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 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 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353
|
/**
* Copyright (c) 2016-present, Gregory Szorc
* All rights reserved.
*
* This software may be modified and distributed under the terms
* of the BSD license. See the LICENSE file for details.
*/
/* A Python C extension for Zstandard. */
#if defined(_WIN32)
#define WIN32_LEAN_AND_MEAN
#include <Windows.h>
#elif defined(__APPLE__) || defined(__OpenBSD__) || defined(__FreeBSD__) || \
defined(__NetBSD__) || defined(__DragonFly__)
#include <sys/types.h>
#include <sys/sysctl.h>
#endif
#include "python-zstandard.h"
#include "bufferutil.c"
#include "compressionchunker.c"
#include "compressiondict.c"
#include "compressionparams.c"
#include "compressionreader.c"
#include "compressionwriter.c"
#include "compressobj.c"
#include "compressor.c"
#include "compressoriterator.c"
#include "constants.c"
#include "decompressionreader.c"
#include "decompressionwriter.c"
#include "decompressobj.c"
#include "decompressor.c"
#include "decompressoriterator.c"
#include "frameparams.c"
PyObject *ZstdError;
static PyObject *estimate_decompression_context_size(PyObject *self) {
return PyLong_FromSize_t(ZSTD_estimateDCtxSize());
}
static PyObject *frame_content_size(PyObject *self, PyObject *args,
PyObject *kwargs) {
static char *kwlist[] = {"source", NULL};
Py_buffer source;
PyObject *result = NULL;
unsigned long long size;
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "y*:frame_content_size",
kwlist, &source)) {
return NULL;
}
size = ZSTD_getFrameContentSize(source.buf, source.len);
if (size == ZSTD_CONTENTSIZE_ERROR) {
PyErr_SetString(ZstdError, "error when determining content size");
}
else if (size == ZSTD_CONTENTSIZE_UNKNOWN) {
result = PyLong_FromLong(-1);
}
else {
result = PyLong_FromUnsignedLongLong(size);
}
PyBuffer_Release(&source);
return result;
}
static PyObject *frame_header_size(PyObject *self, PyObject *args,
PyObject *kwargs) {
static char *kwlist[] = {"source", NULL};
Py_buffer source;
PyObject *result = NULL;
size_t zresult;
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "y*:frame_header_size",
kwlist, &source)) {
return NULL;
}
zresult = ZSTD_frameHeaderSize(source.buf, source.len);
if (ZSTD_isError(zresult)) {
PyErr_Format(ZstdError, "could not determine frame header size: %s",
ZSTD_getErrorName(zresult));
}
else {
result = PyLong_FromSize_t(zresult);
}
PyBuffer_Release(&source);
return result;
}
static char zstd_doc[] = "Interface to zstandard";
static PyMethodDef zstd_methods[] = {
{"estimate_decompression_context_size",
(PyCFunction)estimate_decompression_context_size, METH_NOARGS, NULL},
{"frame_content_size", (PyCFunction)frame_content_size,
METH_VARARGS | METH_KEYWORDS, NULL},
{"frame_header_size", (PyCFunction)frame_header_size,
METH_VARARGS | METH_KEYWORDS, NULL},
{"get_frame_parameters", (PyCFunction)get_frame_parameters,
METH_VARARGS | METH_KEYWORDS, NULL},
{"train_dictionary", (PyCFunction)train_dictionary,
METH_VARARGS | METH_KEYWORDS, NULL},
{NULL, NULL}};
void bufferutil_module_init(PyObject *mod);
void compressobj_module_init(PyObject *mod);
void compressor_module_init(PyObject *mod);
void compressionparams_module_init(PyObject *mod);
void constants_module_init(PyObject *mod);
void compressionchunker_module_init(PyObject *mod);
void compressiondict_module_init(PyObject *mod);
void compressionreader_module_init(PyObject *mod);
void compressionwriter_module_init(PyObject *mod);
void compressoriterator_module_init(PyObject *mod);
void decompressor_module_init(PyObject *mod);
void decompressobj_module_init(PyObject *mod);
void decompressionreader_module_init(PyObject *mod);
void decompressionwriter_module_init(PyObject *mod);
void decompressoriterator_module_init(PyObject *mod);
void frameparams_module_init(PyObject *mod);
void zstd_module_init(PyObject *m) {
/* python-zstandard relies on unstable zstd C API features. This means
that changes in zstd may break expectations in python-zstandard.
python-zstandard is distributed with a copy of the zstd sources.
python-zstandard is only guaranteed to work with the bundled version
of zstd.
However, downstream redistributors or packagers may unbundle zstd
from python-zstandard. This can result in a mismatch between zstd
versions and API semantics. This essentially "voids the warranty"
of python-zstandard and may cause undefined behavior.
We detect this mismatch here and refuse to load the module if this
scenario is detected.
*/
PyObject *features = NULL;
PyObject *feature = NULL;
unsigned zstd_ver_no = ZSTD_versionNumber();
unsigned our_hardcoded_version = 10507;
if (ZSTD_VERSION_NUMBER != our_hardcoded_version ||
zstd_ver_no != our_hardcoded_version) {
PyErr_Format(
PyExc_ImportError,
"zstd C API versions mismatch; Python bindings were not "
"compiled/linked against expected zstd version (%u returned by the "
"lib, %u hardcoded in zstd headers, %u hardcoded in the cext)",
zstd_ver_no, ZSTD_VERSION_NUMBER, our_hardcoded_version);
return;
}
features = PySet_New(NULL);
if (NULL == features) {
PyErr_SetString(PyExc_ImportError, "could not create empty set");
return;
}
feature = PyUnicode_FromString("buffer_types");
if (NULL == feature) {
PyErr_SetString(PyExc_ImportError, "could not create feature string");
return;
}
if (PySet_Add(features, feature) == -1) {
return;
}
Py_DECREF(feature);
#ifdef HAVE_ZSTD_POOL_APIS
feature = PyUnicode_FromString("multi_compress_to_buffer");
if (NULL == feature) {
PyErr_SetString(PyExc_ImportError, "could not create feature string");
return;
}
if (PySet_Add(features, feature) == -1) {
return;
}
Py_DECREF(feature);
#endif
#ifdef HAVE_ZSTD_POOL_APIS
feature = PyUnicode_FromString("multi_decompress_to_buffer");
if (NULL == feature) {
PyErr_SetString(PyExc_ImportError, "could not create feature string");
return;
}
if (PySet_Add(features, feature) == -1) {
return;
}
Py_DECREF(feature);
#endif
if (PyObject_SetAttrString(m, "backend_features", features) == -1) {
return;
}
Py_DECREF(features);
bufferutil_module_init(m);
compressionparams_module_init(m);
compressiondict_module_init(m);
compressobj_module_init(m);
compressor_module_init(m);
compressionchunker_module_init(m);
compressionreader_module_init(m);
compressionwriter_module_init(m);
compressoriterator_module_init(m);
constants_module_init(m);
decompressor_module_init(m);
decompressobj_module_init(m);
decompressionreader_module_init(m);
decompressionwriter_module_init(m);
decompressoriterator_module_init(m);
frameparams_module_init(m);
}
#if defined(__GNUC__) && (__GNUC__ >= 4)
#define PYTHON_ZSTD_VISIBILITY __attribute__((visibility("default")))
#else
#define PYTHON_ZSTD_VISIBILITY
#endif
static struct PyModuleDef zstd_module = {PyModuleDef_HEAD_INIT, "backend_c",
zstd_doc, -1, zstd_methods};
PYTHON_ZSTD_VISIBILITY PyMODINIT_FUNC PyInit_backend_c(void) {
PyObject *m = PyModule_Create(&zstd_module);
if (m) {
zstd_module_init(m);
if (PyErr_Occurred()) {
Py_DECREF(m);
m = NULL;
}
}
return m;
}
/* Attempt to resolve the number of CPUs in the system. */
int cpu_count() {
int count = 0;
#if defined(_WIN32)
SYSTEM_INFO si;
si.dwNumberOfProcessors = 0;
GetSystemInfo(&si);
count = si.dwNumberOfProcessors;
#elif defined(__APPLE__)
int num;
size_t size = sizeof(int);
if (0 == sysctlbyname("hw.logicalcpu", &num, &size, NULL, 0)) {
count = num;
}
#elif defined(__linux__)
count = sysconf(_SC_NPROCESSORS_ONLN);
#elif defined(__OpenBSD__) || defined(__FreeBSD__) || defined(__NetBSD__) || \
defined(__DragonFly__)
int mib[2];
size_t len = sizeof(count);
mib[0] = CTL_HW;
mib[1] = HW_NCPU;
if (0 != sysctl(mib, 2, &count, &len, NULL, 0)) {
count = 0;
}
#elif defined(__hpux)
count = mpctl(MPC_GETNUMSPUS, NULL, NULL);
#endif
return count;
}
size_t roundpow2(size_t i) {
i--;
i |= i >> 1;
i |= i >> 2;
i |= i >> 4;
i |= i >> 8;
i |= i >> 16;
i++;
return i;
}
/* Safer version of _PyBytes_Resize().
*
* _PyBytes_Resize() only works if the refcount is 1. In some scenarios,
* we can get an object with a refcount > 1, even if it was just created
* with PyBytes_FromStringAndSize()! That's because (at least) CPython
* pre-allocates PyBytes instances of size 1 for every possible byte value.
*
* If non-0 is returned, obj may or may not be NULL.
*/
int safe_pybytes_resize(PyObject **obj, Py_ssize_t size) {
PyObject *tmp;
if ((*obj)->ob_refcnt == 1) {
return _PyBytes_Resize(obj, size);
}
tmp = PyBytes_FromStringAndSize(NULL, size);
if (!tmp) {
return -1;
}
memcpy(PyBytes_AS_STRING(tmp), PyBytes_AS_STRING(*obj),
PyBytes_GET_SIZE(*obj));
Py_DECREF(*obj);
*obj = tmp;
return 0;
}
// Set/raise an `io.UnsupportedOperation` exception.
void set_io_unsupported_operation(void) {
PyObject *iomod;
PyObject *exc;
iomod = PyImport_ImportModule("io");
if (NULL == iomod) {
return;
}
exc = PyObject_GetAttrString(iomod, "UnsupportedOperation");
if (NULL == exc) {
Py_DECREF(iomod);
return;
}
PyErr_SetNone(exc);
Py_DECREF(exc);
Py_DECREF(iomod);
}
|