cpython/Modules/_zstd/_zstdmodule.c

842 lines
24 KiB
C
Raw Normal View History

gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
/*
Low level interface to Meta's zstd library for use in the compression.zstd
Python module.
*/
#ifndef Py_BUILD_CORE_BUILTIN
# define Py_BUILD_CORE_MODULE 1
#endif
#include "_zstdmodule.h"
/*[clinic input]
module _zstd
[clinic start generated code]*/
/*[clinic end generated code: output=da39a3ee5e6b4b0d input=4b5f5587aac15c14]*/
#include "clinic/_zstdmodule.c.h"
/* Format error message and set ZstdError. */
void
set_zstd_error(const _zstd_state* const state,
error_type type, size_t zstd_ret)
{
char *msg;
assert(ZSTD_isError(zstd_ret));
switch (type)
{
case ERR_DECOMPRESS:
msg = "Unable to decompress zstd data: %s";
break;
case ERR_COMPRESS:
msg = "Unable to compress zstd data: %s";
break;
case ERR_LOAD_D_DICT:
msg = "Unable to load zstd dictionary or prefix for decompression: %s";
break;
case ERR_LOAD_C_DICT:
msg = "Unable to load zstd dictionary or prefix for compression: %s";
break;
case ERR_GET_C_BOUNDS:
msg = "Unable to get zstd compression parameter bounds: %s";
break;
case ERR_GET_D_BOUNDS:
msg = "Unable to get zstd decompression parameter bounds: %s";
break;
case ERR_SET_C_LEVEL:
msg = "Unable to set zstd compression level: %s";
break;
case ERR_TRAIN_DICT:
msg = "Unable to train zstd dictionary: %s";
break;
case ERR_FINALIZE_DICT:
msg = "Unable to finalize zstd dictionary: %s";
break;
default:
Py_UNREACHABLE();
}
PyErr_Format(state->ZstdError, msg, ZSTD_getErrorName(zstd_ret));
}
typedef struct {
int parameter;
char parameter_name[32];
} ParameterInfo;
static const ParameterInfo cp_list[] =
{
{ZSTD_c_compressionLevel, "compression_level"},
{ZSTD_c_windowLog, "window_log"},
{ZSTD_c_hashLog, "hash_log"},
{ZSTD_c_chainLog, "chain_log"},
{ZSTD_c_searchLog, "search_log"},
{ZSTD_c_minMatch, "min_match"},
{ZSTD_c_targetLength, "target_length"},
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{ZSTD_c_strategy, "strategy"},
{ZSTD_c_enableLongDistanceMatching, "enable_long_distance_matching"},
{ZSTD_c_ldmHashLog, "ldm_hash_log"},
{ZSTD_c_ldmMinMatch, "ldm_min_match"},
{ZSTD_c_ldmBucketSizeLog, "ldm_bucket_size_log"},
{ZSTD_c_ldmHashRateLog, "ldm_hash_rate_log"},
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{ZSTD_c_contentSizeFlag, "content_size_flag"},
{ZSTD_c_checksumFlag, "checksum_flag"},
{ZSTD_c_dictIDFlag, "dict_id_flag"},
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{ZSTD_c_nbWorkers, "nb_workers"},
{ZSTD_c_jobSize, "job_size"},
{ZSTD_c_overlapLog, "overlap_log"}
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
};
static const ParameterInfo dp_list[] =
{
{ZSTD_d_windowLogMax, "window_log_max"}
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
};
void
set_parameter_error(const _zstd_state* const state, int is_compress,
int key_v, int value_v)
{
ParameterInfo const *list;
int list_size;
char const *name;
char *type;
ZSTD_bounds bounds;
int i;
char pos_msg[128];
if (is_compress) {
list = cp_list;
list_size = Py_ARRAY_LENGTH(cp_list);
type = "compression";
}
else {
list = dp_list;
list_size = Py_ARRAY_LENGTH(dp_list);
type = "decompression";
}
/* Find parameter's name */
name = NULL;
for (i = 0; i < list_size; i++) {
if (key_v == (list+i)->parameter) {
name = (list+i)->parameter_name;
break;
}
}
/* Unknown parameter */
if (name == NULL) {
PyOS_snprintf(pos_msg, sizeof(pos_msg),
"unknown parameter (key %d)", key_v);
name = pos_msg;
}
/* Get parameter bounds */
if (is_compress) {
bounds = ZSTD_cParam_getBounds(key_v);
}
else {
bounds = ZSTD_dParam_getBounds(key_v);
}
if (ZSTD_isError(bounds.error)) {
PyErr_Format(state->ZstdError,
"Zstd %s parameter \"%s\" is invalid.",
type, name);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return;
}
/* Error message */
PyErr_Format(state->ZstdError,
"Error when setting zstd %s parameter \"%s\", it "
"should %d <= value <= %d, provided value is %d. "
"(%d-bit build)",
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
type, name,
bounds.lowerBound, bounds.upperBound, value_v,
8*(int)sizeof(Py_ssize_t));
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
}
static inline _zstd_state*
get_zstd_state(PyObject *module)
{
void *state = PyModule_GetState(module);
assert(state != NULL);
return (_zstd_state *)state;
}
/*[clinic input]
_zstd._train_dict
samples_bytes: PyBytesObject
Concatenation of samples.
samples_sizes: object(subclass_of='&PyTuple_Type')
Tuple of samples' sizes.
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
dict_size: Py_ssize_t
The size of the dictionary.
/
Internal function, train a zstd dictionary on sample data.
[clinic start generated code]*/
static PyObject *
_zstd__train_dict_impl(PyObject *module, PyBytesObject *samples_bytes,
PyObject *samples_sizes, Py_ssize_t dict_size)
/*[clinic end generated code: output=b5b4f36347c0addd input=2dce5b57d63923e2]*/
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{
// TODO(emmatyping): The preamble and suffix to this function and _finalize_dict
// are pretty similar. We should see if we can refactor them to share that code.
Py_ssize_t chunks_number;
size_t *chunk_sizes = NULL;
PyObject *dst_dict_bytes = NULL;
size_t zstd_ret;
Py_ssize_t sizes_sum;
Py_ssize_t i;
/* Check arguments */
if (dict_size <= 0) {
PyErr_SetString(PyExc_ValueError, "dict_size argument should be positive number.");
return NULL;
}
chunks_number = Py_SIZE(samples_sizes);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
if ((size_t) chunks_number > UINT32_MAX) {
PyErr_Format(PyExc_ValueError,
"The number of samples should be <= %u.", UINT32_MAX);
return NULL;
}
/* Prepare chunk_sizes */
chunk_sizes = PyMem_New(size_t, chunks_number);
if (chunk_sizes == NULL) {
PyErr_NoMemory();
goto error;
}
sizes_sum = 0;
for (i = 0; i < chunks_number; i++) {
PyObject *size = PyTuple_GetItem(samples_sizes, i);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
chunk_sizes[i] = PyLong_AsSize_t(size);
if (chunk_sizes[i] == (size_t)-1 && PyErr_Occurred()) {
PyErr_Format(PyExc_ValueError,
"Items in samples_sizes should be an int "
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
"object, with a value between 0 and %u.", SIZE_MAX);
goto error;
}
sizes_sum += chunk_sizes[i];
}
if (sizes_sum != Py_SIZE(samples_bytes)) {
PyErr_SetString(PyExc_ValueError,
"The samples size tuple doesn't match the concatenation's size.");
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
goto error;
}
/* Allocate dict buffer */
dst_dict_bytes = PyBytes_FromStringAndSize(NULL, dict_size);
if (dst_dict_bytes == NULL) {
goto error;
}
/* Train the dictionary */
char *dst_dict_buffer = PyBytes_AS_STRING(dst_dict_bytes);
char *samples_buffer = PyBytes_AS_STRING(samples_bytes);
Py_BEGIN_ALLOW_THREADS
zstd_ret = ZDICT_trainFromBuffer(dst_dict_buffer, dict_size,
samples_buffer,
chunk_sizes, (uint32_t)chunks_number);
Py_END_ALLOW_THREADS
/* Check zstd dict error */
if (ZDICT_isError(zstd_ret)) {
_zstd_state* const mod_state = get_zstd_state(module);
set_zstd_error(mod_state, ERR_TRAIN_DICT, zstd_ret);
goto error;
}
/* Resize dict_buffer */
if (_PyBytes_Resize(&dst_dict_bytes, zstd_ret) < 0) {
goto error;
}
goto success;
error:
Py_CLEAR(dst_dict_bytes);
success:
PyMem_Free(chunk_sizes);
return dst_dict_bytes;
}
/*[clinic input]
_zstd._finalize_dict
custom_dict_bytes: PyBytesObject
Custom dictionary content.
samples_bytes: PyBytesObject
Concatenation of samples.
samples_sizes: object(subclass_of='&PyTuple_Type')
Tuple of samples' sizes.
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
dict_size: Py_ssize_t
The size of the dictionary.
compression_level: int
Optimize for a specific zstd compression level, 0 means default.
/
Internal function, finalize a zstd dictionary.
[clinic start generated code]*/
static PyObject *
_zstd__finalize_dict_impl(PyObject *module, PyBytesObject *custom_dict_bytes,
PyBytesObject *samples_bytes,
PyObject *samples_sizes, Py_ssize_t dict_size,
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
int compression_level)
/*[clinic end generated code: output=5dc5b520fddba37f input=8afd42a249078460]*/
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{
Py_ssize_t chunks_number;
size_t *chunk_sizes = NULL;
PyObject *dst_dict_bytes = NULL;
size_t zstd_ret;
ZDICT_params_t params;
Py_ssize_t sizes_sum;
Py_ssize_t i;
/* Check arguments */
if (dict_size <= 0) {
PyErr_SetString(PyExc_ValueError, "dict_size argument should be positive number.");
return NULL;
}
chunks_number = Py_SIZE(samples_sizes);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
if ((size_t) chunks_number > UINT32_MAX) {
PyErr_Format(PyExc_ValueError,
"The number of samples should be <= %u.", UINT32_MAX);
return NULL;
}
/* Prepare chunk_sizes */
chunk_sizes = PyMem_New(size_t, chunks_number);
if (chunk_sizes == NULL) {
PyErr_NoMemory();
goto error;
}
sizes_sum = 0;
for (i = 0; i < chunks_number; i++) {
PyObject *size = PyTuple_GetItem(samples_sizes, i);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
chunk_sizes[i] = PyLong_AsSize_t(size);
if (chunk_sizes[i] == (size_t)-1 && PyErr_Occurred()) {
PyErr_Format(PyExc_ValueError,
"Items in samples_sizes should be an int "
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
"object, with a value between 0 and %u.", SIZE_MAX);
goto error;
}
sizes_sum += chunk_sizes[i];
}
if (sizes_sum != Py_SIZE(samples_bytes)) {
PyErr_SetString(PyExc_ValueError,
"The samples size tuple doesn't match the concatenation's size.");
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
goto error;
}
/* Allocate dict buffer */
dst_dict_bytes = PyBytes_FromStringAndSize(NULL, dict_size);
if (dst_dict_bytes == NULL) {
goto error;
}
/* Parameters */
/* Optimize for a specific zstd compression level, 0 means default. */
params.compressionLevel = compression_level;
/* Write log to stderr, 0 = none. */
params.notificationLevel = 0;
/* Force dictID value, 0 means auto mode (32-bits random value). */
params.dictID = 0;
/* Finalize the dictionary */
Py_BEGIN_ALLOW_THREADS
zstd_ret = ZDICT_finalizeDictionary(
PyBytes_AS_STRING(dst_dict_bytes), dict_size,
PyBytes_AS_STRING(custom_dict_bytes), Py_SIZE(custom_dict_bytes),
PyBytes_AS_STRING(samples_bytes), chunk_sizes,
(uint32_t)chunks_number, params);
Py_END_ALLOW_THREADS
/* Check zstd dict error */
if (ZDICT_isError(zstd_ret)) {
_zstd_state* const mod_state = get_zstd_state(module);
set_zstd_error(mod_state, ERR_FINALIZE_DICT, zstd_ret);
goto error;
}
/* Resize dict_buffer */
if (_PyBytes_Resize(&dst_dict_bytes, zstd_ret) < 0) {
goto error;
}
goto success;
error:
Py_CLEAR(dst_dict_bytes);
success:
PyMem_Free(chunk_sizes);
return dst_dict_bytes;
}
/*[clinic input]
_zstd._get_param_bounds
parameter: int
The parameter to get bounds.
is_compress: bool
True for CompressionParameter, False for DecompressionParameter.
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
Internal function, get CompressionParameter/DecompressionParameter bounds.
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
[clinic start generated code]*/
static PyObject *
_zstd__get_param_bounds_impl(PyObject *module, int parameter,
int is_compress)
/*[clinic end generated code: output=9892cd822f937e79 input=884cd1a01125267d]*/
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{
ZSTD_bounds bound;
if (is_compress) {
bound = ZSTD_cParam_getBounds(parameter);
if (ZSTD_isError(bound.error)) {
_zstd_state* const mod_state = get_zstd_state(module);
set_zstd_error(mod_state, ERR_GET_C_BOUNDS, bound.error);
return NULL;
}
}
else {
bound = ZSTD_dParam_getBounds(parameter);
if (ZSTD_isError(bound.error)) {
_zstd_state* const mod_state = get_zstd_state(module);
set_zstd_error(mod_state, ERR_GET_D_BOUNDS, bound.error);
return NULL;
}
}
return Py_BuildValue("ii", bound.lowerBound, bound.upperBound);
}
/*[clinic input]
_zstd.get_frame_size
frame_buffer: Py_buffer
A bytes-like object, it should start from the beginning of a frame,
and contains at least one complete frame.
Get the size of a zstd frame, including frame header and 4-byte checksum if it has one.
It will iterate all blocks' headers within a frame, to accumulate the frame size.
[clinic start generated code]*/
static PyObject *
_zstd_get_frame_size_impl(PyObject *module, Py_buffer *frame_buffer)
/*[clinic end generated code: output=a7384c2f8780f442 input=7d3ad24311893bf3]*/
{
size_t frame_size;
frame_size = ZSTD_findFrameCompressedSize(frame_buffer->buf, frame_buffer->len);
if (ZSTD_isError(frame_size)) {
_zstd_state* const mod_state = get_zstd_state(module);
PyErr_Format(mod_state->ZstdError,
"Error when finding the compressed size of a zstd frame. "
"Make sure the frame_buffer argument starts from the "
"beginning of a frame, and its length not less than this "
"complete frame. Zstd error message: %s.",
ZSTD_getErrorName(frame_size));
return NULL;
}
return PyLong_FromSize_t(frame_size);
}
/*[clinic input]
_zstd._get_frame_info
frame_buffer: Py_buffer
A bytes-like object, containing the header of a zstd frame.
Internal function, get zstd frame infomation from a frame header.
[clinic start generated code]*/
static PyObject *
_zstd__get_frame_info_impl(PyObject *module, Py_buffer *frame_buffer)
/*[clinic end generated code: output=5462855464ecdf81 input=67f1f8e4b7b89c4d]*/
{
uint64_t decompressed_size;
uint32_t dict_id;
/* ZSTD_getFrameContentSize */
decompressed_size = ZSTD_getFrameContentSize(frame_buffer->buf,
frame_buffer->len);
/* #define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1)
#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2) */
if (decompressed_size == ZSTD_CONTENTSIZE_ERROR) {
_zstd_state* const mod_state = get_zstd_state(module);
PyErr_SetString(mod_state->ZstdError,
"Error when getting information from the header of "
"a zstd frame. Make sure the frame_buffer argument "
"starts from the beginning of a frame, and its length "
"not less than the frame header (6~18 bytes).");
return NULL;
}
/* ZSTD_getDictID_fromFrame */
dict_id = ZSTD_getDictID_fromFrame(frame_buffer->buf, frame_buffer->len);
/* Build tuple */
if (decompressed_size == ZSTD_CONTENTSIZE_UNKNOWN) {
return Py_BuildValue("OI", Py_None, dict_id);
}
return Py_BuildValue("KI", decompressed_size, dict_id);
}
/*[clinic input]
_zstd._set_parameter_types
c_parameter_type: object(subclass_of='&PyType_Type')
CompressionParameter IntEnum type object
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
d_parameter_type: object(subclass_of='&PyType_Type')
DecompressionParameter IntEnum type object
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
Internal function, set CompressionParameter/DecompressionParameter types for validity check.
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
[clinic start generated code]*/
static PyObject *
_zstd__set_parameter_types_impl(PyObject *module, PyObject *c_parameter_type,
PyObject *d_parameter_type)
/*[clinic end generated code: output=a13d4890ccbd2873 input=4535545d903853d3]*/
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{
_zstd_state* const mod_state = get_zstd_state(module);
if (!PyType_Check(c_parameter_type) || !PyType_Check(d_parameter_type)) {
PyErr_SetString(PyExc_ValueError,
"The two arguments should be CompressionParameter and "
"DecompressionParameter types.");
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return NULL;
}
Py_XDECREF(mod_state->CParameter_type);
Py_INCREF(c_parameter_type);
mod_state->CParameter_type = (PyTypeObject*)c_parameter_type;
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
Py_XDECREF(mod_state->DParameter_type);
Py_INCREF(d_parameter_type);
mod_state->DParameter_type = (PyTypeObject*)d_parameter_type;
Py_RETURN_NONE;
}
static PyMethodDef _zstd_methods[] = {
_ZSTD__TRAIN_DICT_METHODDEF
_ZSTD__FINALIZE_DICT_METHODDEF
_ZSTD__GET_PARAM_BOUNDS_METHODDEF
_ZSTD_GET_FRAME_SIZE_METHODDEF
_ZSTD__GET_FRAME_INFO_METHODDEF
_ZSTD__SET_PARAMETER_TYPES_METHODDEF
{0}
};
static inline int
add_vars_to_module(PyObject *m)
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{
#define ADD_INT_MACRO(MACRO) \
if (PyModule_AddIntConstant((m), #MACRO, (MACRO)) < 0) { \
return -1; \
}
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
/* zstd_version_number, int */
if (PyModule_AddIntConstant(m, "zstd_version_number",
ZSTD_versionNumber()) < 0) {
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return -1;
}
/* zstd_version, str */
if (PyModule_AddStringConstant(m, "zstd_version",
ZSTD_versionString()) < 0) {
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return -1;
}
/* ZSTD_CLEVEL_DEFAULT, int */
#if ZSTD_VERSION_NUMBER >= 10500
if (PyModule_AddIntConstant(m, "ZSTD_CLEVEL_DEFAULT",
ZSTD_defaultCLevel()) < 0) {
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return -1;
}
#else
ADD_INT_MACRO(ZSTD_CLEVEL_DEFAULT);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
#endif
/* ZSTD_DStreamOutSize, int */
if (PyModule_Add(m, "ZSTD_DStreamOutSize",
PyLong_FromSize_t(ZSTD_DStreamOutSize())) < 0) {
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return -1;
}
/* Add zstd compression parameters. All should also be in cp_list. */
ADD_INT_MACRO(ZSTD_c_compressionLevel);
ADD_INT_MACRO(ZSTD_c_windowLog);
ADD_INT_MACRO(ZSTD_c_hashLog);
ADD_INT_MACRO(ZSTD_c_chainLog);
ADD_INT_MACRO(ZSTD_c_searchLog);
ADD_INT_MACRO(ZSTD_c_minMatch);
ADD_INT_MACRO(ZSTD_c_targetLength);
ADD_INT_MACRO(ZSTD_c_strategy);
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
ADD_INT_MACRO(ZSTD_c_enableLongDistanceMatching);
ADD_INT_MACRO(ZSTD_c_ldmHashLog);
ADD_INT_MACRO(ZSTD_c_ldmMinMatch);
ADD_INT_MACRO(ZSTD_c_ldmBucketSizeLog);
ADD_INT_MACRO(ZSTD_c_ldmHashRateLog);
ADD_INT_MACRO(ZSTD_c_contentSizeFlag);
ADD_INT_MACRO(ZSTD_c_checksumFlag);
ADD_INT_MACRO(ZSTD_c_dictIDFlag);
ADD_INT_MACRO(ZSTD_c_nbWorkers);
ADD_INT_MACRO(ZSTD_c_jobSize);
ADD_INT_MACRO(ZSTD_c_overlapLog);
/* Add zstd decompression parameters. All should also be in dp_list. */
ADD_INT_MACRO(ZSTD_d_windowLogMax);
/* ZSTD_strategy enum */
ADD_INT_MACRO(ZSTD_fast);
ADD_INT_MACRO(ZSTD_dfast);
ADD_INT_MACRO(ZSTD_greedy);
ADD_INT_MACRO(ZSTD_lazy);
ADD_INT_MACRO(ZSTD_lazy2);
ADD_INT_MACRO(ZSTD_btlazy2);
ADD_INT_MACRO(ZSTD_btopt);
ADD_INT_MACRO(ZSTD_btultra);
ADD_INT_MACRO(ZSTD_btultra2);
#undef ADD_INT_MACRO
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
return 0;
}
#define ADD_STR_TO_STATE_MACRO(STR) \
do { \
mod_state->str_##STR = PyUnicode_FromString(#STR); \
if (mod_state->str_##STR == NULL) { \
return -1; \
} \
} while(0)
static inline int
add_type_to_module(PyObject *module, const char *name,
PyType_Spec *type_spec, PyTypeObject **dest)
{
PyObject *temp = PyType_FromModuleAndSpec(module, type_spec, NULL);
if (PyModule_AddObjectRef(module, name, temp) < 0) {
Py_XDECREF(temp);
return -1;
}
*dest = (PyTypeObject*) temp;
return 0;
}
static inline int
add_constant_to_type(PyTypeObject *type, const char *name, long value)
{
PyObject *temp;
temp = PyLong_FromLong(value);
if (temp == NULL) {
return -1;
}
int rc = PyObject_SetAttrString((PyObject*) type, name, temp);
Py_DECREF(temp);
return rc;
}
static int _zstd_exec(PyObject *module) {
_zstd_state* const mod_state = get_zstd_state(module);
/* Reusable objects & variables */
mod_state->empty_bytes = PyBytes_FromStringAndSize(NULL, 0);
if (mod_state->empty_bytes == NULL) {
return -1;
}
mod_state->empty_readonly_memoryview =
PyMemoryView_FromMemory((char*)mod_state, 0, PyBUF_READ);
if (mod_state->empty_readonly_memoryview == NULL) {
return -1;
}
/* Add str to module state */
ADD_STR_TO_STATE_MACRO(read);
ADD_STR_TO_STATE_MACRO(readinto);
ADD_STR_TO_STATE_MACRO(write);
ADD_STR_TO_STATE_MACRO(flush);
mod_state->CParameter_type = NULL;
mod_state->DParameter_type = NULL;
/* Add variables to module */
if (add_vars_to_module(module) < 0) {
return -1;
}
/* ZstdError */
mod_state->ZstdError = PyErr_NewExceptionWithDoc(
"_zstd.ZstdError",
"Call to the underlying zstd library failed.",
NULL, NULL);
if (mod_state->ZstdError == NULL) {
return -1;
}
if (PyModule_AddObjectRef(module, "ZstdError", mod_state->ZstdError) < 0) {
Py_DECREF(mod_state->ZstdError);
return -1;
}
/* ZstdDict */
if (add_type_to_module(module,
"ZstdDict",
&zstddict_type_spec,
&mod_state->ZstdDict_type) < 0) {
return -1;
}
// ZstdCompressor
if (add_type_to_module(module,
"ZstdCompressor",
&zstdcompressor_type_spec,
&mod_state->ZstdCompressor_type) < 0) {
return -1;
}
// Add EndDirective enum to ZstdCompressor
if (add_constant_to_type(mod_state->ZstdCompressor_type,
"CONTINUE",
ZSTD_e_continue) < 0) {
return -1;
}
if (add_constant_to_type(mod_state->ZstdCompressor_type,
"FLUSH_BLOCK",
ZSTD_e_flush) < 0) {
return -1;
}
if (add_constant_to_type(mod_state->ZstdCompressor_type,
"FLUSH_FRAME",
ZSTD_e_end) < 0) {
return -1;
}
// ZstdDecompressor
if (add_type_to_module(module,
"ZstdDecompressor",
&zstddecompressor_type_spec,
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
&mod_state->ZstdDecompressor_type) < 0) {
return -1;
}
return 0;
}
static int
_zstd_traverse(PyObject *module, visitproc visit, void *arg)
{
_zstd_state* const mod_state = get_zstd_state(module);
Py_VISIT(mod_state->empty_bytes);
Py_VISIT(mod_state->empty_readonly_memoryview);
Py_VISIT(mod_state->str_read);
Py_VISIT(mod_state->str_readinto);
Py_VISIT(mod_state->str_write);
Py_VISIT(mod_state->str_flush);
Py_VISIT(mod_state->ZstdDict_type);
Py_VISIT(mod_state->ZstdCompressor_type);
Py_VISIT(mod_state->ZstdDecompressor_type);
Py_VISIT(mod_state->ZstdError);
Py_VISIT(mod_state->CParameter_type);
Py_VISIT(mod_state->DParameter_type);
return 0;
}
static int
_zstd_clear(PyObject *module)
{
_zstd_state* const mod_state = get_zstd_state(module);
Py_CLEAR(mod_state->empty_bytes);
Py_CLEAR(mod_state->empty_readonly_memoryview);
Py_CLEAR(mod_state->str_read);
Py_CLEAR(mod_state->str_readinto);
Py_CLEAR(mod_state->str_write);
Py_CLEAR(mod_state->str_flush);
Py_CLEAR(mod_state->ZstdDict_type);
Py_CLEAR(mod_state->ZstdCompressor_type);
Py_CLEAR(mod_state->ZstdDecompressor_type);
Py_CLEAR(mod_state->ZstdError);
Py_CLEAR(mod_state->CParameter_type);
Py_CLEAR(mod_state->DParameter_type);
return 0;
}
static void
_zstd_free(void *module)
{
(void)_zstd_clear((PyObject *)module);
}
static struct PyModuleDef_Slot _zstd_slots[] = {
{Py_mod_exec, _zstd_exec},
{Py_mod_multiple_interpreters, Py_MOD_PER_INTERPRETER_GIL_SUPPORTED},
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
{Py_mod_gil, Py_MOD_GIL_NOT_USED},
{0, NULL},
gh-132983: Introduce `_zstd` bindings module (GH-133027) * Add _zstd module for https://peps.python.org/pep-0784/ This commit introduces the `_zstd` module, with bindings to libzstd from the pyzstd project. It also includes the unix build system configuration. Windows build system support will be integrated independently as it depends on integration with cpython-source-deps. * Add _zstd to modules * Fix path for compression.zstd module * Ignore _zstd module like _io * Expand module state macros to improve code quality Also removes module state references from the classes in the _zstd module and instead uses PyType_GetModuleState() * Remove backticks suggested in review Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> * Use critical sections to lock object state This should avoid races and deadlocks. * Remove compress/decompress and mark module as not reliant on the GIL The `compress`/`decompress` functions will be moved to Python code for simplicity. C implementations can always be re-added in the future. Also, mark _zstd as not requiring the GIL. * Lift critical section to avoid clang warning * Respond to comments by picnixz * Call out pyzstd explicitly in license description Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> * Use a much more robust implementation... ... for `get_zstd_state_from_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Use PyList_GetItemRef for thread safety purposes * Use a macro for the minimum supported version * remove const from primivite types * Use PyMem_New in another spot * Simplify error handling in _get_frame_size * Another simplification of error handling in get_frame_info * Rename _module_state to mod_state * Rewrite comment explaining the context of the code * Add link to pyzstd * Add TODO about refactoring dict training code * Use PyModule_AddObjectRef over PyModule_AddObject PyModule_AddObject is soft-deprecated, so we should use PyModule_AddObjectRef * Check result of OutputBufferGrow * Simplify return logic in `add_constant_to_type` Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Ignore return value of _zstd_clear() Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> * Remove redundant comments * Remove __reduce__ from ZstdDict We should instead document that to pickle a dictionary a user should use the `.dict_content` attribute. * Use PyUnicode_FromFormat instead of a buffer * Don't use C constants/types in error messages * Make error messages easier to understand for Python users * Lower minimum required version 1.4.0 * Use casts and make slot function signatures correct * Be consistent with CPython on const usage * Make else clauses in line with PEP 7 * Fix over-indented blocks in argument clinic * Add critical section around ZSTD_DCtx_setParameter * Add a TODO about refactoring critical sections * Use Py_UNREACHABLE * Move bytes operations out of Py_BEGIN_ALLOW_THREADS * Add TODO about ensuring a lock is held * Remove asserts that may not be correct * Add TODO to make ZstdDict and others GC objects * Make objects GC tracked * Remove unused include * Fix some memory issues * Fix refleaks on module and in ZstdDict * Update configure to check for ZDICT_finalizeDictionary * Properly check version in configure * exit(1) if check fails * Use AC_RUN_IFELSE * Use a define() to re-use version check * Actually properly set _zstd module status based on version --------- Co-authored-by: Stan Ulbrych <89152624+StanFromIreland@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
2025-05-03 18:29:55 -07:00
};
struct PyModuleDef _zstdmodule = {
PyModuleDef_HEAD_INIT,
.m_name = "_zstd",
.m_size = sizeof(_zstd_state),
.m_slots = _zstd_slots,
.m_methods = _zstd_methods,
.m_traverse = _zstd_traverse,
.m_clear = _zstd_clear,
.m_free = _zstd_free
};
PyMODINIT_FUNC
PyInit__zstd(void)
{
return PyModuleDef_Init(&_zstdmodule);
}