Use PyCapsule for internal datetime functions #51525
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,178 @@ | ||
| /* | ||
|
|
||
| Copyright (c) 2023, PyData Development Team | ||
| All rights reserved. | ||
|
|
||
| Distributed under the terms of the BSD Simplified License. | ||
|
|
||
| */ | ||
| #define _PANDAS_PARSER_IMPL | ||
|
|
||
| #include "pd_parser.h" | ||
| #include "src/parser/io.h" | ||
|
|
||
| static int to_double(char *item, double *p_value, char sci, char decimal, | ||
|
There was a problem hiding this comment. This and floatify were moved from |
||
| int *maybe_int) { | ||
| char *p_end = NULL; | ||
| int error = 0; | ||
|
|
||
| /* Switch to precise xstrtod GH 31364 */ | ||
| *p_value = | ||
| precise_xstrtod(item, &p_end, decimal, sci, '\0', 1, &error, maybe_int); | ||
|
|
||
| return (error == 0) && (!*p_end); | ||
| } | ||
|
|
||
| static int floatify(PyObject *str, double *result, int *maybe_int) { | ||
| int status; | ||
| char *data; | ||
| PyObject *tmp = NULL; | ||
| const char sci = 'E'; | ||
| const char dec = '.'; | ||
|
|
||
| if (PyBytes_Check(str)) { | ||
| data = PyBytes_AS_STRING(str); | ||
| } else if (PyUnicode_Check(str)) { | ||
| tmp = PyUnicode_AsUTF8String(str); | ||
| if (tmp == NULL) { | ||
| return -1; | ||
| } | ||
| data = PyBytes_AS_STRING(tmp); | ||
| } else { | ||
| PyErr_SetString(PyExc_TypeError, "Invalid object type"); | ||
| return -1; | ||
| } | ||
|
|
||
| status = to_double(data, result, sci, dec, maybe_int); | ||
|
|
||
| if (!status) { | ||
| /* handle inf/-inf infinity/-infinity */ | ||
| if (strlen(data) == 3) { | ||
| if (0 == strcasecmp(data, "inf")) { | ||
| *result = HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else { | ||
| goto parsingerror; | ||
| } | ||
| } else if (strlen(data) == 4) { | ||
| if (0 == strcasecmp(data, "-inf")) { | ||
| *result = -HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else if (0 == strcasecmp(data, "+inf")) { | ||
| *result = HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else { | ||
| goto parsingerror; | ||
| } | ||
| } else if (strlen(data) == 8) { | ||
| if (0 == strcasecmp(data, "infinity")) { | ||
| *result = HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else { | ||
| goto parsingerror; | ||
| } | ||
| } else if (strlen(data) == 9) { | ||
| if (0 == strcasecmp(data, "-infinity")) { | ||
| *result = -HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else if (0 == strcasecmp(data, "+infinity")) { | ||
| *result = HUGE_VAL; | ||
| *maybe_int = 0; | ||
| } else { | ||
| goto parsingerror; | ||
| } | ||
| } else { | ||
| goto parsingerror; | ||
| } | ||
| } | ||
|
|
||
| Py_XDECREF(tmp); | ||
| return 0; | ||
|
|
||
| parsingerror: | ||
| PyErr_Format(PyExc_ValueError, "Unable to parse string \"%s\"", data); | ||
| Py_XDECREF(tmp); | ||
| return -1; | ||
| } | ||
|
|
||
|
|
||
| static void pandas_parser_destructor(PyObject *op) { | ||
| void *ptr = PyCapsule_GetPointer(op, PandasParser_CAPSULE_NAME); | ||
| PyMem_Free(ptr); | ||
| } | ||
|
|
||
| static int pandas_parser_exec(PyObject *module) { | ||
| PandasParser_CAPI *capi = PyMem_Malloc(sizeof(PandasParser_CAPI)); | ||
| if (capi == NULL) { | ||
| PyErr_NoMemory(); | ||
| return -1; | ||
| } | ||
|
|
||
| capi->to_double = to_double; | ||
| capi->floatify = floatify; | ||
| capi->new_rd_source = new_rd_source; | ||
| capi->del_rd_source = del_rd_source; | ||
| capi->buffer_rd_bytes = buffer_rd_bytes; | ||
| capi->uint_state_init = uint_state_init; | ||
| capi->uint64_conflict = uint64_conflict; | ||
| capi->coliter_setup = coliter_setup; | ||
| capi->parser_new = parser_new; | ||
|
There was a problem hiding this comment. Is it possible to separate actual parsing stuff from string to numeric parsing more? There was a problem hiding this comment. Yea that's a great question. The vast majority of these are used in parsers.pyx; we could implement different capsules for different purposes. It probably would be paired well with a refactor of that Cython module. It is hard to tell from the history of parsers.pyx what was placed there as a matter of code organization versus tricks to prevent the build system from getting too heavy handed |
||
| capi->parser_init = parser_init; | ||
| capi->parser_free = parser_free; | ||
| capi->parser_del = parser_del; | ||
| capi->parser_add_skiprow = parser_add_skiprow; | ||
| capi->parser_set_skipfirstnrows = parser_set_skipfirstnrows; | ||
| capi->parser_set_default_options = parser_set_default_options; | ||
| capi->parser_consume_rows = parser_consume_rows; | ||
| capi->parser_trim_buffers = parser_trim_buffers; | ||
| capi->tokenize_all_rows = tokenize_all_rows; | ||
| capi->tokenize_nrows = tokenize_nrows; | ||
| capi->str_to_int64 = str_to_int64; | ||
| capi->str_to_uint64 = str_to_uint64; | ||
| capi->xstrtod = xstrtod; | ||
| capi->precise_xstrtod = precise_xstrtod; | ||
| capi->round_trip = round_trip; | ||
| capi->to_boolean = to_boolean; | ||
|
|
||
| PyObject *capsule = | ||
| PyCapsule_New(capi, PandasParser_CAPSULE_NAME, pandas_parser_destructor); | ||
| if (capsule == NULL) { | ||
| PyMem_Free(capi); | ||
| return -1; | ||
| } | ||
|
|
||
| // Monkeypatch the top level pandas module to have an attribute for the | ||
| // C-API. This is required because Python capsules do not support setting | ||
| // this attribute on anything but the top level package. Ideally not | ||
| // done when cpython gh-6898 gets implemented | ||
| PyObject *pandas = PyImport_ImportModule("pandas"); | ||
| if (!pandas) { | ||
| PyErr_SetString(PyExc_ImportError, | ||
| "pd_parser.c could not import module pandas"); | ||
| Py_DECREF(capsule); | ||
| return -1; | ||
| } | ||
|
|
||
| if (PyModule_AddObject(pandas, "_pandas_parser_CAPI", capsule) < 0) { | ||
|
There was a problem hiding this comment. I'm interested why the parser needs to have this exposed to Python (I can't see to ctrl+F anything useful out of the diff). There was a problem hiding this comment. I think this is just a requirement of the CPython implementation: https://docs.python.org/3/extending/extending.html#providing-a-c-api-for-an-extension-module Ideally we wouldn't add this to the pandas namespace and attach it solely to the extension module itself. From what I could see in the CPython source though I don't think that would be possible without python/cpython#6898 |
||
| Py_DECREF(capsule); | ||
| return -1; | ||
| } | ||
|
|
||
| return 0; | ||
| } | ||
|
|
||
| static PyModuleDef_Slot pandas_parser_slots[] = { | ||
| {Py_mod_exec, pandas_parser_exec}, {0, NULL}}; | ||
|
|
||
| static struct PyModuleDef pandas_parsermodule = { | ||
| PyModuleDef_HEAD_INIT, | ||
| .m_name = "pandas._libs.pandas_parser", | ||
|
|
||
| .m_doc = "Internal module with parser support for other extensions", | ||
| .m_size = 0, | ||
| .m_methods = NULL, | ||
| .m_slots = pandas_parser_slots}; | ||
|
|
||
| PyMODINIT_FUNC PyInit_pandas_parser(void) { | ||
| return PyModuleDef_Init(&pandas_parsermodule); | ||
| } | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This function actually returns a double; not sure why Cython was OK with the previous declaration but had to explictly change it as part of this PR for things to cythonize