zstd: vendor python-zstandard 0.7.0 Commit 3054ae3a66112970a091d3939fee32c2d0c1a23e from https://github.com/indygreg/python-zstandard is imported without modifications (other than removing unwanted files). The vendored zstd library within has been upgraded from 1.1.2 to 1.1.3. This version introduced new APIs for threads, thread pools, multi-threaded compression, and a new dictionary builder (COVER). These features are not yet used by python-zstandard (or Mercurial for that matter). However, that will likely change in the next python-zstandard release (and I think there are opportunities for Mercurial to take advantage of the multi-threaded APIs). Relevant to Mercurial, the CFFI bindings are now fully implemented. This means zstd should "just work" with PyPy (although I haven't tried). The python-zstandard test suite also runs all tests against both the C extension and CFFI bindings to ensure feature parity. There is also a "decompress_content_dict_chain()" API. This was derived from discussions with Yann Collet on list about alternate ways of encoding delta chains. The change most relevant to Mercurial is a performance enhancement in the simple decompression API to reuse a data structure across operations. This makes decompression of multiple inputs significantly faster. (This scenario occurs when reading revlog delta chains, for example.) Using python-zstandard's bench.py to measure the performance difference... On changelog chunks in the mozilla-unified repo: decompress discrete decompress() reuse zctx 1.262243 wall; 1.260000 CPU; 1.260000 user; 0.000000 sys 170.43 MB/s (best of 3) 0.949106 wall; 0.950000 CPU; 0.950000 user; 0.000000 sys 226.66 MB/s (best of 4) decompress discrete dict decompress() reuse zctx 0.692170 wall; 0.690000 CPU; 0.690000 user; 0.000000 sys 310.80 MB/s (best of 5) 0.437088 wall; 0.440000 CPU; 0.440000 user; 0.000000 sys 492.17 MB/s (best of 7) On manifest chunks in the mozilla-unified repo: decompress discrete decompress() reuse zctx 1.367284 wall; 1.370000 CPU; 1.370000 user; 0.000000 sys 274.01 MB/s (best of 3) 1.086831 wall; 1.080000 CPU; 1.080000 user; 0.000000 sys 344.72 MB/s (best of 3) decompress discrete dict decompress() reuse zctx 0.993272 wall; 0.990000 CPU; 0.990000 user; 0.000000 sys 377.19 MB/s (best of 3) 0.678651 wall; 0.680000 CPU; 0.680000 user; 0.000000 sys 552.06 MB/s (best of 5) That should make reads on zstd revlogs a bit faster ;) # no-check-commit

     1

# Copyright (c) 2016-present, Gregory Szorc

     2

# All rights reserved.

     3

#

     4

# This software may be modified and distributed under the terms

     5

# of the BSD license. See the LICENSE file for details.

     6

     7

from __future__ import absolute_import

     8

     9

import cffi

    10

import distutils.ccompiler

    11

import os

    12

import re

    13

import subprocess

    14

import tempfile

    15

    16

    17

HERE = os.path.abspath(os.path.dirname(__file__))

    18

    19

SOURCES = ['zstd/%s' % p for p in (

    20

    'common/entropy_common.c',

    21

    'common/error_private.c',

    22

    'common/fse_decompress.c',

    23

    'common/pool.c',

    24

    'common/threading.c',

    25

    'common/xxhash.c',

    26

    'common/zstd_common.c',

    27

    'compress/fse_compress.c',

    28

    'compress/huf_compress.c',

    29

    'compress/zstd_compress.c',

    30

    'decompress/huf_decompress.c',

    31

    'decompress/zstd_decompress.c',

    32

    'dictBuilder/cover.c',

    33

    'dictBuilder/divsufsort.c',

    34

    'dictBuilder/zdict.c',

    35

)]

    36

    37

HEADERS = [os.path.join(HERE, 'zstd', *p) for p in (

    38

    ('zstd.h',),

    39

    ('common', 'pool.h'),

    40

    ('dictBuilder', 'zdict.h'),

    41

)]

    42

    43

INCLUDE_DIRS = [os.path.join(HERE, d) for d in (

    44

    'zstd',

    45

    'zstd/common',

    46

    'zstd/compress',

    47

    'zstd/decompress',

    48

    'zstd/dictBuilder',

    49

)]

    50

    51

# cffi can't parse some of the primitives in zstd.h. So we invoke the

    52

# preprocessor and feed its output into cffi.

    53

compiler = distutils.ccompiler.new_compiler()

    54

    55

# Needed for MSVC.

    56

if hasattr(compiler, 'initialize'):

    57

    compiler.initialize()

    58

    59

# Distutils doesn't set compiler.preprocessor, so invoke the preprocessor

    60

# manually.

    61

if compiler.compiler_type == 'unix':

    62

    args = list(compiler.executables['compiler'])

    63

    args.extend([

    64

        '-E',

    65

        '-DZSTD_STATIC_LINKING_ONLY',

    66

        '-DZDICT_STATIC_LINKING_ONLY',

    67

    ])

    68

elif compiler.compiler_type == 'msvc':

    69

    args = [compiler.cc]

    70

    args.extend([

    71

        '/EP',

    72

        '/DZSTD_STATIC_LINKING_ONLY',

    73

        '/DZDICT_STATIC_LINKING_ONLY',

    74

    ])

    75

else:

    76

    raise Exception('unsupported compiler type: %s' % compiler.compiler_type)

    77

    78

def preprocess(path):

    79

    # zstd.h includes <stddef.h>, which is also included by cffi's boilerplate.

    80

    # This can lead to duplicate declarations. So we strip this include from the

    81

    # preprocessor invocation.

    82

    with open(path, 'rb') as fh:

    83

        lines = [l for l in fh if not l.startswith(b'#include <stddef.h>')]

    84

    85

    fd, input_file = tempfile.mkstemp(suffix='.h')

    86

    os.write(fd, b''.join(lines))

    87

    os.close(fd)

    88

    89

    try:

    90

        process = subprocess.Popen(args + [input_file], stdout=subprocess.PIPE)

    91

        output = process.communicate()[0]

    92

        ret = process.poll()

    93

        if ret:

    94

            raise Exception('preprocessor exited with error')

    95

    96

        return output

    97

    finally:

    98

        os.unlink(input_file)

    99

   100

   101

def normalize_output(output):

   102

    lines = []

   103

    for line in output.splitlines():

   104

        # CFFI's parser doesn't like __attribute__ on UNIX compilers.

   105

        if line.startswith(b'__attribute__ ((visibility ("default"))) '):

   106

            line = line[len(b'__attribute__ ((visibility ("default"))) '):]

   107

   108

        if line.startswith(b'__attribute__((deprecated('):

   109

            continue

   110

        elif b'__declspec(deprecated(' in line:

   111

            continue

   112

   113

        lines.append(line)

   114

   115

    return b'\n'.join(lines)

   116

   117

   118

ffi = cffi.FFI()

   119

ffi.set_source('_zstd_cffi', '''

   120

#include "mem.h"

   121

#define ZSTD_STATIC_LINKING_ONLY

   122

#include "zstd.h"

   123

#define ZDICT_STATIC_LINKING_ONLY

   124

#include "pool.h"

   125

#include "zdict.h"

   126

''', sources=SOURCES, include_dirs=INCLUDE_DIRS)

   127

   128

DEFINE = re.compile(b'^\\#define ([a-zA-Z0-9_]+) ')

   129

   130

sources = []

   131

   132

for header in HEADERS:

   133

    preprocessed = preprocess(header)

   134

    sources.append(normalize_output(preprocessed))

   135

   136

    # Do another pass over source and find constants that were preprocessed

   137

    # away.

   138

    with open(header, 'rb') as fh:

   139

        for line in fh:

   140

            line = line.strip()

   141

            m = DEFINE.match(line)

   142

            if not m:

   143

                continue

   144

   145

            # The parser doesn't like some constants with complex values.

   146

            if m.group(1) in (b'ZSTD_LIB_VERSION', b'ZSTD_VERSION_STRING'):

   147

                continue

   148

   149

            sources.append(m.group(0) + b' ...')

   150

   151

ffi.cdef(u'\n'.join(s.decode('latin1') for s in sources))

   152

   153

if __name__ == '__main__':

   154

    ffi.compile()