__msan_print_shadow((U8*)ws->initOnceStart + offset - 8, 32);
}
#endif
assert(offset==-1);
};
#endif
}
/**
* Align must be a power of 2.
*/
MEM_STATIC size_t ZSTD_cwksp_align(size_t size, size_t const align) {
size_t const mask = align - 1;
assert((align & mask) == 0);
return (size + mask) & ~mask;
}
/**
* Use this to determine how much space in the workspace we will consume to
* allocate this object. (Normally it should be exactly the size of the object,
* but under special conditions, like ASAN, where we pad each object, it might
* be larger.)
*
* Since tables aren't currently redzoned, you don't need to call through this
* to figure out how much space you need for the matchState tables. Everything
* else is though.
*
* Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size().
*/
MEM_STATIC size_t ZSTD_cwksp_alloc_size(size_t size) {
if (size == 0)
return 0;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
return size + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#else
return size;
#endif
}
/**
* Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes.
* Used to determine the number of bytes required for a given "aligned".
*/
MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) {
return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES));
}
/**
* Returns the amount of additional space the cwksp must allocate
* for internal purposes (currently only alignment).
*/
MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) {
/* For alignment, the wksp will always allocate an additional 2*ZSTD_CWKSP_ALIGNMENT_BYTES
* bytes to align the beginning of tables section and end of buffers;
*/
size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES * 2;
return slackSpace;
}
/**
* Return the number of additional bytes required to align a pointer to the given number of bytes.
* alignBytes must be a power of two.
*/
MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) {
size_t const alignBytesMask = alignBytes - 1;
size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask;
assert((alignBytes & alignBytesMask) == 0);
assert(bytes < alignBytes);
return bytes;
}
/**
* Returns the initial value for allocStart which is used to determine the position from
* which we can allocate from the end of the workspace.
*/
MEM_STATIC void* ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws) {
return (void*)((size_t)ws->workspaceEnd & ~(ZSTD_CWKSP_ALIGNMENT_BYTES-1));
}
/**
* Internal function. Do not use directly.
* Reserves the given number of bytes within the aligned/buffer segment of the wksp,
* which counts from the end of the wksp (as opposed to the object/table segment).
*
* Returns a pointer to the beginning of that space.
*/
MEM_STATIC void*
ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes)
{
void* const alloc = (BYTE*)ws->allocStart - bytes;
void* const bottom = ws->tableEnd;
DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining",
alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
ZSTD_cwksp_assert_internal_consistency(ws);
assert(alloc >= bottom);
if (alloc < bottom) {
DEBUGLOG(4, "cwksp: alloc failed!");
ws->allocFailed = 1;
return NULL;
}
/* the area is reserved from the end of wksp.
* If it overlaps with tableValidEnd, it voids guarantees on values' range */
if (alloc < ws->tableValidEnd) {
ws->tableValidEnd = alloc;
}
ws->allocStart = alloc;
return alloc;
}
/**
* Moves the cwksp to the next phase, and does any necessary allocations.
* cwksp initialization must necessarily go through each phase in order.
* Returns a 0 on success, or zstd error
*/
MEM_STATIC size_t
ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase)
{
assert(phase >= ws->phase);
if (phase > ws->phase) {
/* Going from allocating objects to allocating initOnce / tables */
if (ws->phase < ZSTD_cwksp_alloc_aligned_init_once &&
phase >= ZSTD_cwksp_alloc_aligned_init_once) {
ws->tableValidEnd = ws->objectEnd;
ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
{ /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */
void *const alloc = ws->objectEnd;
size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES);
void *const objectEnd = (BYTE *) alloc + bytesToAlign;
DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign);
RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation,
"table phase - alignment initial allocation failed!");
ws->objectEnd = objectEnd;
ws->tableEnd = objectEnd; /* table area starts being empty */
if (ws->tableValidEnd < ws->tableEnd) {
ws->tableValidEnd = ws->tableEnd;
}
}
}
ws->phase = phase;
ZSTD_cwksp_assert_internal_consistency(ws);
}
return 0;
}
/**
* Returns whether this object/buffer/etc was allocated in this workspace.
*/
MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr)
{
return (ptr != NULL) && (ws->workspace <= ptr) && (ptr < ws->workspaceEnd);
}
/**
* Internal function. Do not use directly.
*/
MEM_STATIC void*
ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase)
{
void* alloc;
if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) {
return NULL;
}
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* over-reserve space */
bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#endif
alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes);
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
* either size. */
if (alloc) {
alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
/* We need to keep the redzone poisoned while unpoisoning the bytes that
* are actually allocated. */
__asan_unpoison_memory_region(alloc, bytes - 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE);
}
}
#endif
return alloc;
}
/**
* Reserves and returns unaligned memory.
*/
MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes)
{
return (BYTE*)ZSTD_cwksp_reserve_internal(ws, bytes, ZSTD_cwksp_alloc_buffers);
}
/**
* Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
* This memory has been initialized at least once in the past.
* This doesn't mean it has been initialized this time, and it might contain data from previous
* operations.
* The main usage is for algorithms that might need read access into uninitialized memory.
* The algorithm must maintain safety under these conditions and must make sure it doesn't
* leak any of the past data (directly or in side channels).
*/
MEM_STATIC void* ZSTD_cwksp_reserve_aligned_init_once(ZSTD_cwksp* ws, size_t bytes)
{
size_t const alignedBytes = ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES);
void* ptr = ZSTD_cwksp_reserve_internal(ws, alignedBytes, ZSTD_cwksp_alloc_aligned_init_once);
assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
if(ptr && ptr < ws->initOnceStart) {
/* We assume the memory following the current allocation is either:
* 1. Not usable as initOnce memory (end of workspace)
* 2. Another initOnce buffer that has been allocated before (and so was previously memset)
* 3. An ASAN redzone, in which case we don't want to write on it
* For these reasons it should be fine to not explicitly zero every byte up to ws->initOnceStart.
* Note that we assume here that MSAN and ASAN cannot run in the same time. */
ZSTD_memset(ptr, 0, MIN((size_t)((U8*)ws->initOnceStart - (U8*)ptr), alignedBytes));
ws->initOnceStart = ptr;
}
#if ZSTD_MEMORY_SANITIZER
assert(__msan_test_shadow(ptr, bytes) == -1);
#endif
return ptr;
}
/**
* Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
*/
MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes)
{
void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES),
ZSTD_cwksp_alloc_aligned);
assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
return ptr;
}
/**
* Aligned on 64 bytes. These buffers have the special property that
* their values remain constrained, allowing us to re-use them without
* memset()-ing them.
*/
MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes)
{
const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned_init_once;
void* alloc;
void* end;
void* top;
/* We can only start allocating tables after we are done reserving space for objects at the
* start of the workspace */
if(ws->phase < phase) {
if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) {
return NULL;
}
}
alloc = ws->tableEnd;
end = (BYTE *)alloc + bytes;
top = ws->allocStart;
DEBUGLOG(5, "cwksp: reserving %p table %zd bytes, %zd bytes remaining",
alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
assert((bytes & (sizeof(U32)-1)) == 0);
ZSTD_cwksp_assert_internal_consistency(ws);
assert(end <= top);
if (end > top) {
DEBUGLOG(4, "cwksp: table alloc failed!");
ws->allocFailed = 1;
return NULL;
}
ws->tableEnd = end;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
__asan_unpoison_memory_region(alloc, bytes);
}
#endif
assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0);
assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
return alloc;
}
/**
* Aligned on sizeof(void*).
* Note : should happen only once, at workspace first initialization
*/
MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes)
{
size_t const roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
void* alloc = ws->objectEnd;
void* end = (BYTE*)alloc + roundedBytes;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* over-reserve space */
end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#endif
DEBUGLOG(4,
"cwksp: reserving %p object %zd bytes (rounded to %zd), %zd bytes remaining",
alloc, bytes, roundedBytes, ZSTD_cwksp_available_space(ws) - roundedBytes);
assert((size_t)alloc % ZSTD_ALIGNOF(void*) == 0);
assert(bytes % ZSTD_ALIGNOF(void*) == 0);
ZSTD_cwksp_assert_internal_consistency(ws);
/* we must be in the first phase, no advance is possible */
if (ws->phase != ZSTD_cwksp_alloc_objects || end > ws->workspaceEnd) {
DEBUGLOG(3, "cwksp: object alloc failed!");
ws->allocFailed = 1;
return NULL;
}
ws->objectEnd = end;
ws->tableEnd = end;
ws->tableValidEnd = end;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
* either size. */
alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
__asan_unpoison_memory_region(alloc, bytes);
}
#endif
return alloc;
}
MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws)
{
DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty");
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* To validate that the table re-use logic is sound, and that we don't
* access table space that we haven't cleaned, we re-"poison" the table
* space every time we mark it dirty.
* Since tableValidEnd space and initOnce space may overlap we don't poison
* the initOnce portion as it break its promise. This means that this poisoning
* check isn't always applied fully. */
{
size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
assert(__msan_test_shadow(ws->objectEnd, size) == -1);
if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
__msan_poison(ws->objectEnd, size);
} else {
assert(ws->initOnceStart >= ws->objectEnd);
__msan_poison(ws->objectEnd, (BYTE*)ws->initOnceStart - (BYTE*)ws->objectEnd);
}
}
#endif
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
ws->tableValidEnd = ws->objectEnd;
ZSTD_cwksp_assert_internal_consistency(ws);
}
MEM_STATIC void ZSTD_cwksp_mark_tables_clean(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_clean");
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
if (ws->tableValidEnd < ws->tableEnd) {
ws->tableValidEnd = ws->tableEnd;
}
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* Zero the part of the allocated tables not already marked clean.
*/
MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: ZSTD_cwksp_clean_tables");
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
if (ws->tableValidEnd < ws->tableEnd) {
ZSTD_memset(ws->tableValidEnd, 0, (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd));
}
ZSTD_cwksp_mark_tables_clean(ws);
}
/**
* Invalidates table allocations.
* All other allocations remain valid.
*/
MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: clearing tables!");
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* We don't do this when the workspace is statically allocated, because
* when that is the case, we have no capability to hook into the end of the
* workspace's lifecycle to unpoison the memory.
*/
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
__asan_poison_memory_region(ws->objectEnd, size);
}
#endif
ws->tableEnd = ws->objectEnd;
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* Invalidates all buffer, aligned, and table allocations.
* Object allocations remain valid.
*/
MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: clearing!");
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* To validate that the context re-use logic is sound, and that we don't
* access stuff that this compression hasn't initialized, we re-"poison"
* the workspace except for the areas in which we expect memory re-use
* without initialization (objects, valid tables area and init once
* memory). */
{
if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
size_t size = (BYTE*)ws->initOnceStart - (BYTE*)ws->tableValidEnd;
__msan_poison(ws->tableValidEnd, size);
}
}
#endif
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* We don't do this when the workspace is statically allocated, because
* when that is the case, we have no capability to hook into the end of the
* workspace's lifecycle to unpoison the memory.
*/
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->objectEnd;
__asan_poison_memory_region(ws->objectEnd, size);
}
#endif
ws->tableEnd = ws->objectEnd;
ws->allocStart = ZSTD_cwksp_initialAllocStart(ws);
ws->allocFailed = 0;
if (ws->phase > ZSTD_cwksp_alloc_aligned_init_once) {
ws->phase = ZSTD_cwksp_alloc_aligned_init_once;
}
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* The provided workspace takes ownership of the buffer [start, start+size).
* Any existing values in the workspace are ignored (the previously managed
* buffer, if present, must be separately freed).
*/
MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) {
DEBUGLOG(4, "cwksp: init'ing workspace with %zd bytes", size);
assert(((size_t)start & (sizeof(void*)-1)) == 0); /* ensure correct alignment */
ws->workspace = start;
ws->workspaceEnd = (BYTE*)start + size;
ws->objectEnd = ws->workspace;
ws->tableValidEnd = ws->objectEnd;
ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
ws->phase = ZSTD_cwksp_alloc_objects;
ws->isStatic = isStatic;
ZSTD_cwksp_clear(ws);
ws->workspaceOversizedDuration = 0;
ZSTD_cwksp_assert_internal_consistency(ws);
}
MEM_STATIC size_t ZSTD_cwksp_create(ZSTD_cwksp* ws, size_t size, ZSTD_customMem customMem) {
void* workspace = ZSTD_customMalloc(size, customMem);
DEBUGLOG(4, "cwksp: creating new workspace with %zd bytes", size);
RETURN_ERROR_IF(workspace == NULL, memory_allocation, "NULL pointer!");
ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc);
return 0;
}
MEM_STATIC void ZSTD_cwksp_free(ZSTD_cwksp* ws, ZSTD_customMem customMem) {
void *ptr = ws->workspace;
DEBUGLOG(4, "cwksp: freeing workspace");
ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp));
ZSTD_customFree(ptr, customMem);
}
/**
* Moves the management of a workspace from one cwksp to another. The src cwksp
* is left in an invalid state (src must be re-init()'ed before it's used again).
*/
MEM_STATIC void ZSTD_cwksp_move(ZSTD_cwksp* dst, ZSTD_cwksp* src) {
*dst = *src;
ZSTD_memset(src, 0, sizeof(ZSTD_cwksp));
}
MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace);
}
MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace)
+ (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart);
}
MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) {
return ws->allocFailed;
}
/*-*************************************
* Functions Checking Free Space
***************************************/
/* ZSTD_alignmentSpaceWithinBounds() :
* Returns if the estimated space needed for a wksp is within an acceptable limit of the
* actual amount of space used.
*/
MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp *const ws, size_t const estimatedSpace) {
/* We have an alignment space between objects and tables between tables and buffers, so we can have up to twice
* the alignment bytes difference between estimation and actual usage */
return (estimatedSpace - ZSTD_cwksp_slack_space_required()) <= ZSTD_cwksp_used(ws) &&
ZSTD_cwksp_used(ws) <= estimatedSpace;
}
MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->allocStart - (BYTE*)ws->tableEnd);
}
MEM_STATIC int ZSTD_cwksp_check_available(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_available_space(ws) >= additionalNeededSpace;
}
MEM_STATIC int ZSTD_cwksp_check_too_large(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_check_available(
ws, additionalNeededSpace * ZSTD_WORKSPACETOOLARGE_FACTOR);
}
MEM_STATIC int ZSTD_cwksp_check_wasteful(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_check_too_large(ws, additionalNeededSpace)
&& ws->workspaceOversizedDuration > ZSTD_WORKSPACETOOLARGE_MAXDURATION;
}
MEM_STATIC void ZSTD_cwksp_bump_oversized_duration(
ZSTD_cwksp* ws, size_t additionalNeededSpace) {
if (ZSTD_cwksp_check_too_large(ws, additionalNeededSpace)) {
ws->workspaceOversizedDuration++;
} else {
ws->workspaceOversizedDuration = 0;
}
}
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_CWKSP_H */

758
ext/zstd/lib/compress/zstd_double_fast.c vendored Normal file
View File

@@ -0,0 +1,758 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_compress_internal.h"
#include "zstd_double_fast.h"
static void ZSTD_fillDoubleHashTableForCDict(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
U32 const mls = cParams->minMatch;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Always insert every fastHashFillStep position into the hash tables.
* Insert the other positions into the large hash table if their entry
* is empty.
*/
for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
U32 i;
for (i = 0; i < fastHashFillStep; ++i) {
size_t const smHashAndTag = ZSTD_hashPtr(ip + i, hBitsS, mls);
size_t const lgHashAndTag = ZSTD_hashPtr(ip + i, hBitsL, 8);
if (i == 0) {
ZSTD_writeTaggedIndex(hashSmall, smHashAndTag, curr + i);
}
if (i == 0 || hashLarge[lgHashAndTag >> ZSTD_SHORT_CACHE_TAG_BITS] == 0) {
ZSTD_writeTaggedIndex(hashLarge, lgHashAndTag, curr + i);
}
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
} }
}
static void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
U32 const mls = cParams->minMatch;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Always insert every fastHashFillStep position into the hash tables.
* Insert the other positions into the large hash table if their entry
* is empty.
*/
for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
U32 i;
for (i = 0; i < fastHashFillStep; ++i) {
size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls);
size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8);
if (i == 0)
hashSmall[smHash] = curr + i;
if (i == 0 || hashLarge[lgHash] == 0)
hashLarge[lgHash] = curr + i;
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
} }
}
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
{
if (tfp == ZSTD_tfp_forCDict) {
ZSTD_fillDoubleHashTableForCDict(ms, end, dtlm);
} else {
ZSTD_fillDoubleHashTableForCCtx(ms, end, dtlm);
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_doubleFast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
const U32 hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
const U32 hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
/* presumes that, if there is a dictionary, it must be using Attach mode */
const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixLowest = base + prefixLowestIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
size_t mLength;
U32 offset;
U32 curr;
/* how many positions to search before increasing step size */
const size_t kStepIncr = 1 << kSearchStrength;
/* the position at which to increment the step size if no match is found */
const BYTE* nextStep;
size_t step; /* the current step size */
size_t hl0; /* the long hash at ip */
size_t hl1; /* the long hash at ip1 */
U32 idxl0; /* the long match index for ip */
U32 idxl1; /* the long match index for ip1 */
const BYTE* matchl0; /* the long match for ip */
const BYTE* matchs0; /* the short match for ip */
const BYTE* matchl1; /* the long match for ip1 */
const BYTE* ip = istart; /* the current position */
const BYTE* ip1; /* the next position */
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_noDict_generic");
/* init */
ip += ((ip - prefixLowest) == 0);
{
U32 const current = (U32)(ip - base);
U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, current, cParams->windowLog);
U32 const maxRep = current - windowLow;
if (offset_2 > maxRep) offsetSaved2 = offset_2, offset_2 = 0;
if (offset_1 > maxRep) offsetSaved1 = offset_1, offset_1 = 0;
}
/* Outer Loop: one iteration per match found and stored */
while (1) {
step = 1;
nextStep = ip + kStepIncr;
ip1 = ip + step;
if (ip1 > ilimit) {
goto _cleanup;
}
hl0 = ZSTD_hashPtr(ip, hBitsL, 8);
idxl0 = hashLong[hl0];
matchl0 = base + idxl0;
/* Inner Loop: one iteration per search / position */
do {
const size_t hs0 = ZSTD_hashPtr(ip, hBitsS, mls);
const U32 idxs0 = hashSmall[hs0];
curr = (U32)(ip-base);
matchs0 = base + idxs0;
hashLong[hl0] = hashSmall[hs0] = curr; /* update hash tables */
/* check noDict repcode */
if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
goto _match_stored;
}
hl1 = ZSTD_hashPtr(ip1, hBitsL, 8);
if (idxl0 > prefixLowestIndex) {
/* check prefix long match */
if (MEM_read64(matchl0) == MEM_read64(ip)) {
mLength = ZSTD_count(ip+8, matchl0+8, iend) + 8;
offset = (U32)(ip-matchl0);
while (((ip>anchor) & (matchl0>prefixLowest)) && (ip[-1] == matchl0[-1])) { ip--; matchl0--; mLength++; } /* catch up */
goto _match_found;
}
}
idxl1 = hashLong[hl1];
matchl1 = base + idxl1;
if (idxs0 > prefixLowestIndex) {
/* check prefix short match */
if (MEM_read32(matchs0) == MEM_read32(ip)) {
goto _search_next_long;
}
}
if (ip1 >= nextStep) {
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
step++;
nextStep += kStepIncr;
}
ip = ip1;
ip1 += step;
hl0 = hl1;
idxl0 = idxl1;
matchl0 = matchl1;
#if defined(__aarch64__)
PREFETCH_L1(ip+256);
#endif
} while (ip1 <= ilimit);
_cleanup:
/* If offset_1 started invalid (offsetSaved1 != 0) and became valid (offset_1 != 0),
* rotate saved offsets. See comment in ZSTD_compressBlock_fast_noDict for more context. */
offsetSaved2 = ((offsetSaved1 != 0) && (offset_1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved1;
rep[1] = offset_2 ? offset_2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_search_next_long:
/* check prefix long +1 match */
if (idxl1 > prefixLowestIndex) {
if (MEM_read64(matchl1) == MEM_read64(ip1)) {
ip = ip1;
mLength = ZSTD_count(ip+8, matchl1+8, iend) + 8;
offset = (U32)(ip-matchl1);
while (((ip>anchor) & (matchl1>prefixLowest)) && (ip[-1] == matchl1[-1])) { ip--; matchl1--; mLength++; } /* catch up */
goto _match_found;
}
}
/* if no long +1 match, explore the short match we found */
mLength = ZSTD_count(ip+4, matchs0+4, iend) + 4;
offset = (U32)(ip - matchs0);
while (((ip>anchor) & (matchs0>prefixLowest)) && (ip[-1] == matchs0[-1])) { ip--; matchs0--; mLength++; } /* catch up */
/* fall-through */
_match_found: /* requires ip, offset, mLength */
offset_2 = offset_1;
offset_1 = offset;
if (step < 4) {
/* It is unsafe to write this value back to the hashtable when ip1 is
* greater than or equal to the new ip we will have after we're done
* processing this match. Rather than perform that test directly
* (ip1 >= ip + mLength), which costs speed in practice, we do a simpler
* more predictable test. The minmatch even if we take a short match is
* 4 bytes, so as long as step, the distance between ip and ip1
* (initially) is less than 4, we know ip1 < new ip. */
hashLong[hl1] = (U32)(ip1 - base);
}
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
_match_stored:
/* match found */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while ( (ip <= ilimit)
&& ( (offset_2>0)
& (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
/* store sequence */
size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; /* swap offset_2 <=> offset_1 */
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, rLength);
ip += rLength;
anchor = ip;
continue; /* faster when present ... (?) */
}
}
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
const U32 hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
const U32 hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
/* presumes that, if there is a dictionary, it must be using Attach mode */
const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixLowest = base + prefixLowestIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams;
const U32* const dictHashLong = dms->hashTable;
const U32* const dictHashSmall = dms->chainTable;
const U32 dictStartIndex = dms->window.dictLimit;
const BYTE* const dictBase = dms->window.base;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dms->window.nextSrc;
const U32 dictIndexDelta = prefixLowestIndex - (U32)(dictEnd - dictBase);
const U32 dictHBitsL = dictCParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
const U32 dictHBitsS = dictCParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictStart));
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_dictMatchState_generic");
/* if a dictionary is attached, it must be within window range */
assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex);
if (ms->prefetchCDictTables) {
size_t const hashTableBytes = (((size_t)1) << dictCParams->hashLog) * sizeof(U32);
size_t const chainTableBytes = (((size_t)1) << dictCParams->chainLog) * sizeof(U32);
PREFETCH_AREA(dictHashLong, hashTableBytes)
PREFETCH_AREA(dictHashSmall, chainTableBytes)
}
/* init */
ip += (dictAndPrefixLength == 0);
/* dictMatchState repCode checks don't currently handle repCode == 0
* disabling. */
assert(offset_1 <= dictAndPrefixLength);
assert(offset_2 <= dictAndPrefixLength);
/* Main Search Loop */
while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
size_t mLength;
U32 offset;
size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
size_t const dictHashAndTagL = ZSTD_hashPtr(ip, dictHBitsL, 8);
size_t const dictHashAndTagS = ZSTD_hashPtr(ip, dictHBitsS, mls);
U32 const dictMatchIndexAndTagL = dictHashLong[dictHashAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS];
U32 const dictMatchIndexAndTagS = dictHashSmall[dictHashAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS];
int const dictTagsMatchL = ZSTD_comparePackedTags(dictMatchIndexAndTagL, dictHashAndTagL);
int const dictTagsMatchS = ZSTD_comparePackedTags(dictMatchIndexAndTagS, dictHashAndTagS);
U32 const curr = (U32)(ip-base);
U32 const matchIndexL = hashLong[h2];
U32 matchIndexS = hashSmall[h];
const BYTE* matchLong = base + matchIndexL;
const BYTE* match = base + matchIndexS;
const U32 repIndex = curr + 1 - offset_1;
const BYTE* repMatch = (repIndex < prefixLowestIndex) ?
dictBase + (repIndex - dictIndexDelta) :
base + repIndex;
hashLong[h2] = hashSmall[h] = curr; /* update hash tables */
/* check repcode */
if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
&& (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
goto _match_stored;
}
if (matchIndexL > prefixLowestIndex) {
/* check prefix long match */
if (MEM_read64(matchLong) == MEM_read64(ip)) {
mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8;
offset = (U32)(ip-matchLong);
while (((ip>anchor) & (matchLong>prefixLowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
goto _match_found;
}
} else if (dictTagsMatchL) {
/* check dictMatchState long match */
U32 const dictMatchIndexL = dictMatchIndexAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatchL = dictBase + dictMatchIndexL;
assert(dictMatchL < dictEnd);
if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) {
mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8;
offset = (U32)(curr - dictMatchIndexL - dictIndexDelta);
while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */
goto _match_found;
} }
if (matchIndexS > prefixLowestIndex) {
/* check prefix short match */
if (MEM_read32(match) == MEM_read32(ip)) {
goto _search_next_long;
}
} else if (dictTagsMatchS) {
/* check dictMatchState short match */
U32 const dictMatchIndexS = dictMatchIndexAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS;
match = dictBase + dictMatchIndexS;
matchIndexS = dictMatchIndexS + dictIndexDelta;
if (match > dictStart && MEM_read32(match) == MEM_read32(ip)) {
goto _search_next_long;
} }
ip += ((ip-anchor) >> kSearchStrength) + 1;
#if defined(__aarch64__)
PREFETCH_L1(ip+256);
#endif
continue;
_search_next_long:
{ size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
size_t const dictHashAndTagL3 = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
U32 const matchIndexL3 = hashLong[hl3];
U32 const dictMatchIndexAndTagL3 = dictHashLong[dictHashAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS];
int const dictTagsMatchL3 = ZSTD_comparePackedTags(dictMatchIndexAndTagL3, dictHashAndTagL3);
const BYTE* matchL3 = base + matchIndexL3;
hashLong[hl3] = curr + 1;
/* check prefix long +1 match */
if (matchIndexL3 > prefixLowestIndex) {
if (MEM_read64(matchL3) == MEM_read64(ip+1)) {
mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8;
ip++;
offset = (U32)(ip-matchL3);
while (((ip>anchor) & (matchL3>prefixLowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
goto _match_found;
}
} else if (dictTagsMatchL3) {
/* check dict long +1 match */
U32 const dictMatchIndexL3 = dictMatchIndexAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatchL3 = dictBase + dictMatchIndexL3;
assert(dictMatchL3 < dictEnd);
if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) {
mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8;
ip++;
offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta);
while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */
goto _match_found;
} } }
/* if no long +1 match, explore the short match we found */
if (matchIndexS < prefixLowestIndex) {
mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4;
offset = (U32)(curr - matchIndexS);
while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
} else {
mLength = ZSTD_count(ip+4, match+4, iend) + 4;
offset = (U32)(ip - match);
while (((ip>anchor) & (match>prefixLowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
_match_found:
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
_match_stored:
/* match found */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixLowestIndex ?
dictBase + repIndex2 - dictIndexDelta :
base + repIndex2;
if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
&& (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
anchor = ip;
continue;
}
break;
}
}
} /* while (ip < ilimit) */
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
#define ZSTD_GEN_DFAST_FN(dictMode, mls) \
static size_t ZSTD_compressBlock_doubleFast_##dictMode##_##mls( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_doubleFast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls); \
}
ZSTD_GEN_DFAST_FN(noDict, 4)
ZSTD_GEN_DFAST_FN(noDict, 5)
ZSTD_GEN_DFAST_FN(noDict, 6)
ZSTD_GEN_DFAST_FN(noDict, 7)
ZSTD_GEN_DFAST_FN(dictMatchState, 4)
ZSTD_GEN_DFAST_FN(dictMatchState, 5)
ZSTD_GEN_DFAST_FN(dictMatchState, 6)
ZSTD_GEN_DFAST_FN(dictMatchState, 7)
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_noDict_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_noDict_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_noDict_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_noDict_7(ms, seqStore, rep, src, srcSize);
}
}
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_dictMatchState_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_dictMatchState_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_dictMatchState_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_dictMatchState_7(ms, seqStore, rep, src, srcSize);
}
}
static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
const BYTE* const base = ms->window.base;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
const U32 dictStartIndex = lowLimit;
const U32 dictLimit = ms->window.dictLimit;
const U32 prefixStartIndex = (dictLimit > lowLimit) ? dictLimit : lowLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dictBase + prefixStartIndex;
U32 offset_1=rep[0], offset_2=rep[1];
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_extDict_generic (srcSize=%zu)", srcSize);
/* if extDict is invalidated due to maxDistance, switch to "regular" variant */
if (prefixStartIndex == dictStartIndex)
return ZSTD_compressBlock_doubleFast(ms, seqStore, rep, src, srcSize);
/* Search Loop */
while (ip < ilimit) { /* < instead of <=, because (ip+1) */
const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
const U32 matchIndex = hashSmall[hSmall];
const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base;
const BYTE* match = matchBase + matchIndex;
const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8);
const U32 matchLongIndex = hashLong[hLong];
const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base;
const BYTE* matchLong = matchLongBase + matchLongIndex;
const U32 curr = (U32)(ip-base);
const U32 repIndex = curr + 1 - offset_1; /* offset_1 expected <= curr +1 */
const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
const BYTE* const repMatch = repBase + repIndex;
size_t mLength;
hashSmall[hSmall] = hashLong[hLong] = curr; /* update hash table */
if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
& (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */
&& (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
} else {
if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart;
U32 offset;
mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8;
offset = curr - matchLongIndex;
while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
} else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
U32 const matchIndex3 = hashLong[h3];
const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base;
const BYTE* match3 = match3Base + matchIndex3;
U32 offset;
hashLong[h3] = curr + 1;
if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart;
mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8;
ip++;
offset = curr+1 - matchIndex3;
while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
} else {
const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
offset = curr - matchIndex;
while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
} else {
ip += ((ip-anchor) >> kSearchStrength) + 1;
continue;
} }
/* move to next sequence start */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */
& (offset_2 <= current2 - dictStartIndex))
&& (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
anchor = ip;
continue;
}
break;
} } }
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
ZSTD_GEN_DFAST_FN(extDict, 4)
ZSTD_GEN_DFAST_FN(extDict, 5)
ZSTD_GEN_DFAST_FN(extDict, 6)
ZSTD_GEN_DFAST_FN(extDict, 7)
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_extDict_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_extDict_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_extDict_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_extDict_7(ms, seqStore, rep, src, srcSize);
}
}

39
ext/zstd/lib/compress/zstd_double_fast.h Normal file
View File

@@ -0,0 +1,39 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_DOUBLE_FAST_H
#define ZSTD_DOUBLE_FAST_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "../common/mem.h" /* U32 */
#include "zstd_compress_internal.h" /* ZSTD_CCtx, size_t */
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_DOUBLE_FAST_H */

960
ext/zstd/lib/compress/zstd_fast.c vendored Normal file
View File

@@ -0,0 +1,960 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_compress_internal.h" /* ZSTD_hashPtr, ZSTD_count, ZSTD_storeSeq */
#include "zstd_fast.h"
static void ZSTD_fillHashTableForCDict(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
U32 const mls = cParams->minMatch;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Currently, we always use ZSTD_dtlm_full for filling CDict tables.
* Feel free to remove this assert if there's a good reason! */
assert(dtlm == ZSTD_dtlm_full);
/* Always insert every fastHashFillStep position into the hash table.
* Insert the other positions if their hash entry is empty.
*/
for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
{ size_t const hashAndTag = ZSTD_hashPtr(ip, hBits, mls);
ZSTD_writeTaggedIndex(hashTable, hashAndTag, curr); }
if (dtlm == ZSTD_dtlm_fast) continue;
/* Only load extra positions for ZSTD_dtlm_full */
{ U32 p;
for (p = 1; p < fastHashFillStep; ++p) {
size_t const hashAndTag = ZSTD_hashPtr(ip + p, hBits, mls);
if (hashTable[hashAndTag >> ZSTD_SHORT_CACHE_TAG_BITS] == 0) { /* not yet filled */
ZSTD_writeTaggedIndex(hashTable, hashAndTag, curr + p);
} } } }
}
static void ZSTD_fillHashTableForCCtx(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog;
U32 const mls = cParams->minMatch;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Currently, we always use ZSTD_dtlm_fast for filling CCtx tables.
* Feel free to remove this assert if there's a good reason! */
assert(dtlm == ZSTD_dtlm_fast);
/* Always insert every fastHashFillStep position into the hash table.
* Insert the other positions if their hash entry is empty.
*/
for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls);
hashTable[hash0] = curr;
if (dtlm == ZSTD_dtlm_fast) continue;
/* Only load extra positions for ZSTD_dtlm_full */
{ U32 p;
for (p = 1; p < fastHashFillStep; ++p) {
size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls);
if (hashTable[hash] == 0) { /* not yet filled */
hashTable[hash] = curr + p;
} } } }
}
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
{
if (tfp == ZSTD_tfp_forCDict) {
ZSTD_fillHashTableForCDict(ms, end, dtlm);
} else {
ZSTD_fillHashTableForCCtx(ms, end, dtlm);
}
}
/**
* If you squint hard enough (and ignore repcodes), the search operation at any
* given position is broken into 4 stages:
*
* 1. Hash (map position to hash value via input read)
* 2. Lookup (map hash val to index via hashtable read)
* 3. Load (map index to value at that position via input read)
* 4. Compare
*
* Each of these steps involves a memory read at an address which is computed
* from the previous step. This means these steps must be sequenced and their
* latencies are cumulative.
*
* Rather than do 1->2->3->4 sequentially for a single position before moving
* onto the next, this implementation interleaves these operations across the
* next few positions:
*
* R = Repcode Read & Compare
* H = Hash
* T = Table Lookup
* M = Match Read & Compare
*
* Pos | Time -->
* ----+-------------------
* N | ... M
* N+1 | ... TM
* N+2 | R H T M
* N+3 | H TM
* N+4 | R H T M
* N+5 | H ...
* N+6 | R ...
*
* This is very much analogous to the pipelining of execution in a CPU. And just
* like a CPU, we have to dump the pipeline when we find a match (i.e., take a
* branch).
*
* When this happens, we throw away our current state, and do the following prep
* to re-enter the loop:
*
* Pos | Time -->
* ----+-------------------
* N | H T
* N+1 | H
*
* This is also the work we do at the beginning to enter the loop initially.
*/
FORCE_INLINE_TEMPLATE size_t
ZSTD_compressBlock_fast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
size_t const stepSize = hasStep ? (cParams->targetLength + !(cParams->targetLength) + 1) : 2;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 prefixStartIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
const BYTE* anchor = istart;
const BYTE* ip0 = istart;
const BYTE* ip1;
const BYTE* ip2;
const BYTE* ip3;
U32 current0;
U32 rep_offset1 = rep[0];
U32 rep_offset2 = rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
size_t hash0; /* hash for ip0 */
size_t hash1; /* hash for ip1 */
U32 idx; /* match idx for ip0 */
U32 mval; /* src value at match idx */
U32 offcode;
const BYTE* match0;
size_t mLength;
/* ip0 and ip1 are always adjacent. The targetLength skipping and
* uncompressibility acceleration is applied to every other position,
* matching the behavior of #1562. step therefore represents the gap
* between pairs of positions, from ip0 to ip2 or ip1 to ip3. */
size_t step;
const BYTE* nextStep;
const size_t kStepIncr = (1 << (kSearchStrength - 1));
DEBUGLOG(5, "ZSTD_compressBlock_fast_generic");
ip0 += (ip0 == prefixStart);
{ U32 const curr = (U32)(ip0 - base);
U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog);
U32 const maxRep = curr - windowLow;
if (rep_offset2 > maxRep) offsetSaved2 = rep_offset2, rep_offset2 = 0;
if (rep_offset1 > maxRep) offsetSaved1 = rep_offset1, rep_offset1 = 0;
}
/* start each op */
_start: /* Requires: ip0 */
step = stepSize;
nextStep = ip0 + kStepIncr;
/* calculate positions, ip0 - anchor == 0, so we skip step calc */
ip1 = ip0 + 1;
ip2 = ip0 + step;
ip3 = ip2 + 1;
if (ip3 >= ilimit) {
goto _cleanup;
}
hash0 = ZSTD_hashPtr(ip0, hlog, mls);
hash1 = ZSTD_hashPtr(ip1, hlog, mls);
idx = hashTable[hash0];
do {
/* load repcode match for ip[2]*/
const U32 rval = MEM_read32(ip2 - rep_offset1);
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* check repcode at ip[2] */
if ((MEM_read32(ip2) == rval) & (rep_offset1 > 0)) {
ip0 = ip2;
match0 = ip0 - rep_offset1;
mLength = ip0[-1] == match0[-1];
ip0 -= mLength;
match0 -= mLength;
offcode = REPCODE1_TO_OFFBASE;
mLength += 4;
/* First write next hash table entry; we've already calculated it.
* This write is known to be safe because the ip1 is before the
* repcode (ip2). */
hashTable[hash1] = (U32)(ip1 - base);
goto _match;
}
/* load match for ip[0] */
if (idx >= prefixStartIndex) {
mval = MEM_read32(base + idx);
} else {
mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */
}
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
/* First write next hash table entry; we've already calculated it.
* This write is known to be safe because the ip1 == ip0 + 1, so
* we know we will resume searching after ip1 */
hashTable[hash1] = (U32)(ip1 - base);
goto _offset;
}
/* lookup ip[1] */
idx = hashTable[hash1];
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip3;
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* load match for ip[0] */
if (idx >= prefixStartIndex) {
mval = MEM_read32(base + idx);
} else {
mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */
}
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
/* first write next hash table entry; we've already calculated it */
if (step <= 4) {
/* We need to avoid writing an index into the hash table >= the
* position at which we will pick up our searching after we've
* taken this match.
*
* The minimum possible match has length 4, so the earliest ip0
* can be after we take this match will be the current ip0 + 4.
* ip1 is ip0 + step - 1. If ip1 is >= ip0 + 4, we can't safely
* write this position.
*/
hashTable[hash1] = (U32)(ip1 - base);
}
goto _offset;
}
/* lookup ip[1] */
idx = hashTable[hash1];
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip0 + step;
ip3 = ip1 + step;
/* calculate step */
if (ip2 >= nextStep) {
step++;
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
nextStep += kStepIncr;
}
} while (ip3 < ilimit);
_cleanup:
/* Note that there are probably still a couple positions we could search.
* However, it seems to be a meaningful performance hit to try to search
* them. So let's not. */
/* When the repcodes are outside of the prefix, we set them to zero before the loop.
* When the offsets are still zero, we need to restore them after the block to have a correct
* repcode history. If only one offset was invalid, it is easy. The tricky case is when both
* offsets were invalid. We need to figure out which offset to refill with.
* - If both offsets are zero they are in the same order.
* - If both offsets are non-zero, we won't restore the offsets from `offsetSaved[12]`.
* - If only one is zero, we need to decide which offset to restore.
* - If rep_offset1 is non-zero, then rep_offset2 must be offsetSaved1.
* - It is impossible for rep_offset2 to be non-zero.
*
* So if rep_offset1 started invalid (offsetSaved1 != 0) and became valid (rep_offset1 != 0), then
* set rep[0] = rep_offset1 and rep[1] = offsetSaved1.
*/
offsetSaved2 = ((offsetSaved1 != 0) && (rep_offset1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = rep_offset1 ? rep_offset1 : offsetSaved1;
rep[1] = rep_offset2 ? rep_offset2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_offset: /* Requires: ip0, idx */
/* Compute the offset code. */
match0 = base + idx;
rep_offset2 = rep_offset1;
rep_offset1 = (U32)(ip0-match0);
offcode = OFFSET_TO_OFFBASE(rep_offset1);
mLength = 4;
/* Count the backwards match length. */
while (((ip0>anchor) & (match0>prefixStart)) && (ip0[-1] == match0[-1])) {
ip0--;
match0--;
mLength++;
}
_match: /* Requires: ip0, match0, offcode */
/* Count the forward length. */
mLength += ZSTD_count(ip0 + mLength, match0 + mLength, iend);
ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength);
ip0 += mLength;
anchor = ip0;
/* Fill table and check for immediate repcode. */
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+current0+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
if (rep_offset2 > 0) { /* rep_offset2==0 means rep_offset2 is invalidated */
while ( (ip0 <= ilimit) && (MEM_read32(ip0) == MEM_read32(ip0 - rep_offset2)) ) {
/* store sequence */
size_t const rLength = ZSTD_count(ip0+4, ip0+4-rep_offset2, iend) + 4;
{ U32 const tmpOff = rep_offset2; rep_offset2 = rep_offset1; rep_offset1 = tmpOff; } /* swap rep_offset2 <=> rep_offset1 */
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
ip0 += rLength;
ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, REPCODE1_TO_OFFBASE, rLength);
anchor = ip0;
continue; /* faster when present (confirmed on gcc-8) ... (?) */
} } }
goto _start;
}
#define ZSTD_GEN_FAST_FN(dictMode, mls, step) \
static size_t ZSTD_compressBlock_fast_##dictMode##_##mls##_##step( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_fast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls, step); \
}
ZSTD_GEN_FAST_FN(noDict, 4, 1)
ZSTD_GEN_FAST_FN(noDict, 5, 1)
ZSTD_GEN_FAST_FN(noDict, 6, 1)
ZSTD_GEN_FAST_FN(noDict, 7, 1)
ZSTD_GEN_FAST_FN(noDict, 4, 0)
ZSTD_GEN_FAST_FN(noDict, 5, 0)
ZSTD_GEN_FAST_FN(noDict, 6, 0)
ZSTD_GEN_FAST_FN(noDict, 7, 0)
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState == NULL);
if (ms->cParams.targetLength > 1) {
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_noDict_4_1(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_noDict_5_1(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_noDict_6_1(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_noDict_7_1(ms, seqStore, rep, src, srcSize);
}
} else {
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_noDict_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_noDict_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_noDict_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_noDict_7_0(ms, seqStore, rep, src, srcSize);
}
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_fast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
U32 const stepSize = cParams->targetLength + !(cParams->targetLength);
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip0 = istart;
const BYTE* ip1 = ip0 + stepSize; /* we assert below that stepSize >= 1 */
const BYTE* anchor = istart;
const U32 prefixStartIndex = ms->window.dictLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams ;
const U32* const dictHashTable = dms->hashTable;
const U32 dictStartIndex = dms->window.dictLimit;
const BYTE* const dictBase = dms->window.base;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dms->window.nextSrc;
const U32 dictIndexDelta = prefixStartIndex - (U32)(dictEnd - dictBase);
const U32 dictAndPrefixLength = (U32)(istart - prefixStart + dictEnd - dictStart);
const U32 dictHBits = dictCParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
/* if a dictionary is still attached, it necessarily means that
* it is within window size. So we just check it. */
const U32 maxDistance = 1U << cParams->windowLog;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
assert(endIndex - prefixStartIndex <= maxDistance);
(void)maxDistance; (void)endIndex; /* these variables are not used when assert() is disabled */
(void)hasStep; /* not currently specialized on whether it's accelerated */
/* ensure there will be no underflow
* when translating a dict index into a local index */
assert(prefixStartIndex >= (U32)(dictEnd - dictBase));
if (ms->prefetchCDictTables) {
size_t const hashTableBytes = (((size_t)1) << dictCParams->hashLog) * sizeof(U32);
PREFETCH_AREA(dictHashTable, hashTableBytes)
}
/* init */
DEBUGLOG(5, "ZSTD_compressBlock_fast_dictMatchState_generic");
ip0 += (dictAndPrefixLength == 0);
/* dictMatchState repCode checks don't currently handle repCode == 0
* disabling. */
assert(offset_1 <= dictAndPrefixLength);
assert(offset_2 <= dictAndPrefixLength);
/* Outer search loop */
assert(stepSize >= 1);
while (ip1 <= ilimit) { /* repcode check at (ip0 + 1) is safe because ip0 < ip1 */
size_t mLength;
size_t hash0 = ZSTD_hashPtr(ip0, hlog, mls);
size_t const dictHashAndTag0 = ZSTD_hashPtr(ip0, dictHBits, mls);
U32 dictMatchIndexAndTag = dictHashTable[dictHashAndTag0 >> ZSTD_SHORT_CACHE_TAG_BITS];
int dictTagsMatch = ZSTD_comparePackedTags(dictMatchIndexAndTag, dictHashAndTag0);
U32 matchIndex = hashTable[hash0];
U32 curr = (U32)(ip0 - base);
size_t step = stepSize;
const size_t kStepIncr = 1 << kSearchStrength;
const BYTE* nextStep = ip0 + kStepIncr;
/* Inner search loop */
while (1) {
const BYTE* match = base + matchIndex;
const U32 repIndex = curr + 1 - offset_1;
const BYTE* repMatch = (repIndex < prefixStartIndex) ?
dictBase + (repIndex - dictIndexDelta) :
base + repIndex;
const size_t hash1 = ZSTD_hashPtr(ip1, hlog, mls);
size_t const dictHashAndTag1 = ZSTD_hashPtr(ip1, dictHBits, mls);
hashTable[hash0] = curr; /* update hash table */
if (((U32) ((prefixStartIndex - 1) - repIndex) >=
3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */
&& (MEM_read32(repMatch) == MEM_read32(ip0 + 1))) {
const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip0 + 1 + 4, repMatch + 4, iend, repMatchEnd, prefixStart) + 4;
ip0++;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
break;
}
if (dictTagsMatch) {
/* Found a possible dict match */
const U32 dictMatchIndex = dictMatchIndexAndTag >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatch = dictBase + dictMatchIndex;
if (dictMatchIndex > dictStartIndex &&
MEM_read32(dictMatch) == MEM_read32(ip0)) {
/* To replicate extDict parse behavior, we only use dict matches when the normal matchIndex is invalid */
if (matchIndex <= prefixStartIndex) {
U32 const offset = (U32) (curr - dictMatchIndex - dictIndexDelta);
mLength = ZSTD_count_2segments(ip0 + 4, dictMatch + 4, iend, dictEnd, prefixStart) + 4;
while (((ip0 > anchor) & (dictMatch > dictStart))
&& (ip0[-1] == dictMatch[-1])) {
ip0--;
dictMatch--;
mLength++;
} /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
break;
}
}
}
if (matchIndex > prefixStartIndex && MEM_read32(match) == MEM_read32(ip0)) {
/* found a regular match */
U32 const offset = (U32) (ip0 - match);
mLength = ZSTD_count(ip0 + 4, match + 4, iend) + 4;
while (((ip0 > anchor) & (match > prefixStart))
&& (ip0[-1] == match[-1])) {
ip0--;
match--;
mLength++;
} /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
break;
}
/* Prepare for next iteration */
dictMatchIndexAndTag = dictHashTable[dictHashAndTag1 >> ZSTD_SHORT_CACHE_TAG_BITS];
dictTagsMatch = ZSTD_comparePackedTags(dictMatchIndexAndTag, dictHashAndTag1);
matchIndex = hashTable[hash1];
if (ip1 >= nextStep) {
step++;
nextStep += kStepIncr;
}
ip0 = ip1;
ip1 = ip1 + step;
if (ip1 > ilimit) goto _cleanup;
curr = (U32)(ip0 - base);
hash0 = hash1;
} /* end inner search loop */
/* match found */
assert(mLength);
ip0 += mLength;
anchor = ip0;
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+curr+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; /* here because curr+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
/* check immediate repcode */
while (ip0 <= ilimit) {
U32 const current2 = (U32)(ip0-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixStartIndex ?
dictBase - dictIndexDelta + repIndex2 :
base + repIndex2;
if ( ((U32)((prefixStartIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
&& (MEM_read32(repMatch2) == MEM_read32(ip0))) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip0+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = current2;
ip0 += repLength2;
anchor = ip0;
continue;
}
break;
}
}
/* Prepare for next iteration */
assert(ip0 == anchor);
ip1 = ip0 + stepSize;
}
_cleanup:
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
ZSTD_GEN_FAST_FN(dictMatchState, 4, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 5, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 6, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 7, 0)
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState != NULL);
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_dictMatchState_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_dictMatchState_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_dictMatchState_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_dictMatchState_7_0(ms, seqStore, rep, src, srcSize);
}
}
static size_t ZSTD_compressBlock_fast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
size_t const stepSize = cParams->targetLength + !(cParams->targetLength) + 1;
const BYTE* const base = ms->window.base;
const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const istart = (const BYTE*)src;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
const U32 dictStartIndex = lowLimit;
const BYTE* const dictStart = dictBase + dictStartIndex;
const U32 dictLimit = ms->window.dictLimit;
const U32 prefixStartIndex = dictLimit < lowLimit ? lowLimit : dictLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const dictEnd = dictBase + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
const BYTE* ip0 = istart;
const BYTE* ip1;
const BYTE* ip2;
const BYTE* ip3;
U32 current0;
size_t hash0; /* hash for ip0 */
size_t hash1; /* hash for ip1 */
U32 idx; /* match idx for ip0 */
const BYTE* idxBase; /* base pointer for idx */
U32 offcode;
const BYTE* match0;
size_t mLength;
const BYTE* matchEnd = 0; /* initialize to avoid warning, assert != 0 later */
size_t step;
const BYTE* nextStep;
const size_t kStepIncr = (1 << (kSearchStrength - 1));
(void)hasStep; /* not currently specialized on whether it's accelerated */
DEBUGLOG(5, "ZSTD_compressBlock_fast_extDict_generic (offset_1=%u)", offset_1);
/* switch to "regular" variant if extDict is invalidated due to maxDistance */
if (prefixStartIndex == dictStartIndex)
return ZSTD_compressBlock_fast(ms, seqStore, rep, src, srcSize);
{ U32 const curr = (U32)(ip0 - base);
U32 const maxRep = curr - dictStartIndex;
if (offset_2 >= maxRep) offsetSaved2 = offset_2, offset_2 = 0;
if (offset_1 >= maxRep) offsetSaved1 = offset_1, offset_1 = 0;
}
/* start each op */
_start: /* Requires: ip0 */
step = stepSize;
nextStep = ip0 + kStepIncr;
/* calculate positions, ip0 - anchor == 0, so we skip step calc */
ip1 = ip0 + 1;
ip2 = ip0 + step;
ip3 = ip2 + 1;
if (ip3 >= ilimit) {
goto _cleanup;
}
hash0 = ZSTD_hashPtr(ip0, hlog, mls);
hash1 = ZSTD_hashPtr(ip1, hlog, mls);
idx = hashTable[hash0];
idxBase = idx < prefixStartIndex ? dictBase : base;
do {
{ /* load repcode match for ip[2] */
U32 const current2 = (U32)(ip2 - base);
U32 const repIndex = current2 - offset_1;
const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
U32 rval;
if ( ((U32)(prefixStartIndex - repIndex) >= 4) /* intentional underflow */
& (offset_1 > 0) ) {
rval = MEM_read32(repBase + repIndex);
} else {
rval = MEM_read32(ip2) ^ 1; /* guaranteed to not match. */
}
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* check repcode at ip[2] */
if (MEM_read32(ip2) == rval) {
ip0 = ip2;
match0 = repBase + repIndex;
matchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
assert((match0 != prefixStart) & (match0 != dictStart));
mLength = ip0[-1] == match0[-1];
ip0 -= mLength;
match0 -= mLength;
offcode = REPCODE1_TO_OFFBASE;
mLength += 4;
goto _match;
} }
{ /* load match for ip[0] */
U32 const mval = idx >= dictStartIndex ?
MEM_read32(idxBase + idx) :
MEM_read32(ip0) ^ 1; /* guaranteed not to match */
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
goto _offset;
} }
/* lookup ip[1] */
idx = hashTable[hash1];
idxBase = idx < prefixStartIndex ? dictBase : base;
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip3;
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
{ /* load match for ip[0] */
U32 const mval = idx >= dictStartIndex ?
MEM_read32(idxBase + idx) :
MEM_read32(ip0) ^ 1; /* guaranteed not to match */
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
goto _offset;
} }
/* lookup ip[1] */
idx = hashTable[hash1];
idxBase = idx < prefixStartIndex ? dictBase : base;
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip0 + step;
ip3 = ip1 + step;
/* calculate step */
if (ip2 >= nextStep) {
step++;
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
nextStep += kStepIncr;
}
} while (ip3 < ilimit);
_cleanup:
/* Note that there are probably still a couple positions we could search.
* However, it seems to be a meaningful performance hit to try to search
* them. So let's not. */
/* If offset_1 started invalid (offsetSaved1 != 0) and became valid (offset_1 != 0),
* rotate saved offsets. See comment in ZSTD_compressBlock_fast_noDict for more context. */
offsetSaved2 = ((offsetSaved1 != 0) && (offset_1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved1;
rep[1] = offset_2 ? offset_2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_offset: /* Requires: ip0, idx, idxBase */
/* Compute the offset code. */
{ U32 const offset = current0 - idx;
const BYTE* const lowMatchPtr = idx < prefixStartIndex ? dictStart : prefixStart;
matchEnd = idx < prefixStartIndex ? dictEnd : iend;
match0 = idxBase + idx;
offset_2 = offset_1;
offset_1 = offset;
offcode = OFFSET_TO_OFFBASE(offset);
mLength = 4;
/* Count the backwards match length. */
while (((ip0>anchor) & (match0>lowMatchPtr)) && (ip0[-1] == match0[-1])) {
ip0--;
match0--;
mLength++;
} }
_match: /* Requires: ip0, match0, offcode, matchEnd */
/* Count the forward length. */
assert(matchEnd != 0);
mLength += ZSTD_count_2segments(ip0 + mLength, match0 + mLength, iend, matchEnd, prefixStart);
ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength);
ip0 += mLength;
anchor = ip0;
/* write next hash table entry */
if (ip1 < ip0) {
hashTable[hash1] = (U32)(ip1 - base);
}
/* Fill table and check for immediate repcode. */
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+current0+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
while (ip0 <= ilimit) {
U32 const repIndex2 = (U32)(ip0-base) - offset_2;
const BYTE* const repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (offset_2 > 0)) /* intentional underflow */
&& (MEM_read32(repMatch2) == MEM_read32(ip0)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip0+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
{ U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; } /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
ip0 += repLength2;
anchor = ip0;
continue;
}
break;
} }
goto _start;
}
ZSTD_GEN_FAST_FN(extDict, 4, 0)
ZSTD_GEN_FAST_FN(extDict, 5, 0)
ZSTD_GEN_FAST_FN(extDict, 6, 0)
ZSTD_GEN_FAST_FN(extDict, 7, 0)
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState == NULL);
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_extDict_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_extDict_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_extDict_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_extDict_7_0(ms, seqStore, rep, src, srcSize);
}
}

38
ext/zstd/lib/compress/zstd_fast.h Normal file
View File

@@ -0,0 +1,38 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_FAST_H
#define ZSTD_FAST_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "../common/mem.h" /* U32 */
#include "zstd_compress_internal.h"
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_FAST_H */

2157
ext/zstd/lib/compress/zstd_lazy.c vendored Normal file
View File

File diff suppressed because it is too large Load Diff

127
ext/zstd/lib/compress/zstd_lazy.h Normal file
View File

@@ -0,0 +1,127 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_LAZY_H
#define ZSTD_LAZY_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_compress_internal.h"
/**
* Dedicated Dictionary Search Structure bucket log. In the
* ZSTD_dedicatedDictSearch mode, the hashTable has
* 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just
* one.
*/
#define ZSTD_LAZY_DDSS_BUCKET_LOG 2
#define ZSTD_ROW_HASH_TAG_BITS 8 /* nb bits to use for the tag */
U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip);
void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip);
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip);
void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_LAZY_H */

724
ext/zstd/lib/compress/zstd_ldm.c vendored Normal file
View File

@@ -0,0 +1,724 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_ldm.h"
#include "../common/debug.h"
#include "../common/xxhash.h"
#include "zstd_fast.h" /* ZSTD_fillHashTable() */
#include "zstd_double_fast.h" /* ZSTD_fillDoubleHashTable() */
#include "zstd_ldm_geartab.h"
#define LDM_BUCKET_SIZE_LOG 3
#define LDM_MIN_MATCH_LENGTH 64
#define LDM_HASH_RLOG 7
typedef struct {
U64 rolling;
U64 stopMask;
} ldmRollingHashState_t;
/** ZSTD_ldm_gear_init():
*
* Initializes the rolling hash state such that it will honor the
* settings in params. */
static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params)
{
unsigned maxBitsInMask = MIN(params->minMatchLength, 64);
unsigned hashRateLog = params->hashRateLog;
state->rolling = ~(U32)0;
/* The choice of the splitting criterion is subject to two conditions:
* 1. it has to trigger on average every 2^(hashRateLog) bytes;
* 2. ideally, it has to depend on a window of minMatchLength bytes.
*
* In the gear hash algorithm, bit n depends on the last n bytes;
* so in order to obtain a good quality splitting criterion it is
* preferable to use bits with high weight.
*
* To match condition 1 we use a mask with hashRateLog bits set
* and, because of the previous remark, we make sure these bits
* have the highest possible weight while still respecting
* condition 2.
*/
if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) {
state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog);
} else {
/* In this degenerate case we simply honor the hash rate. */
state->stopMask = ((U64)1 << hashRateLog) - 1;
}
}
/** ZSTD_ldm_gear_reset()
* Feeds [data, data + minMatchLength) into the hash without registering any
* splits. This effectively resets the hash state. This is used when skipping
* over data, either at the beginning of a block, or skipping sections.
*/
static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state,
BYTE const* data, size_t minMatchLength)
{
U64 hash = state->rolling;
size_t n = 0;
#define GEAR_ITER_ONCE() do { \
hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
n += 1; \
} while (0)
while (n + 3 < minMatchLength) {
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
}
while (n < minMatchLength) {
GEAR_ITER_ONCE();
}
#undef GEAR_ITER_ONCE
}
/** ZSTD_ldm_gear_feed():
*
* Registers in the splits array all the split points found in the first
* size bytes following the data pointer. This function terminates when
* either all the data has been processed or LDM_BATCH_SIZE splits are
* present in the splits array.
*
* Precondition: The splits array must not be full.
* Returns: The number of bytes processed. */
static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state,
BYTE const* data, size_t size,
size_t* splits, unsigned* numSplits)
{
size_t n;
U64 hash, mask;
hash = state->rolling;
mask = state->stopMask;
n = 0;
#define GEAR_ITER_ONCE() do { \
hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
n += 1; \
if (UNLIKELY((hash & mask) == 0)) { \
splits[*numSplits] = n; \
*numSplits += 1; \
if (*numSplits == LDM_BATCH_SIZE) \
goto done; \
} \
} while (0)
while (n + 3 < size) {
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
}
while (n < size) {
GEAR_ITER_ONCE();
}
#undef GEAR_ITER_ONCE
done:
state->rolling = hash;
return n;
}
void ZSTD_ldm_adjustParameters(ldmParams_t* params,
ZSTD_compressionParameters const* cParams)
{
params->windowLog = cParams->windowLog;
ZSTD_STATIC_ASSERT(LDM_BUCKET_SIZE_LOG <= ZSTD_LDM_BUCKETSIZELOG_MAX);
DEBUGLOG(4, "ZSTD_ldm_adjustParameters");
if (!params->bucketSizeLog) params->bucketSizeLog = LDM_BUCKET_SIZE_LOG;
if (!params->minMatchLength) params->minMatchLength = LDM_MIN_MATCH_LENGTH;
if (params->hashLog == 0) {
params->hashLog = MAX(ZSTD_HASHLOG_MIN, params->windowLog - LDM_HASH_RLOG);
assert(params->hashLog <= ZSTD_HASHLOG_MAX);
}
if (params->hashRateLog == 0) {
params->hashRateLog = params->windowLog < params->hashLog
? 0
: params->windowLog - params->hashLog;
}
params->bucketSizeLog = MIN(params->bucketSizeLog, params->hashLog);
}
size_t ZSTD_ldm_getTableSize(ldmParams_t params)
{
size_t const ldmHSize = ((size_t)1) << params.hashLog;
size_t const ldmBucketSizeLog = MIN(params.bucketSizeLog, params.hashLog);
size_t const ldmBucketSize = ((size_t)1) << (params.hashLog - ldmBucketSizeLog);
size_t const totalSize = ZSTD_cwksp_alloc_size(ldmBucketSize)
+ ZSTD_cwksp_alloc_size(ldmHSize * sizeof(ldmEntry_t));
return params.enableLdm == ZSTD_ps_enable ? totalSize : 0;
}
size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize)
{
return params.enableLdm == ZSTD_ps_enable ? (maxChunkSize / params.minMatchLength) : 0;
}
/** ZSTD_ldm_getBucket() :
* Returns a pointer to the start of the bucket associated with hash. */
static ldmEntry_t* ZSTD_ldm_getBucket(
ldmState_t* ldmState, size_t hash, ldmParams_t const ldmParams)
{
return ldmState->hashTable + (hash << ldmParams.bucketSizeLog);
}
/** ZSTD_ldm_insertEntry() :
* Insert the entry with corresponding hash into the hash table */
static void ZSTD_ldm_insertEntry(ldmState_t* ldmState,
size_t const hash, const ldmEntry_t entry,
ldmParams_t const ldmParams)
{
BYTE* const pOffset = ldmState->bucketOffsets + hash;
unsigned const offset = *pOffset;
*(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry;
*pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1));
}
/** ZSTD_ldm_countBackwardsMatch() :
* Returns the number of bytes that match backwards before pIn and pMatch.
*
* We count only bytes where pMatch >= pBase and pIn >= pAnchor. */
static size_t ZSTD_ldm_countBackwardsMatch(
const BYTE* pIn, const BYTE* pAnchor,
const BYTE* pMatch, const BYTE* pMatchBase)
{
size_t matchLength = 0;
while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) {
pIn--;
pMatch--;
matchLength++;
}
return matchLength;
}
/** ZSTD_ldm_countBackwardsMatch_2segments() :
* Returns the number of bytes that match backwards from pMatch,
* even with the backwards match spanning 2 different segments.
*
* On reaching `pMatchBase`, start counting from mEnd */
static size_t ZSTD_ldm_countBackwardsMatch_2segments(
const BYTE* pIn, const BYTE* pAnchor,
const BYTE* pMatch, const BYTE* pMatchBase,
const BYTE* pExtDictStart, const BYTE* pExtDictEnd)
{
size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase);
if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) {
/* If backwards match is entirely in the extDict or prefix, immediately return */
return matchLength;
}
DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength);
matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart);
DEBUGLOG(7, "final backwards match length = %zu", matchLength);
return matchLength;
}
/** ZSTD_ldm_fillFastTables() :
*
* Fills the relevant tables for the ZSTD_fast and ZSTD_dfast strategies.
* This is similar to ZSTD_loadDictionaryContent.
*
* The tables for the other strategies are filled within their
* block compressors. */
static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
void const* end)
{
const BYTE* const iend = (const BYTE*)end;
switch(ms->cParams.strategy)
{
case ZSTD_fast:
ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_dfast:
ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_greedy:
case ZSTD_lazy:
case ZSTD_lazy2:
case ZSTD_btlazy2:
case ZSTD_btopt:
case ZSTD_btultra:
case ZSTD_btultra2:
break;
default:
assert(0); /* not possible : not a valid strategy id */
}
return 0;
}
void ZSTD_ldm_fillHashTable(
ldmState_t* ldmState, const BYTE* ip,
const BYTE* iend, ldmParams_t const* params)
{
U32 const minMatchLength = params->minMatchLength;
U32 const hBits = params->hashLog - params->bucketSizeLog;
BYTE const* const base = ldmState->window.base;
BYTE const* const istart = ip;
ldmRollingHashState_t hashState;
size_t* const splits = ldmState->splitIndices;
unsigned numSplits;
DEBUGLOG(5, "ZSTD_ldm_fillHashTable");
ZSTD_ldm_gear_init(&hashState, params);
while (ip < iend) {
size_t hashed;
unsigned n;
numSplits = 0;
hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits);
for (n = 0; n < numSplits; n++) {
if (ip + splits[n] >= istart + minMatchLength) {
BYTE const* const split = ip + splits[n] - minMatchLength;
U64 const xxhash = XXH64(split, minMatchLength, 0);
U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
ldmEntry_t entry;
entry.offset = (U32)(split - base);
entry.checksum = (U32)(xxhash >> 32);
ZSTD_ldm_insertEntry(ldmState, hash, entry, *params);
}
}
ip += hashed;
}
}
/** ZSTD_ldm_limitTableUpdate() :
*
* Sets cctx->nextToUpdate to a position corresponding closer to anchor
* if it is far way
* (after a long match, only update tables a limited amount). */
static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor)
{
U32 const curr = (U32)(anchor - ms->window.base);
if (curr > ms->nextToUpdate + 1024) {
ms->nextToUpdate =
curr - MIN(512, curr - ms->nextToUpdate - 1024);
}
}
static size_t ZSTD_ldm_generateSequences_internal(
ldmState_t* ldmState, rawSeqStore_t* rawSeqStore,
ldmParams_t const* params, void const* src, size_t srcSize)
{
/* LDM parameters */
int const extDict = ZSTD_window_hasExtDict(ldmState->window);
U32 const minMatchLength = params->minMatchLength;
U32 const entsPerBucket = 1U << params->bucketSizeLog;
U32 const hBits = params->hashLog - params->bucketSizeLog;
/* Prefix and extDict parameters */
U32 const dictLimit = ldmState->window.dictLimit;
U32 const lowestIndex = extDict ? ldmState->window.lowLimit : dictLimit;
BYTE const* const base = ldmState->window.base;
BYTE const* const dictBase = extDict ? ldmState->window.dictBase : NULL;
BYTE const* const dictStart = extDict ? dictBase + lowestIndex : NULL;
BYTE const* const dictEnd = extDict ? dictBase + dictLimit : NULL;
BYTE const* const lowPrefixPtr = base + dictLimit;
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
BYTE const* const ilimit = iend - HASH_READ_SIZE;
/* Input positions */
BYTE const* anchor = istart;
BYTE const* ip = istart;
/* Rolling hash state */
ldmRollingHashState_t hashState;
/* Arrays for staged-processing */
size_t* const splits = ldmState->splitIndices;
ldmMatchCandidate_t* const candidates = ldmState->matchCandidates;
unsigned numSplits;
if (srcSize < minMatchLength)
return iend - anchor;
/* Initialize the rolling hash state with the first minMatchLength bytes */
ZSTD_ldm_gear_init(&hashState, params);
ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength);
ip += minMatchLength;
while (ip < ilimit) {
size_t hashed;
unsigned n;
numSplits = 0;
hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip,
splits, &numSplits);
for (n = 0; n < numSplits; n++) {
BYTE const* const split = ip + splits[n] - minMatchLength;
U64 const xxhash = XXH64(split, minMatchLength, 0);
U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
candidates[n].split = split;
candidates[n].hash = hash;
candidates[n].checksum = (U32)(xxhash >> 32);
candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params);
PREFETCH_L1(candidates[n].bucket);
}
for (n = 0; n < numSplits; n++) {
size_t forwardMatchLength = 0, backwardMatchLength = 0,
bestMatchLength = 0, mLength;
U32 offset;
BYTE const* const split = candidates[n].split;
U32 const checksum = candidates[n].checksum;
U32 const hash = candidates[n].hash;
ldmEntry_t* const bucket = candidates[n].bucket;
ldmEntry_t const* cur;
ldmEntry_t const* bestEntry = NULL;
ldmEntry_t newEntry;
newEntry.offset = (U32)(split - base);
newEntry.checksum = checksum;
/* If a split point would generate a sequence overlapping with
* the previous one, we merely register it in the hash table and
* move on */
if (split < anchor) {
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
continue;
}
for (cur = bucket; cur < bucket + entsPerBucket; cur++) {
size_t curForwardMatchLength, curBackwardMatchLength,
curTotalMatchLength;
if (cur->checksum != checksum || cur->offset <= lowestIndex) {
continue;
}
if (extDict) {
BYTE const* const curMatchBase =
cur->offset < dictLimit ? dictBase : base;
BYTE const* const pMatch = curMatchBase + cur->offset;
BYTE const* const matchEnd =
cur->offset < dictLimit ? dictEnd : iend;
BYTE const* const lowMatchPtr =
cur->offset < dictLimit ? dictStart : lowPrefixPtr;
curForwardMatchLength =
ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr);
if (curForwardMatchLength < minMatchLength) {
continue;
}
curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments(
split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd);
} else { /* !extDict */
BYTE const* const pMatch = base + cur->offset;
curForwardMatchLength = ZSTD_count(split, pMatch, iend);
if (curForwardMatchLength < minMatchLength) {
continue;
}
curBackwardMatchLength =
ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr);
}
curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength;
if (curTotalMatchLength > bestMatchLength) {
bestMatchLength = curTotalMatchLength;
forwardMatchLength = curForwardMatchLength;
backwardMatchLength = curBackwardMatchLength;
bestEntry = cur;
}
}
/* No match found -- insert an entry into the hash table
* and process the next candidate match */
if (bestEntry == NULL) {
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
continue;
}
/* Match found */
offset = (U32)(split - base) - bestEntry->offset;
mLength = forwardMatchLength + backwardMatchLength;
{
rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size;
/* Out of sequence storage */
if (rawSeqStore->size == rawSeqStore->capacity)
return ERROR(dstSize_tooSmall);
seq->litLength = (U32)(split - backwardMatchLength - anchor);
seq->matchLength = (U32)mLength;
seq->offset = offset;
rawSeqStore->size++;
}
/* Insert the current entry into the hash table --- it must be
* done after the previous block to avoid clobbering bestEntry */
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
anchor = split + forwardMatchLength;
/* If we find a match that ends after the data that we've hashed
* then we have a repeating, overlapping, pattern. E.g. all zeros.
* If one repetition of the pattern matches our `stopMask` then all
* repetitions will. We don't need to insert them all into out table,
* only the first one. So skip over overlapping matches.
* This is a major speed boost (20x) for compressing a single byte
* repeated, when that byte ends up in the table.
*/
if (anchor > ip + hashed) {
ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength);
/* Continue the outer loop at anchor (ip + hashed == anchor). */
ip = anchor - hashed;
break;
}
}
ip += hashed;
}
return iend - anchor;
}
/*! ZSTD_ldm_reduceTable() :
* reduce table indexes by `reducerValue` */
static void ZSTD_ldm_reduceTable(ldmEntry_t* const table, U32 const size,
U32 const reducerValue)
{
U32 u;
for (u = 0; u < size; u++) {
if (table[u].offset < reducerValue) table[u].offset = 0;
else table[u].offset -= reducerValue;
}
}
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldmState, rawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize)
{
U32 const maxDist = 1U << params->windowLog;
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
size_t const kMaxChunkSize = 1 << 20;
size_t const nbChunks = (srcSize / kMaxChunkSize) + ((srcSize % kMaxChunkSize) != 0);
size_t chunk;
size_t leftoverSize = 0;
assert(ZSTD_CHUNKSIZE_MAX >= kMaxChunkSize);
/* Check that ZSTD_window_update() has been called for this chunk prior
* to passing it to this function.
*/
assert(ldmState->window.nextSrc >= (BYTE const*)src + srcSize);
/* The input could be very large (in zstdmt), so it must be broken up into
* chunks to enforce the maximum distance and handle overflow correction.
*/
assert(sequences->pos <= sequences->size);
assert(sequences->size <= sequences->capacity);
for (chunk = 0; chunk < nbChunks && sequences->size < sequences->capacity; ++chunk) {
BYTE const* const chunkStart = istart + chunk * kMaxChunkSize;
size_t const remaining = (size_t)(iend - chunkStart);
BYTE const *const chunkEnd =
(remaining < kMaxChunkSize) ? iend : chunkStart + kMaxChunkSize;
size_t const chunkSize = chunkEnd - chunkStart;
size_t newLeftoverSize;
size_t const prevSize = sequences->size;
assert(chunkStart < iend);
/* 1. Perform overflow correction if necessary. */
if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) {
U32 const ldmHSize = 1U << params->hashLog;
U32 const correction = ZSTD_window_correctOverflow(
&ldmState->window, /* cycleLog */ 0, maxDist, chunkStart);
ZSTD_ldm_reduceTable(ldmState->hashTable, ldmHSize, correction);
/* invalidate dictionaries on overflow correction */
ldmState->loadedDictEnd = 0;
}
/* 2. We enforce the maximum offset allowed.
*
* kMaxChunkSize should be small enough that we don't lose too much of
* the window through early invalidation.
* TODO: * Test the chunk size.
* * Try invalidation after the sequence generation and test the
* offset against maxDist directly.
*
* NOTE: Because of dictionaries + sequence splitting we MUST make sure
* that any offset used is valid at the END of the sequence, since it may
* be split into two sequences. This condition holds when using
* ZSTD_window_enforceMaxDist(), but if we move to checking offsets
* against maxDist directly, we'll have to carefully handle that case.
*/
ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, &ldmState->loadedDictEnd, NULL);
/* 3. Generate the sequences for the chunk, and get newLeftoverSize. */
newLeftoverSize = ZSTD_ldm_generateSequences_internal(
ldmState, sequences, params, chunkStart, chunkSize);
if (ZSTD_isError(newLeftoverSize))
return newLeftoverSize;
/* 4. We add the leftover literals from previous iterations to the first
* newly generated sequence, or add the `newLeftoverSize` if none are
* generated.
*/
/* Prepend the leftover literals from the last call */
if (prevSize < sequences->size) {
sequences->seq[prevSize].litLength += (U32)leftoverSize;
leftoverSize = newLeftoverSize;
} else {
assert(newLeftoverSize == chunkSize);
leftoverSize += chunkSize;
}
}
return 0;
}
void
ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch)
{
while (srcSize > 0 && rawSeqStore->pos < rawSeqStore->size) {
rawSeq* seq = rawSeqStore->seq + rawSeqStore->pos;
if (srcSize <= seq->litLength) {
/* Skip past srcSize literals */
seq->litLength -= (U32)srcSize;
return;
}
srcSize -= seq->litLength;
seq->litLength = 0;
if (srcSize < seq->matchLength) {
/* Skip past the first srcSize of the match */
seq->matchLength -= (U32)srcSize;
if (seq->matchLength < minMatch) {
/* The match is too short, omit it */
if (rawSeqStore->pos + 1 < rawSeqStore->size) {
seq[1].litLength += seq[0].matchLength;
}
rawSeqStore->pos++;
}
return;
}
srcSize -= seq->matchLength;
seq->matchLength = 0;
rawSeqStore->pos++;
}
}
/**
* If the sequence length is longer than remaining then the sequence is split
* between this block and the next.
*
* Returns the current sequence to handle, or if the rest of the block should
* be literals, it returns a sequence with offset == 0.
*/
static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore,
U32 const remaining, U32 const minMatch)
{
rawSeq sequence = rawSeqStore->seq[rawSeqStore->pos];
assert(sequence.offset > 0);
/* Likely: No partial sequence */
if (remaining >= sequence.litLength + sequence.matchLength) {
rawSeqStore->pos++;
return sequence;
}
/* Cut the sequence short (offset == 0 ==> rest is literals). */
if (remaining <= sequence.litLength) {
sequence.offset = 0;
} else if (remaining < sequence.litLength + sequence.matchLength) {
sequence.matchLength = remaining - sequence.litLength;
if (sequence.matchLength < minMatch) {
sequence.offset = 0;
}
}
/* Skip past `remaining` bytes for the future sequences. */
ZSTD_ldm_skipSequences(rawSeqStore, remaining, minMatch);
return sequence;
}
void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes);
while (currPos && rawSeqStore->pos < rawSeqStore->size) {
rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos];
if (currPos >= currSeq.litLength + currSeq.matchLength) {
currPos -= currSeq.litLength + currSeq.matchLength;
rawSeqStore->pos++;
} else {
rawSeqStore->posInSequence = currPos;
break;
}
}
if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) {
rawSeqStore->posInSequence = 0;
}
}
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_paramSwitch_e useRowMatchFinder,
void const* src, size_t srcSize)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
unsigned const minMatch = cParams->minMatch;
ZSTD_blockCompressor const blockCompressor =
ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms));
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
/* Input positions */
BYTE const* ip = istart;
DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize);
/* If using opt parser, use LDMs only as candidates rather than always accepting them */
if (cParams->strategy >= ZSTD_btopt) {
size_t lastLLSize;
ms->ldmSeqStore = rawSeqStore;
lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize);
ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize);
return lastLLSize;
}
assert(rawSeqStore->pos <= rawSeqStore->size);
assert(rawSeqStore->size <= rawSeqStore->capacity);
/* Loop through each sequence and apply the block compressor to the literals */
while (rawSeqStore->pos < rawSeqStore->size && ip < iend) {
/* maybeSplitSequence updates rawSeqStore->pos */
rawSeq const sequence = maybeSplitSequence(rawSeqStore,
(U32)(iend - ip), minMatch);
int i;
/* End signal */
if (sequence.offset == 0)
break;
assert(ip + sequence.litLength + sequence.matchLength <= iend);
/* Fill tables for block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
ZSTD_ldm_fillFastTables(ms, ip);
/* Run the block compressor */
DEBUGLOG(5, "pos %u : calling block compressor on segment of size %u", (unsigned)(ip-istart), sequence.litLength);
{
size_t const newLitLength =
blockCompressor(ms, seqStore, rep, ip, sequence.litLength);
ip += sequence.litLength;
/* Update the repcodes */
for (i = ZSTD_REP_NUM - 1; i > 0; i--)
rep[i] = rep[i-1];
rep[0] = sequence.offset;
/* Store the sequence */
ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend,
OFFSET_TO_OFFBASE(sequence.offset),
sequence.matchLength);
ip += sequence.matchLength;
}
}
/* Fill the tables for the block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
ZSTD_ldm_fillFastTables(ms, ip);
/* Compress the last literals */
return blockCompressor(ms, seqStore, rep, ip, iend - ip);
}

Some files were not shown because too many files have changed in this diff Show More