1#ifndef Py_UNICODEOBJECT_H
2#define Py_UNICODEOBJECT_H
3
4#include <stdarg.h>
5
6/*
7
8Unicode implementation based on original code by Fredrik Lundh,
9modified by Marc-Andre Lemburg ([email protected]) according to the
10Unicode Integration Proposal. (See
11http://www.egenix.com/files/python/unicode-proposal.txt).
12
13Copyright (c) Corporation for National Research Initiatives.
14
15
16 Original header:
17 --------------------------------------------------------------------
18
19 * Yet another Unicode string type for Python. This type supports the
20 * 16-bit Basic Multilingual Plane (BMP) only.
21 *
22 * Written by Fredrik Lundh, January 1999.
23 *
24 * Copyright (c) 1999 by Secret Labs AB.
25 * Copyright (c) 1999 by Fredrik Lundh.
26 *
27 * [email protected]
28 * http://www.pythonware.com
29 *
30 * --------------------------------------------------------------------
31 * This Unicode String Type is
32 *
33 * Copyright (c) 1999 by Secret Labs AB
34 * Copyright (c) 1999 by Fredrik Lundh
35 *
36 * By obtaining, using, and/or copying this software and/or its
37 * associated documentation, you agree that you have read, understood,
38 * and will comply with the following terms and conditions:
39 *
40 * Permission to use, copy, modify, and distribute this software and its
41 * associated documentation for any purpose and without fee is hereby
42 * granted, provided that the above copyright notice appears in all
43 * copies, and that both that copyright notice and this permission notice
44 * appear in supporting documentation, and that the name of Secret Labs
45 * AB or the author not be used in advertising or publicity pertaining to
46 * distribution of the software without specific, written prior
47 * permission.
48 *
49 * SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO
50 * THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
51 * FITNESS. IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR
52 * ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
53 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
54 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
55 * OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
56 * -------------------------------------------------------------------- */
57
58#include <ctype.h>
59
60/* === Internal API ======================================================= */
61
62/* --- Internal Unicode Format -------------------------------------------- */
63
64/* Python 3.x requires unicode */
65#define Py_USING_UNICODE
66
67#ifndef SIZEOF_WCHAR_T
68#error Must define SIZEOF_WCHAR_T
69#endif
70
71#define Py_UNICODE_SIZE SIZEOF_WCHAR_T
72
73/* If wchar_t can be used for UCS-4 storage, set Py_UNICODE_WIDE.
74 Otherwise, Unicode strings are stored as UCS-2 (with limited support
75 for UTF-16) */
76
77#if Py_UNICODE_SIZE >= 4
78#define Py_UNICODE_WIDE
79#endif
80
81/* Set these flags if the platform has "wchar.h" and the
82 wchar_t type is a 16-bit unsigned type */
83/* #define HAVE_WCHAR_H */
84/* #define HAVE_USABLE_WCHAR_T */
85
86/* If the compiler provides a wchar_t type we try to support it
87 through the interface functions PyUnicode_FromWideChar(),
88 PyUnicode_AsWideChar() and PyUnicode_AsWideCharString(). */
89
90#ifdef HAVE_USABLE_WCHAR_T
91# ifndef HAVE_WCHAR_H
92# define HAVE_WCHAR_H
93# endif
94#endif
95
96#ifdef HAVE_WCHAR_H
97# include <wchar.h>
98#endif
99
100/* Py_UCS4 and Py_UCS2 are typedefs for the respective
101 unicode representations. */
102typedef uint32_t Py_UCS4;
103typedef uint16_t Py_UCS2;
104typedef uint8_t Py_UCS1;
105
106#ifdef __cplusplus
107extern "C" {
108#endif
109
110
111PyAPI_DATA(PyTypeObject) PyUnicode_Type;
112PyAPI_DATA(PyTypeObject) PyUnicodeIter_Type;
113
114#define PyUnicode_Check(op) \
115 PyType_FastSubclass(Py_TYPE(op), Py_TPFLAGS_UNICODE_SUBCLASS)
116#define PyUnicode_CheckExact(op) Py_IS_TYPE(op, &PyUnicode_Type)
117
118/* --- Constants ---------------------------------------------------------- */
119
120/* This Unicode character will be used as replacement character during
121 decoding if the errors argument is set to "replace". Note: the
122 Unicode character U+FFFD is the official REPLACEMENT CHARACTER in
123 Unicode 3.0. */
124
125#define Py_UNICODE_REPLACEMENT_CHARACTER ((Py_UCS4) 0xFFFD)
126
127/* === Public API ========================================================= */
128
129/* Similar to PyUnicode_FromUnicode(), but u points to UTF-8 encoded bytes */
130PyAPI_FUNC(PyObject*) PyUnicode_FromStringAndSize(
131 const char *u, /* UTF-8 encoded string */
132 Py_ssize_t size /* size of buffer */
133 );
134
135/* Similar to PyUnicode_FromUnicode(), but u points to null-terminated
136 UTF-8 encoded bytes. The size is determined with strlen(). */
137PyAPI_FUNC(PyObject*) PyUnicode_FromString(
138 const char *u /* UTF-8 encoded string */
139 );
140
141#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
142PyAPI_FUNC(PyObject*) PyUnicode_Substring(
143 PyObject *str,
144 Py_ssize_t start,
145 Py_ssize_t end);
146#endif
147
148#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
149/* Copy the string into a UCS4 buffer including the null character if copy_null
150 is set. Return NULL and raise an exception on error. Raise a SystemError if
151 the buffer is smaller than the string. Return buffer on success.
152
153 buflen is the length of the buffer in (Py_UCS4) characters. */
154PyAPI_FUNC(Py_UCS4*) PyUnicode_AsUCS4(
155 PyObject *unicode,
156 Py_UCS4* buffer,
157 Py_ssize_t buflen,
158 int copy_null);
159
160/* Copy the string into a UCS4 buffer. A new buffer is allocated using
161 * PyMem_Malloc; if this fails, NULL is returned with a memory error
162 exception set. */
163PyAPI_FUNC(Py_UCS4*) PyUnicode_AsUCS4Copy(PyObject *unicode);
164#endif
165
166#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
167/* Get the length of the Unicode object. */
168
169PyAPI_FUNC(Py_ssize_t) PyUnicode_GetLength(
170 PyObject *unicode
171);
172#endif
173
174/* Get the number of Py_UNICODE units in the
175 string representation. */
176
177Py_DEPRECATED(3.3) PyAPI_FUNC(Py_ssize_t) PyUnicode_GetSize(
178 PyObject *unicode /* Unicode object */
179 );
180
181#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
182/* Read a character from the string. */
183
184PyAPI_FUNC(Py_UCS4) PyUnicode_ReadChar(
185 PyObject *unicode,
186 Py_ssize_t index
187 );
188
189/* Write a character to the string. The string must have been created through
190 PyUnicode_New, must not be shared, and must not have been hashed yet.
191
192 Return 0 on success, -1 on error. */
193
194PyAPI_FUNC(int) PyUnicode_WriteChar(
195 PyObject *unicode,
196 Py_ssize_t index,
197 Py_UCS4 character
198 );
199#endif
200
201/* Resize a Unicode object. The length is the number of characters, except
202 if the kind of the string is PyUnicode_WCHAR_KIND: in this case, the length
203 is the number of Py_UNICODE characters.
204
205 *unicode is modified to point to the new (resized) object and 0
206 returned on success.
207
208 Try to resize the string in place (which is usually faster than allocating
209 a new string and copy characters), or create a new string.
210
211 Error handling is implemented as follows: an exception is set, -1
212 is returned and *unicode left untouched.
213
214 WARNING: The function doesn't check string content, the result may not be a
215 string in canonical representation. */
216
217PyAPI_FUNC(int) PyUnicode_Resize(
218 PyObject **unicode, /* Pointer to the Unicode object */
219 Py_ssize_t length /* New length */
220 );
221
222/* Decode obj to a Unicode object.
223
224 bytes, bytearray and other bytes-like objects are decoded according to the
225 given encoding and error handler. The encoding and error handler can be
226 NULL to have the interface use UTF-8 and "strict".
227
228 All other objects (including Unicode objects) raise an exception.
229
230 The API returns NULL in case of an error. The caller is responsible
231 for decref'ing the returned objects.
232
233*/
234
235PyAPI_FUNC(PyObject*) PyUnicode_FromEncodedObject(
236 PyObject *obj, /* Object */
237 const char *encoding, /* encoding */
238 const char *errors /* error handling */
239 );
240
241/* Copy an instance of a Unicode subtype to a new true Unicode object if
242 necessary. If obj is already a true Unicode object (not a subtype), return
243 the reference with *incremented* refcount.
244
245 The API returns NULL in case of an error. The caller is responsible
246 for decref'ing the returned objects.
247
248*/
249
250PyAPI_FUNC(PyObject*) PyUnicode_FromObject(
251 PyObject *obj /* Object */
252 );
253
254PyAPI_FUNC(PyObject *) PyUnicode_FromFormatV(
255 const char *format, /* ASCII-encoded string */
256 va_list vargs
257 );
258PyAPI_FUNC(PyObject *) PyUnicode_FromFormat(
259 const char *format, /* ASCII-encoded string */
260 ...
261 );
262
263PyAPI_FUNC(void) PyUnicode_InternInPlace(PyObject **);
264PyAPI_FUNC(PyObject *) PyUnicode_InternFromString(
265 const char *u /* UTF-8 encoded string */
266 );
267
268// PyUnicode_InternImmortal() is deprecated since Python 3.10
269// and will be removed in Python 3.12. Use PyUnicode_InternInPlace() instead.
270Py_DEPRECATED(3.10) PyAPI_FUNC(void) PyUnicode_InternImmortal(PyObject **);
271
272/* Use only if you know it's a string */
273#define PyUnicode_CHECK_INTERNED(op) \
274 (((PyASCIIObject *)(op))->state.interned)
275
276/* --- wchar_t support for platforms which support it --------------------- */
277
278#ifdef HAVE_WCHAR_H
279
280/* Create a Unicode Object from the wchar_t buffer w of the given
281 size.
282
283 The buffer is copied into the new object. */
284
285PyAPI_FUNC(PyObject*) PyUnicode_FromWideChar(
286 const wchar_t *w, /* wchar_t buffer */
287 Py_ssize_t size /* size of buffer */
288 );
289
290/* Copies the Unicode Object contents into the wchar_t buffer w. At
291 most size wchar_t characters are copied.
292
293 Note that the resulting wchar_t string may or may not be
294 0-terminated. It is the responsibility of the caller to make sure
295 that the wchar_t string is 0-terminated in case this is required by
296 the application.
297
298 Returns the number of wchar_t characters copied (excluding a
299 possibly trailing 0-termination character) or -1 in case of an
300 error. */
301
302PyAPI_FUNC(Py_ssize_t) PyUnicode_AsWideChar(
303 PyObject *unicode, /* Unicode object */
304 wchar_t *w, /* wchar_t buffer */
305 Py_ssize_t size /* size of buffer */
306 );
307
308/* Convert the Unicode object to a wide character string. The output string
309 always ends with a nul character. If size is not NULL, write the number of
310 wide characters (excluding the null character) into *size.
311
312 Returns a buffer allocated by PyMem_Malloc() (use PyMem_Free() to free it)
313 on success. On error, returns NULL, *size is undefined and raises a
314 MemoryError. */
315
316PyAPI_FUNC(wchar_t*) PyUnicode_AsWideCharString(
317 PyObject *unicode, /* Unicode object */
318 Py_ssize_t *size /* number of characters of the result */
319 );
320
321#endif
322
323/* --- Unicode ordinals --------------------------------------------------- */
324
325/* Create a Unicode Object from the given Unicode code point ordinal.
326
327 The ordinal must be in range(0x110000). A ValueError is
328 raised in case it is not.
329
330*/
331
332PyAPI_FUNC(PyObject*) PyUnicode_FromOrdinal(int ordinal);
333
334/* === Builtin Codecs =====================================================
335
336 Many of these APIs take two arguments encoding and errors. These
337 parameters encoding and errors have the same semantics as the ones
338 of the builtin str() API.
339
340 Setting encoding to NULL causes the default encoding (UTF-8) to be used.
341
342 Error handling is set by errors which may also be set to NULL
343 meaning to use the default handling defined for the codec. Default
344 error handling for all builtin codecs is "strict" (ValueErrors are
345 raised).
346
347 The codecs all use a similar interface. Only deviation from the
348 generic ones are documented.
349
350*/
351
352/* --- Manage the default encoding ---------------------------------------- */
353
354/* Returns "utf-8". */
355PyAPI_FUNC(const char*) PyUnicode_GetDefaultEncoding(void);
356
357/* --- Generic Codecs ----------------------------------------------------- */
358
359/* Create a Unicode object by decoding the encoded string s of the
360 given size. */
361
362PyAPI_FUNC(PyObject*) PyUnicode_Decode(
363 const char *s, /* encoded string */
364 Py_ssize_t size, /* size of buffer */
365 const char *encoding, /* encoding */
366 const char *errors /* error handling */
367 );
368
369/* Decode a Unicode object unicode and return the result as Python
370 object.
371
372 This API is DEPRECATED. The only supported standard encoding is rot13.
373 Use PyCodec_Decode() to decode with rot13 and non-standard codecs
374 that decode from str. */
375
376Py_DEPRECATED(3.6) PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedObject(
377 PyObject *unicode, /* Unicode object */
378 const char *encoding, /* encoding */
379 const char *errors /* error handling */
380 );
381
382/* Decode a Unicode object unicode and return the result as Unicode
383 object.
384
385 This API is DEPRECATED. The only supported standard encoding is rot13.
386 Use PyCodec_Decode() to decode with rot13 and non-standard codecs
387 that decode from str to str. */
388
389Py_DEPRECATED(3.6) PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedUnicode(
390 PyObject *unicode, /* Unicode object */
391 const char *encoding, /* encoding */
392 const char *errors /* error handling */
393 );
394
395/* Encodes a Unicode object and returns the result as Python
396 object.
397
398 This API is DEPRECATED. It is superseded by PyUnicode_AsEncodedString()
399 since all standard encodings (except rot13) encode str to bytes.
400 Use PyCodec_Encode() for encoding with rot13 and non-standard codecs
401 that encode form str to non-bytes. */
402
403Py_DEPRECATED(3.6) PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedObject(
404 PyObject *unicode, /* Unicode object */
405 const char *encoding, /* encoding */
406 const char *errors /* error handling */
407 );
408
409/* Encodes a Unicode object and returns the result as Python string
410 object. */
411
412PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedString(
413 PyObject *unicode, /* Unicode object */
414 const char *encoding, /* encoding */
415 const char *errors /* error handling */
416 );
417
418/* Encodes a Unicode object and returns the result as Unicode
419 object.
420
421 This API is DEPRECATED. The only supported standard encodings is rot13.
422 Use PyCodec_Encode() to encode with rot13 and non-standard codecs
423 that encode from str to str. */
424
425Py_DEPRECATED(3.6) PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedUnicode(
426 PyObject *unicode, /* Unicode object */
427 const char *encoding, /* encoding */
428 const char *errors /* error handling */
429 );
430
431/* Build an encoding map. */
432
433PyAPI_FUNC(PyObject*) PyUnicode_BuildEncodingMap(
434 PyObject* string /* 256 character map */
435 );
436
437/* --- UTF-7 Codecs ------------------------------------------------------- */
438
439PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7(
440 const char *string, /* UTF-7 encoded string */
441 Py_ssize_t length, /* size of string */
442 const char *errors /* error handling */
443 );
444
445PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7Stateful(
446 const char *string, /* UTF-7 encoded string */
447 Py_ssize_t length, /* size of string */
448 const char *errors, /* error handling */
449 Py_ssize_t *consumed /* bytes consumed */
450 );
451
452/* --- UTF-8 Codecs ------------------------------------------------------- */
453
454PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8(
455 const char *string, /* UTF-8 encoded string */
456 Py_ssize_t length, /* size of string */
457 const char *errors /* error handling */
458 );
459
460PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8Stateful(
461 const char *string, /* UTF-8 encoded string */
462 Py_ssize_t length, /* size of string */
463 const char *errors, /* error handling */
464 Py_ssize_t *consumed /* bytes consumed */
465 );
466
467PyAPI_FUNC(PyObject*) PyUnicode_AsUTF8String(
468 PyObject *unicode /* Unicode object */
469 );
470
471/* Returns a pointer to the default encoding (UTF-8) of the
472 Unicode object unicode and the size of the encoded representation
473 in bytes stored in *size.
474
475 In case of an error, no *size is set.
476
477 This function caches the UTF-8 encoded string in the unicodeobject
478 and subsequent calls will return the same string. The memory is released
479 when the unicodeobject is deallocated.
480*/
481
482#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x030A0000
483PyAPI_FUNC(const char *) PyUnicode_AsUTF8AndSize(
484 PyObject *unicode,
485 Py_ssize_t *size);
486#endif
487
488/* --- UTF-32 Codecs ------------------------------------------------------ */
489
490/* Decodes length bytes from a UTF-32 encoded buffer string and returns
491 the corresponding Unicode object.
492
493 errors (if non-NULL) defines the error handling. It defaults
494 to "strict".
495
496 If byteorder is non-NULL, the decoder starts decoding using the
497 given byte order:
498
499 *byteorder == -1: little endian
500 *byteorder == 0: native order
501 *byteorder == 1: big endian
502
503 In native mode, the first four bytes of the stream are checked for a
504 BOM mark. If found, the BOM mark is analysed, the byte order
505 adjusted and the BOM skipped. In the other modes, no BOM mark
506 interpretation is done. After completion, *byteorder is set to the
507 current byte order at the end of input data.
508
509 If byteorder is NULL, the codec starts in native order mode.
510
511*/
512
513PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32(
514 const char *string, /* UTF-32 encoded string */
515 Py_ssize_t length, /* size of string */
516 const char *errors, /* error handling */
517 int *byteorder /* pointer to byteorder to use
518 0=native;-1=LE,1=BE; updated on
519 exit */
520 );
521
522PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32Stateful(
523 const char *string, /* UTF-32 encoded string */
524 Py_ssize_t length, /* size of string */
525 const char *errors, /* error handling */
526 int *byteorder, /* pointer to byteorder to use
527 0=native;-1=LE,1=BE; updated on
528 exit */
529 Py_ssize_t *consumed /* bytes consumed */
530 );
531
532/* Returns a Python string using the UTF-32 encoding in native byte
533 order. The string always starts with a BOM mark. */
534
535PyAPI_FUNC(PyObject*) PyUnicode_AsUTF32String(
536 PyObject *unicode /* Unicode object */
537 );
538
539/* Returns a Python string object holding the UTF-32 encoded value of
540 the Unicode data.
541
542 If byteorder is not 0, output is written according to the following
543 byte order:
544
545 byteorder == -1: little endian
546 byteorder == 0: native byte order (writes a BOM mark)
547 byteorder == 1: big endian
548
549 If byteorder is 0, the output string will always start with the
550 Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is
551 prepended.
552
553*/
554
555/* --- UTF-16 Codecs ------------------------------------------------------ */
556
557/* Decodes length bytes from a UTF-16 encoded buffer string and returns
558 the corresponding Unicode object.
559
560 errors (if non-NULL) defines the error handling. It defaults
561 to "strict".
562
563 If byteorder is non-NULL, the decoder starts decoding using the
564 given byte order:
565
566 *byteorder == -1: little endian
567 *byteorder == 0: native order
568 *byteorder == 1: big endian
569
570 In native mode, the first two bytes of the stream are checked for a
571 BOM mark. If found, the BOM mark is analysed, the byte order
572 adjusted and the BOM skipped. In the other modes, no BOM mark
573 interpretation is done. After completion, *byteorder is set to the
574 current byte order at the end of input data.
575
576 If byteorder is NULL, the codec starts in native order mode.
577
578*/
579
580PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16(
581 const char *string, /* UTF-16 encoded string */
582 Py_ssize_t length, /* size of string */
583 const char *errors, /* error handling */
584 int *byteorder /* pointer to byteorder to use
585 0=native;-1=LE,1=BE; updated on
586 exit */
587 );
588
589PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16Stateful(
590 const char *string, /* UTF-16 encoded string */
591 Py_ssize_t length, /* size of string */
592 const char *errors, /* error handling */
593 int *byteorder, /* pointer to byteorder to use
594 0=native;-1=LE,1=BE; updated on
595 exit */
596 Py_ssize_t *consumed /* bytes consumed */
597 );
598
599/* Returns a Python string using the UTF-16 encoding in native byte
600 order. The string always starts with a BOM mark. */
601
602PyAPI_FUNC(PyObject*) PyUnicode_AsUTF16String(
603 PyObject *unicode /* Unicode object */
604 );
605
606/* --- Unicode-Escape Codecs ---------------------------------------------- */
607
608PyAPI_FUNC(PyObject*) PyUnicode_DecodeUnicodeEscape(
609 const char *string, /* Unicode-Escape encoded string */
610 Py_ssize_t length, /* size of string */
611 const char *errors /* error handling */
612 );
613
614PyAPI_FUNC(PyObject*) PyUnicode_AsUnicodeEscapeString(
615 PyObject *unicode /* Unicode object */
616 );
617
618/* --- Raw-Unicode-Escape Codecs ------------------------------------------ */
619
620PyAPI_FUNC(PyObject*) PyUnicode_DecodeRawUnicodeEscape(
621 const char *string, /* Raw-Unicode-Escape encoded string */
622 Py_ssize_t length, /* size of string */
623 const char *errors /* error handling */
624 );
625
626PyAPI_FUNC(PyObject*) PyUnicode_AsRawUnicodeEscapeString(
627 PyObject *unicode /* Unicode object */
628 );
629
630/* --- Latin-1 Codecs -----------------------------------------------------
631
632 Note: Latin-1 corresponds to the first 256 Unicode ordinals. */
633
634PyAPI_FUNC(PyObject*) PyUnicode_DecodeLatin1(
635 const char *string, /* Latin-1 encoded string */
636 Py_ssize_t length, /* size of string */
637 const char *errors /* error handling */
638 );
639
640PyAPI_FUNC(PyObject*) PyUnicode_AsLatin1String(
641 PyObject *unicode /* Unicode object */
642 );
643
644/* --- ASCII Codecs -------------------------------------------------------
645
646 Only 7-bit ASCII data is excepted. All other codes generate errors.
647
648*/
649
650PyAPI_FUNC(PyObject*) PyUnicode_DecodeASCII(
651 const char *string, /* ASCII encoded string */
652 Py_ssize_t length, /* size of string */
653 const char *errors /* error handling */
654 );
655
656PyAPI_FUNC(PyObject*) PyUnicode_AsASCIIString(
657 PyObject *unicode /* Unicode object */
658 );
659
660/* --- Character Map Codecs -----------------------------------------------
661
662 This codec uses mappings to encode and decode characters.
663
664 Decoding mappings must map byte ordinals (integers in the range from 0 to
665 255) to Unicode strings, integers (which are then interpreted as Unicode
666 ordinals) or None. Unmapped data bytes (ones which cause a LookupError)
667 as well as mapped to None, 0xFFFE or '\ufffe' are treated as "undefined
668 mapping" and cause an error.
669
670 Encoding mappings must map Unicode ordinal integers to bytes objects,
671 integers in the range from 0 to 255 or None. Unmapped character
672 ordinals (ones which cause a LookupError) as well as mapped to
673 None are treated as "undefined mapping" and cause an error.
674
675*/
676
677PyAPI_FUNC(PyObject*) PyUnicode_DecodeCharmap(
678 const char *string, /* Encoded string */
679 Py_ssize_t length, /* size of string */
680 PyObject *mapping, /* decoding mapping */
681 const char *errors /* error handling */
682 );
683
684PyAPI_FUNC(PyObject*) PyUnicode_AsCharmapString(
685 PyObject *unicode, /* Unicode object */
686 PyObject *mapping /* encoding mapping */
687 );
688
689/* --- MBCS codecs for Windows -------------------------------------------- */
690
691#ifdef MS_WINDOWS
692PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCS(
693 const char *string, /* MBCS encoded string */
694 Py_ssize_t length, /* size of string */
695 const char *errors /* error handling */
696 );
697
698PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCSStateful(
699 const char *string, /* MBCS encoded string */
700 Py_ssize_t length, /* size of string */
701 const char *errors, /* error handling */
702 Py_ssize_t *consumed /* bytes consumed */
703 );
704
705#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
706PyAPI_FUNC(PyObject*) PyUnicode_DecodeCodePageStateful(
707 int code_page, /* code page number */
708 const char *string, /* encoded string */
709 Py_ssize_t length, /* size of string */
710 const char *errors, /* error handling */
711 Py_ssize_t *consumed /* bytes consumed */
712 );
713#endif
714
715PyAPI_FUNC(PyObject*) PyUnicode_AsMBCSString(
716 PyObject *unicode /* Unicode object */
717 );
718
719#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
720PyAPI_FUNC(PyObject*) PyUnicode_EncodeCodePage(
721 int code_page, /* code page number */
722 PyObject *unicode, /* Unicode object */
723 const char *errors /* error handling */
724 );
725#endif
726
727#endif /* MS_WINDOWS */
728
729/* --- Locale encoding --------------------------------------------------- */
730
731#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
732/* Decode a string from the current locale encoding. The decoder is strict if
733 *surrogateescape* is equal to zero, otherwise it uses the 'surrogateescape'
734 error handler (PEP 383) to escape undecodable bytes. If a byte sequence can
735 be decoded as a surrogate character and *surrogateescape* is not equal to
736 zero, the byte sequence is escaped using the 'surrogateescape' error handler
737 instead of being decoded. *str* must end with a null character but cannot
738 contain embedded null characters. */
739
740PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocaleAndSize(
741 const char *str,
742 Py_ssize_t len,
743 const char *errors);
744
745/* Similar to PyUnicode_DecodeLocaleAndSize(), but compute the string
746 length using strlen(). */
747
748PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocale(
749 const char *str,
750 const char *errors);
751
752/* Encode a Unicode object to the current locale encoding. The encoder is
753 strict is *surrogateescape* is equal to zero, otherwise the
754 "surrogateescape" error handler is used. Return a bytes object. The string
755 cannot contain embedded null characters. */
756
757PyAPI_FUNC(PyObject*) PyUnicode_EncodeLocale(
758 PyObject *unicode,
759 const char *errors
760 );
761#endif
762
763/* --- File system encoding ---------------------------------------------- */
764
765/* ParseTuple converter: encode str objects to bytes using
766 PyUnicode_EncodeFSDefault(); bytes objects are output as-is. */
767
768PyAPI_FUNC(int) PyUnicode_FSConverter(PyObject*, void*);
769
770/* ParseTuple converter: decode bytes objects to unicode using
771 PyUnicode_DecodeFSDefaultAndSize(); str objects are output as-is. */
772
773PyAPI_FUNC(int) PyUnicode_FSDecoder(PyObject*, void*);
774
775/* Decode a null-terminated string using Py_FileSystemDefaultEncoding
776 and the "surrogateescape" error handler.
777
778 If Py_FileSystemDefaultEncoding is not set, fall back to the locale
779 encoding.
780
781 Use PyUnicode_DecodeFSDefaultAndSize() if the string length is known.
782*/
783
784PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefault(
785 const char *s /* encoded string */
786 );
787
788/* Decode a string using Py_FileSystemDefaultEncoding
789 and the "surrogateescape" error handler.
790
791 If Py_FileSystemDefaultEncoding is not set, fall back to the locale
792 encoding.
793*/
794
795PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefaultAndSize(
796 const char *s, /* encoded string */
797 Py_ssize_t size /* size */
798 );
799
800/* Encode a Unicode object to Py_FileSystemDefaultEncoding with the
801 "surrogateescape" error handler, and return bytes.
802
803 If Py_FileSystemDefaultEncoding is not set, fall back to the locale
804 encoding.
805*/
806
807PyAPI_FUNC(PyObject*) PyUnicode_EncodeFSDefault(
808 PyObject *unicode
809 );
810
811/* --- Methods & Slots ----------------------------------------------------
812
813 These are capable of handling Unicode objects and strings on input
814 (we refer to them as strings in the descriptions) and return
815 Unicode objects or integers as appropriate. */
816
817/* Concat two strings giving a new Unicode string. */
818
819PyAPI_FUNC(PyObject*) PyUnicode_Concat(
820 PyObject *left, /* Left string */
821 PyObject *right /* Right string */
822 );
823
824/* Concat two strings and put the result in *pleft
825 (sets *pleft to NULL on error) */
826
827PyAPI_FUNC(void) PyUnicode_Append(
828 PyObject **pleft, /* Pointer to left string */
829 PyObject *right /* Right string */
830 );
831
832/* Concat two strings, put the result in *pleft and drop the right object
833 (sets *pleft to NULL on error) */
834
835PyAPI_FUNC(void) PyUnicode_AppendAndDel(
836 PyObject **pleft, /* Pointer to left string */
837 PyObject *right /* Right string */
838 );
839
840/* Split a string giving a list of Unicode strings.
841
842 If sep is NULL, splitting will be done at all whitespace
843 substrings. Otherwise, splits occur at the given separator.
844
845 At most maxsplit splits will be done. If negative, no limit is set.
846
847 Separators are not included in the resulting list.
848
849*/
850
851PyAPI_FUNC(PyObject*) PyUnicode_Split(
852 PyObject *s, /* String to split */
853 PyObject *sep, /* String separator */
854 Py_ssize_t maxsplit /* Maxsplit count */
855 );
856
857/* Dito, but split at line breaks.
858
859 CRLF is considered to be one line break. Line breaks are not
860 included in the resulting list. */
861
862PyAPI_FUNC(PyObject*) PyUnicode_Splitlines(
863 PyObject *s, /* String to split */
864 int keepends /* If true, line end markers are included */
865 );
866
867/* Partition a string using a given separator. */
868
869PyAPI_FUNC(PyObject*) PyUnicode_Partition(
870 PyObject *s, /* String to partition */
871 PyObject *sep /* String separator */
872 );
873
874/* Partition a string using a given separator, searching from the end of the
875 string. */
876
877PyAPI_FUNC(PyObject*) PyUnicode_RPartition(
878 PyObject *s, /* String to partition */
879 PyObject *sep /* String separator */
880 );
881
882/* Split a string giving a list of Unicode strings.
883
884 If sep is NULL, splitting will be done at all whitespace
885 substrings. Otherwise, splits occur at the given separator.
886
887 At most maxsplit splits will be done. But unlike PyUnicode_Split
888 PyUnicode_RSplit splits from the end of the string. If negative,
889 no limit is set.
890
891 Separators are not included in the resulting list.
892
893*/
894
895PyAPI_FUNC(PyObject*) PyUnicode_RSplit(
896 PyObject *s, /* String to split */
897 PyObject *sep, /* String separator */
898 Py_ssize_t maxsplit /* Maxsplit count */
899 );
900
901/* Translate a string by applying a character mapping table to it and
902 return the resulting Unicode object.
903
904 The mapping table must map Unicode ordinal integers to Unicode strings,
905 Unicode ordinal integers or None (causing deletion of the character).
906
907 Mapping tables may be dictionaries or sequences. Unmapped character
908 ordinals (ones which cause a LookupError) are left untouched and
909 are copied as-is.
910
911*/
912
913PyAPI_FUNC(PyObject *) PyUnicode_Translate(
914 PyObject *str, /* String */
915 PyObject *table, /* Translate table */
916 const char *errors /* error handling */
917 );
918
919/* Join a sequence of strings using the given separator and return
920 the resulting Unicode string. */
921
922PyAPI_FUNC(PyObject*) PyUnicode_Join(
923 PyObject *separator, /* Separator string */
924 PyObject *seq /* Sequence object */
925 );
926
927/* Return 1 if substr matches str[start:end] at the given tail end, 0
928 otherwise. */
929
930PyAPI_FUNC(Py_ssize_t) PyUnicode_Tailmatch(
931 PyObject *str, /* String */
932 PyObject *substr, /* Prefix or Suffix string */
933 Py_ssize_t start, /* Start index */
934 Py_ssize_t end, /* Stop index */
935 int direction /* Tail end: -1 prefix, +1 suffix */
936 );
937
938/* Return the first position of substr in str[start:end] using the
939 given search direction or -1 if not found. -2 is returned in case
940 an error occurred and an exception is set. */
941
942PyAPI_FUNC(Py_ssize_t) PyUnicode_Find(
943 PyObject *str, /* String */
944 PyObject *substr, /* Substring to find */
945 Py_ssize_t start, /* Start index */
946 Py_ssize_t end, /* Stop index */
947 int direction /* Find direction: +1 forward, -1 backward */
948 );
949
950#if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03030000
951/* Like PyUnicode_Find, but search for single character only. */
952PyAPI_FUNC(Py_ssize_t) PyUnicode_FindChar(
953 PyObject *str,
954 Py_UCS4 ch,
955 Py_ssize_t start,
956 Py_ssize_t end,
957 int direction
958 );
959#endif
960
961/* Count the number of occurrences of substr in str[start:end]. */
962
963PyAPI_FUNC(Py_ssize_t) PyUnicode_Count(
964 PyObject *str, /* String */
965 PyObject *substr, /* Substring to count */
966 Py_ssize_t start, /* Start index */
967 Py_ssize_t end /* Stop index */
968 );
969
970/* Replace at most maxcount occurrences of substr in str with replstr
971 and return the resulting Unicode object. */
972
973PyAPI_FUNC(PyObject *) PyUnicode_Replace(
974 PyObject *str, /* String */
975 PyObject *substr, /* Substring to find */
976 PyObject *replstr, /* Substring to replace */
977 Py_ssize_t maxcount /* Max. number of replacements to apply;
978 -1 = all */
979 );
980
981/* Compare two strings and return -1, 0, 1 for less than, equal,
982 greater than resp.
983 Raise an exception and return -1 on error. */
984
985PyAPI_FUNC(int) PyUnicode_Compare(
986 PyObject *left, /* Left string */
987 PyObject *right /* Right string */
988 );
989
990/* Compare a Unicode object with C string and return -1, 0, 1 for less than,
991 equal, and greater than, respectively. It is best to pass only
992 ASCII-encoded strings, but the function interprets the input string as
993 ISO-8859-1 if it contains non-ASCII characters.
994 This function does not raise exceptions. */
995
996PyAPI_FUNC(int) PyUnicode_CompareWithASCIIString(
997 PyObject *left,
998 const char *right /* ASCII-encoded string */
999 );
1000
1001/* Rich compare two strings and return one of the following:
1002
1003 - NULL in case an exception was raised
1004 - Py_True or Py_False for successful comparisons
1005 - Py_NotImplemented in case the type combination is unknown
1006
1007 Possible values for op:
1008
1009 Py_GT, Py_GE, Py_EQ, Py_NE, Py_LT, Py_LE
1010
1011*/
1012
1013PyAPI_FUNC(PyObject *) PyUnicode_RichCompare(
1014 PyObject *left, /* Left string */
1015 PyObject *right, /* Right string */
1016 int op /* Operation: Py_EQ, Py_NE, Py_GT, etc. */
1017 );
1018
1019/* Apply an argument tuple or dictionary to a format string and return
1020 the resulting Unicode string. */
1021
1022PyAPI_FUNC(PyObject *) PyUnicode_Format(
1023 PyObject *format, /* Format string */
1024 PyObject *args /* Argument tuple or dictionary */
1025 );
1026
1027/* Checks whether element is contained in container and return 1/0
1028 accordingly.
1029
1030 element has to coerce to a one element Unicode string. -1 is
1031 returned in case of an error. */
1032
1033PyAPI_FUNC(int) PyUnicode_Contains(
1034 PyObject *container, /* Container string */
1035 PyObject *element /* Element string */
1036 );
1037
1038/* Checks whether argument is a valid identifier. */
1039
1040PyAPI_FUNC(int) PyUnicode_IsIdentifier(PyObject *s);
1041
1042/* === Characters Type APIs =============================================== */
1043
1044#ifndef Py_LIMITED_API
1045# define Py_CPYTHON_UNICODEOBJECT_H
1046# include "cpython/unicodeobject.h"
1047# undef Py_CPYTHON_UNICODEOBJECT_H
1048#endif
1049
1050#ifdef __cplusplus
1051}
1052#endif
1053#endif /* !Py_UNICODEOBJECT_H */
1054