unicodeobject.h revision bfbc29cb8f5797f239a6ad8944edfea4e153cb75
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 (mal@lemburg.com) 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 * fredrik@pythonware.com 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/* Py_UNICODE was the native Unicode storage format (code unit) used by 87 Python and represents a single Unicode element in the Unicode type. 88 With PEP 393, Py_UNICODE is deprecated and replaced with a 89 typedef to wchar_t. */ 90 91#ifndef Py_LIMITED_API 92#define PY_UNICODE_TYPE wchar_t 93typedef wchar_t Py_UNICODE; 94#endif 95 96/* If the compiler provides a wchar_t type we try to support it 97 through the interface functions PyUnicode_FromWideChar(), 98 PyUnicode_AsWideChar() and PyUnicode_AsWideCharString(). */ 99 100#ifdef HAVE_USABLE_WCHAR_T 101# ifndef HAVE_WCHAR_H 102# define HAVE_WCHAR_H 103# endif 104#endif 105 106#if defined(MS_WINDOWS) 107# define HAVE_MBCS 108#endif 109 110#ifdef HAVE_WCHAR_H 111/* Work around a cosmetic bug in BSDI 4.x wchar.h; thanks to Thomas Wouters */ 112# ifdef _HAVE_BSDI 113# include <time.h> 114# endif 115# include <wchar.h> 116#endif 117 118/* Py_UCS4 and Py_UCS2 are typedefs for the respective 119 unicode representations. */ 120#if SIZEOF_INT == 4 121typedef unsigned int Py_UCS4; 122#elif SIZEOF_LONG == 4 123typedef unsigned long Py_UCS4; 124#else 125#error "Could not find a proper typedef for Py_UCS4" 126#endif 127 128#if SIZEOF_SHORT == 2 129typedef unsigned short Py_UCS2; 130#else 131#error "Could not find a proper typedef for Py_UCS2" 132#endif 133 134typedef unsigned char Py_UCS1; 135 136/* --- Internal Unicode Operations ---------------------------------------- */ 137 138/* Since splitting on whitespace is an important use case, and 139 whitespace in most situations is solely ASCII whitespace, we 140 optimize for the common case by using a quick look-up table 141 _Py_ascii_whitespace (see below) with an inlined check. 142 143 */ 144#ifndef Py_LIMITED_API 145#define Py_UNICODE_ISSPACE(ch) \ 146 ((ch) < 128U ? _Py_ascii_whitespace[(ch)] : _PyUnicode_IsWhitespace(ch)) 147 148#define Py_UNICODE_ISLOWER(ch) _PyUnicode_IsLowercase(ch) 149#define Py_UNICODE_ISUPPER(ch) _PyUnicode_IsUppercase(ch) 150#define Py_UNICODE_ISTITLE(ch) _PyUnicode_IsTitlecase(ch) 151#define Py_UNICODE_ISLINEBREAK(ch) _PyUnicode_IsLinebreak(ch) 152 153#define Py_UNICODE_TOLOWER(ch) _PyUnicode_ToLowercase(ch) 154#define Py_UNICODE_TOUPPER(ch) _PyUnicode_ToUppercase(ch) 155#define Py_UNICODE_TOTITLE(ch) _PyUnicode_ToTitlecase(ch) 156 157#define Py_UNICODE_ISDECIMAL(ch) _PyUnicode_IsDecimalDigit(ch) 158#define Py_UNICODE_ISDIGIT(ch) _PyUnicode_IsDigit(ch) 159#define Py_UNICODE_ISNUMERIC(ch) _PyUnicode_IsNumeric(ch) 160#define Py_UNICODE_ISPRINTABLE(ch) _PyUnicode_IsPrintable(ch) 161 162#define Py_UNICODE_TODECIMAL(ch) _PyUnicode_ToDecimalDigit(ch) 163#define Py_UNICODE_TODIGIT(ch) _PyUnicode_ToDigit(ch) 164#define Py_UNICODE_TONUMERIC(ch) _PyUnicode_ToNumeric(ch) 165 166#define Py_UNICODE_ISALPHA(ch) _PyUnicode_IsAlpha(ch) 167 168#define Py_UNICODE_ISALNUM(ch) \ 169 (Py_UNICODE_ISALPHA(ch) || \ 170 Py_UNICODE_ISDECIMAL(ch) || \ 171 Py_UNICODE_ISDIGIT(ch) || \ 172 Py_UNICODE_ISNUMERIC(ch)) 173 174#define Py_UNICODE_COPY(target, source, length) \ 175 Py_MEMCPY((target), (source), (length)*sizeof(Py_UNICODE)) 176 177#define Py_UNICODE_FILL(target, value, length) \ 178 do {Py_ssize_t i_; Py_UNICODE *t_ = (target); Py_UNICODE v_ = (value);\ 179 for (i_ = 0; i_ < (length); i_++) t_[i_] = v_;\ 180 } while (0) 181 182/* macros to work with surrogates */ 183#define Py_UNICODE_IS_SURROGATE(ch) (0xD800 <= (ch) && (ch) <= 0xDFFF) 184#define Py_UNICODE_IS_HIGH_SURROGATE(ch) (0xD800 <= (ch) && (ch) <= 0xDBFF) 185#define Py_UNICODE_IS_LOW_SURROGATE(ch) (0xDC00 <= (ch) && (ch) <= 0xDFFF) 186/* Join two surrogate characters and return a single Py_UCS4 value. */ 187#define Py_UNICODE_JOIN_SURROGATES(high, low) \ 188 (((((Py_UCS4)(high) & 0x03FF) << 10) | \ 189 ((Py_UCS4)(low) & 0x03FF)) + 0x10000) 190/* high surrogate = top 10 bits added to D800 */ 191#define Py_UNICODE_HIGH_SURROGATE(ch) (0xD800 - (0x10000 >> 10) + ((ch) >> 10)) 192/* low surrogate = bottom 10 bits added to DC00 */ 193#define Py_UNICODE_LOW_SURROGATE(ch) (0xDC00 + ((ch) & 0x3FF)) 194 195/* Check if substring matches at given offset. The offset must be 196 valid, and the substring must not be empty. */ 197 198#define Py_UNICODE_MATCH(string, offset, substring) \ 199 ((*((string)->wstr + (offset)) == *((substring)->wstr)) && \ 200 ((*((string)->wstr + (offset) + (substring)->wstr_length-1) == *((substring)->wstr + (substring)->wstr_length-1))) && \ 201 !memcmp((string)->wstr + (offset), (substring)->wstr, (substring)->wstr_length*sizeof(Py_UNICODE))) 202 203#endif /* Py_LIMITED_API */ 204 205#ifdef __cplusplus 206extern "C" { 207#endif 208 209/* --- Unicode Type ------------------------------------------------------- */ 210 211#ifndef Py_LIMITED_API 212 213/* ASCII-only strings created through PyUnicode_New use the PyASCIIObject 214 structure. state.ascii and state.compact are set, and the data 215 immediately follow the structure. utf8_length and wstr_length can be found 216 in the length field; the utf8 pointer is equal to the data pointer. */ 217typedef struct { 218 /* There are 4 forms of Unicode strings: 219 220 - compact ascii: 221 222 * structure = PyASCIIObject 223 * test: PyUnicode_IS_COMPACT_ASCII(op) 224 * kind = PyUnicode_1BYTE_KIND 225 * compact = 1 226 * ascii = 1 227 * ready = 1 228 * (length is the length of the utf8 and wstr strings) 229 * (data starts just after the structure) 230 * (since ASCII is decoded from UTF-8, the utf8 string are the data) 231 232 - compact: 233 234 * structure = PyCompactUnicodeObject 235 * test: PyUnicode_IS_COMPACT(op) && !PyUnicode_IS_ASCII(op) 236 * kind = PyUnicode_1BYTE_KIND, PyUnicode_2BYTE_KIND or 237 PyUnicode_4BYTE_KIND 238 * compact = 1 239 * ready = 1 240 * ascii = 0 241 * utf8 is not shared with data 242 * utf8_length = 0 if utf8 is NULL 243 * wstr is shared with data and wstr_length=length 244 if kind=PyUnicode_2BYTE_KIND and sizeof(wchar_t)=2 245 or if kind=PyUnicode_4BYTE_KIND and sizeof(wchar_t)=4 246 * wstr_length = 0 if wstr is NULL 247 * (data starts just after the structure) 248 249 - legacy string, not ready: 250 251 * structure = PyUnicodeObject 252 * test: kind == PyUnicode_WCHAR_KIND 253 * length = 0 (use wstr_length) 254 * hash = -1 255 * kind = PyUnicode_WCHAR_KIND 256 * compact = 0 257 * ascii = 0 258 * ready = 0 259 * interned = SSTATE_NOT_INTERNED 260 * wstr is not NULL 261 * data.any is NULL 262 * utf8 is NULL 263 * utf8_length = 0 264 265 - legacy string, ready: 266 267 * structure = PyUnicodeObject structure 268 * test: !PyUnicode_IS_COMPACT(op) && kind != PyUnicode_WCHAR_KIND 269 * kind = PyUnicode_1BYTE_KIND, PyUnicode_2BYTE_KIND or 270 PyUnicode_4BYTE_KIND 271 * compact = 0 272 * ready = 1 273 * data.any is not NULL 274 * utf8 is shared and utf8_length = length with data.any if ascii = 1 275 * utf8_length = 0 if utf8 is NULL 276 * wstr is shared with data.any and wstr_length = length 277 if kind=PyUnicode_2BYTE_KIND and sizeof(wchar_t)=2 278 or if kind=PyUnicode_4BYTE_KIND and sizeof(wchar_4)=4 279 * wstr_length = 0 if wstr is NULL 280 281 Compact strings use only one memory block (structure + characters), 282 whereas legacy strings use one block for the structure and one block 283 for characters. 284 285 Legacy strings are created by PyUnicode_FromUnicode() and 286 PyUnicode_FromStringAndSize(NULL, size) functions. They become ready 287 when PyUnicode_READY() is called. 288 289 See also _PyUnicode_CheckConsistency(). 290 */ 291 PyObject_HEAD 292 Py_ssize_t length; /* Number of code points in the string */ 293 Py_hash_t hash; /* Hash value; -1 if not set */ 294 struct { 295 /* 296 SSTATE_NOT_INTERNED (0) 297 SSTATE_INTERNED_MORTAL (1) 298 SSTATE_INTERNED_IMMORTAL (2) 299 300 If interned != SSTATE_NOT_INTERNED, the two references from the 301 dictionary to this object are *not* counted in ob_refcnt. 302 */ 303 unsigned int interned:2; 304 /* Character size: 305 306 - PyUnicode_WCHAR_KIND (0): 307 308 * character type = wchar_t (16 or 32 bits, depending on the 309 platform) 310 311 - PyUnicode_1BYTE_KIND (1): 312 313 * character type = Py_UCS1 (8 bits, unsigned) 314 * all characters are in the range U+0000-U+00FF (latin1) 315 * if ascii is set, all characters are in the range U+0000-U+007F 316 (ASCII), otherwise at least one character is in the range 317 U+0080-U+00FF 318 319 - PyUnicode_2BYTE_KIND (2): 320 321 * character type = Py_UCS2 (16 bits, unsigned) 322 * all characters are in the range U+0000-U+FFFF (BMP) 323 * at least one character is in the range U+0100-U+FFFF 324 325 - PyUnicode_4BYTE_KIND (4): 326 327 * character type = Py_UCS4 (32 bits, unsigned) 328 * all characters are in the range U+0000-U+10FFFF 329 * at least one character is in the range U+10000-U+10FFFF 330 */ 331 unsigned int kind:3; 332 /* Compact is with respect to the allocation scheme. Compact unicode 333 objects only require one memory block while non-compact objects use 334 one block for the PyUnicodeObject struct and another for its data 335 buffer. */ 336 unsigned int compact:1; 337 /* The string only contains characters in the range U+0000-U+007F (ASCII) 338 and the kind is PyUnicode_1BYTE_KIND. If ascii is set and compact is 339 set, use the PyASCIIObject structure. */ 340 unsigned int ascii:1; 341 /* The ready flag indicates whether the object layout is initialized 342 completely. This means that this is either a compact object, or 343 the data pointer is filled out. The bit is redundant, and helps 344 to minimize the test in PyUnicode_IS_READY(). */ 345 unsigned int ready:1; 346 /* Padding to ensure that PyUnicode_DATA() is always aligned to 347 4 bytes (see issue #19537 on m68k). */ 348 unsigned int :24; 349 } state; 350 wchar_t *wstr; /* wchar_t representation (null-terminated) */ 351} PyASCIIObject; 352 353/* Non-ASCII strings allocated through PyUnicode_New use the 354 PyCompactUnicodeObject structure. state.compact is set, and the data 355 immediately follow the structure. */ 356typedef struct { 357 PyASCIIObject _base; 358 Py_ssize_t utf8_length; /* Number of bytes in utf8, excluding the 359 * terminating \0. */ 360 char *utf8; /* UTF-8 representation (null-terminated) */ 361 Py_ssize_t wstr_length; /* Number of code points in wstr, possible 362 * surrogates count as two code points. */ 363} PyCompactUnicodeObject; 364 365/* Strings allocated through PyUnicode_FromUnicode(NULL, len) use the 366 PyUnicodeObject structure. The actual string data is initially in the wstr 367 block, and copied into the data block using _PyUnicode_Ready. */ 368typedef struct { 369 PyCompactUnicodeObject _base; 370 union { 371 void *any; 372 Py_UCS1 *latin1; 373 Py_UCS2 *ucs2; 374 Py_UCS4 *ucs4; 375 } data; /* Canonical, smallest-form Unicode buffer */ 376} PyUnicodeObject; 377#endif 378 379PyAPI_DATA(PyTypeObject) PyUnicode_Type; 380PyAPI_DATA(PyTypeObject) PyUnicodeIter_Type; 381 382#define PyUnicode_Check(op) \ 383 PyType_FastSubclass(Py_TYPE(op), Py_TPFLAGS_UNICODE_SUBCLASS) 384#define PyUnicode_CheckExact(op) (Py_TYPE(op) == &PyUnicode_Type) 385 386/* Fast access macros */ 387#ifndef Py_LIMITED_API 388 389#define PyUnicode_WSTR_LENGTH(op) \ 390 (PyUnicode_IS_COMPACT_ASCII(op) ? \ 391 ((PyASCIIObject*)op)->length : \ 392 ((PyCompactUnicodeObject*)op)->wstr_length) 393 394/* Returns the deprecated Py_UNICODE representation's size in code units 395 (this includes surrogate pairs as 2 units). 396 If the Py_UNICODE representation is not available, it will be computed 397 on request. Use PyUnicode_GET_LENGTH() for the length in code points. */ 398 399#define PyUnicode_GET_SIZE(op) \ 400 (assert(PyUnicode_Check(op)), \ 401 (((PyASCIIObject *)(op))->wstr) ? \ 402 PyUnicode_WSTR_LENGTH(op) : \ 403 ((void)PyUnicode_AsUnicode((PyObject *)(op)), \ 404 assert(((PyASCIIObject *)(op))->wstr), \ 405 PyUnicode_WSTR_LENGTH(op))) 406 407#define PyUnicode_GET_DATA_SIZE(op) \ 408 (PyUnicode_GET_SIZE(op) * Py_UNICODE_SIZE) 409 410/* Alias for PyUnicode_AsUnicode(). This will create a wchar_t/Py_UNICODE 411 representation on demand. Using this macro is very inefficient now, 412 try to port your code to use the new PyUnicode_*BYTE_DATA() macros or 413 use PyUnicode_WRITE() and PyUnicode_READ(). */ 414 415#define PyUnicode_AS_UNICODE(op) \ 416 (assert(PyUnicode_Check(op)), \ 417 (((PyASCIIObject *)(op))->wstr) ? (((PyASCIIObject *)(op))->wstr) : \ 418 PyUnicode_AsUnicode((PyObject *)(op))) 419 420#define PyUnicode_AS_DATA(op) \ 421 ((const char *)(PyUnicode_AS_UNICODE(op))) 422 423 424/* --- Flexible String Representation Helper Macros (PEP 393) -------------- */ 425 426/* Values for PyASCIIObject.state: */ 427 428/* Interning state. */ 429#define SSTATE_NOT_INTERNED 0 430#define SSTATE_INTERNED_MORTAL 1 431#define SSTATE_INTERNED_IMMORTAL 2 432 433/* Return true if the string contains only ASCII characters, or 0 if not. The 434 string may be compact (PyUnicode_IS_COMPACT_ASCII) or not, but must be 435 ready. */ 436#define PyUnicode_IS_ASCII(op) \ 437 (assert(PyUnicode_Check(op)), \ 438 assert(PyUnicode_IS_READY(op)), \ 439 ((PyASCIIObject*)op)->state.ascii) 440 441/* Return true if the string is compact or 0 if not. 442 No type checks or Ready calls are performed. */ 443#define PyUnicode_IS_COMPACT(op) \ 444 (((PyASCIIObject*)(op))->state.compact) 445 446/* Return true if the string is a compact ASCII string (use PyASCIIObject 447 structure), or 0 if not. No type checks or Ready calls are performed. */ 448#define PyUnicode_IS_COMPACT_ASCII(op) \ 449 (((PyASCIIObject*)op)->state.ascii && PyUnicode_IS_COMPACT(op)) 450 451enum PyUnicode_Kind { 452/* String contains only wstr byte characters. This is only possible 453 when the string was created with a legacy API and _PyUnicode_Ready() 454 has not been called yet. */ 455 PyUnicode_WCHAR_KIND = 0, 456/* Return values of the PyUnicode_KIND() macro: */ 457 PyUnicode_1BYTE_KIND = 1, 458 PyUnicode_2BYTE_KIND = 2, 459 PyUnicode_4BYTE_KIND = 4 460}; 461 462/* Return pointers to the canonical representation cast to unsigned char, 463 Py_UCS2, or Py_UCS4 for direct character access. 464 No checks are performed, use PyUnicode_KIND() before to ensure 465 these will work correctly. */ 466 467#define PyUnicode_1BYTE_DATA(op) ((Py_UCS1*)PyUnicode_DATA(op)) 468#define PyUnicode_2BYTE_DATA(op) ((Py_UCS2*)PyUnicode_DATA(op)) 469#define PyUnicode_4BYTE_DATA(op) ((Py_UCS4*)PyUnicode_DATA(op)) 470 471/* Return one of the PyUnicode_*_KIND values defined above. */ 472#define PyUnicode_KIND(op) \ 473 (assert(PyUnicode_Check(op)), \ 474 assert(PyUnicode_IS_READY(op)), \ 475 ((PyASCIIObject *)(op))->state.kind) 476 477/* Return a void pointer to the raw unicode buffer. */ 478#define _PyUnicode_COMPACT_DATA(op) \ 479 (PyUnicode_IS_ASCII(op) ? \ 480 ((void*)((PyASCIIObject*)(op) + 1)) : \ 481 ((void*)((PyCompactUnicodeObject*)(op) + 1))) 482 483#define _PyUnicode_NONCOMPACT_DATA(op) \ 484 (assert(((PyUnicodeObject*)(op))->data.any), \ 485 ((((PyUnicodeObject *)(op))->data.any))) 486 487#define PyUnicode_DATA(op) \ 488 (assert(PyUnicode_Check(op)), \ 489 PyUnicode_IS_COMPACT(op) ? _PyUnicode_COMPACT_DATA(op) : \ 490 _PyUnicode_NONCOMPACT_DATA(op)) 491 492/* In the access macros below, "kind" may be evaluated more than once. 493 All other macro parameters are evaluated exactly once, so it is safe 494 to put side effects into them (such as increasing the index). */ 495 496/* Write into the canonical representation, this macro does not do any sanity 497 checks and is intended for usage in loops. The caller should cache the 498 kind and data pointers obtained from other macro calls. 499 index is the index in the string (starts at 0) and value is the new 500 code point value which should be written to that location. */ 501#define PyUnicode_WRITE(kind, data, index, value) \ 502 do { \ 503 switch ((kind)) { \ 504 case PyUnicode_1BYTE_KIND: { \ 505 ((Py_UCS1 *)(data))[(index)] = (Py_UCS1)(value); \ 506 break; \ 507 } \ 508 case PyUnicode_2BYTE_KIND: { \ 509 ((Py_UCS2 *)(data))[(index)] = (Py_UCS2)(value); \ 510 break; \ 511 } \ 512 default: { \ 513 assert((kind) == PyUnicode_4BYTE_KIND); \ 514 ((Py_UCS4 *)(data))[(index)] = (Py_UCS4)(value); \ 515 } \ 516 } \ 517 } while (0) 518 519/* Read a code point from the string's canonical representation. No checks 520 or ready calls are performed. */ 521#define PyUnicode_READ(kind, data, index) \ 522 ((Py_UCS4) \ 523 ((kind) == PyUnicode_1BYTE_KIND ? \ 524 ((const Py_UCS1 *)(data))[(index)] : \ 525 ((kind) == PyUnicode_2BYTE_KIND ? \ 526 ((const Py_UCS2 *)(data))[(index)] : \ 527 ((const Py_UCS4 *)(data))[(index)] \ 528 ) \ 529 )) 530 531/* PyUnicode_READ_CHAR() is less efficient than PyUnicode_READ() because it 532 calls PyUnicode_KIND() and might call it twice. For single reads, use 533 PyUnicode_READ_CHAR, for multiple consecutive reads callers should 534 cache kind and use PyUnicode_READ instead. */ 535#define PyUnicode_READ_CHAR(unicode, index) \ 536 (assert(PyUnicode_Check(unicode)), \ 537 assert(PyUnicode_IS_READY(unicode)), \ 538 (Py_UCS4) \ 539 (PyUnicode_KIND((unicode)) == PyUnicode_1BYTE_KIND ? \ 540 ((const Py_UCS1 *)(PyUnicode_DATA((unicode))))[(index)] : \ 541 (PyUnicode_KIND((unicode)) == PyUnicode_2BYTE_KIND ? \ 542 ((const Py_UCS2 *)(PyUnicode_DATA((unicode))))[(index)] : \ 543 ((const Py_UCS4 *)(PyUnicode_DATA((unicode))))[(index)] \ 544 ) \ 545 )) 546 547/* Returns the length of the unicode string. The caller has to make sure that 548 the string has it's canonical representation set before calling 549 this macro. Call PyUnicode_(FAST_)Ready to ensure that. */ 550#define PyUnicode_GET_LENGTH(op) \ 551 (assert(PyUnicode_Check(op)), \ 552 assert(PyUnicode_IS_READY(op)), \ 553 ((PyASCIIObject *)(op))->length) 554 555 556/* Fast check to determine whether an object is ready. Equivalent to 557 PyUnicode_IS_COMPACT(op) || ((PyUnicodeObject*)(op))->data.any) */ 558 559#define PyUnicode_IS_READY(op) (((PyASCIIObject*)op)->state.ready) 560 561/* PyUnicode_READY() does less work than _PyUnicode_Ready() in the best 562 case. If the canonical representation is not yet set, it will still call 563 _PyUnicode_Ready(). 564 Returns 0 on success and -1 on errors. */ 565#define PyUnicode_READY(op) \ 566 (assert(PyUnicode_Check(op)), \ 567 (PyUnicode_IS_READY(op) ? \ 568 0 : _PyUnicode_Ready((PyObject *)(op)))) 569 570/* Return a maximum character value which is suitable for creating another 571 string based on op. This is always an approximation but more efficient 572 than iterating over the string. */ 573#define PyUnicode_MAX_CHAR_VALUE(op) \ 574 (assert(PyUnicode_IS_READY(op)), \ 575 (PyUnicode_IS_ASCII(op) ? \ 576 (0x7f) : \ 577 (PyUnicode_KIND(op) == PyUnicode_1BYTE_KIND ? \ 578 (0xffU) : \ 579 (PyUnicode_KIND(op) == PyUnicode_2BYTE_KIND ? \ 580 (0xffffU) : \ 581 (0x10ffffU))))) 582 583#endif 584 585/* --- Constants ---------------------------------------------------------- */ 586 587/* This Unicode character will be used as replacement character during 588 decoding if the errors argument is set to "replace". Note: the 589 Unicode character U+FFFD is the official REPLACEMENT CHARACTER in 590 Unicode 3.0. */ 591 592#define Py_UNICODE_REPLACEMENT_CHARACTER ((Py_UCS4) 0xFFFD) 593 594/* === Public API ========================================================= */ 595 596/* --- Plain Py_UNICODE --------------------------------------------------- */ 597 598/* With PEP 393, this is the recommended way to allocate a new unicode object. 599 This function will allocate the object and its buffer in a single memory 600 block. Objects created using this function are not resizable. */ 601#ifndef Py_LIMITED_API 602PyAPI_FUNC(PyObject*) PyUnicode_New( 603 Py_ssize_t size, /* Number of code points in the new string */ 604 Py_UCS4 maxchar /* maximum code point value in the string */ 605 ); 606#endif 607 608/* Initializes the canonical string representation from the deprecated 609 wstr/Py_UNICODE representation. This function is used to convert Unicode 610 objects which were created using the old API to the new flexible format 611 introduced with PEP 393. 612 613 Don't call this function directly, use the public PyUnicode_READY() macro 614 instead. */ 615#ifndef Py_LIMITED_API 616PyAPI_FUNC(int) _PyUnicode_Ready( 617 PyObject *unicode /* Unicode object */ 618 ); 619#endif 620 621/* Get a copy of a Unicode string. */ 622#ifndef Py_LIMITED_API 623PyAPI_FUNC(PyObject*) _PyUnicode_Copy( 624 PyObject *unicode 625 ); 626#endif 627 628/* Copy character from one unicode object into another, this function performs 629 character conversion when necessary and falls back to memcpy() if possible. 630 631 Fail if to is too small (smaller than *how_many* or smaller than 632 len(from)-from_start), or if kind(from[from_start:from_start+how_many]) > 633 kind(to), or if *to* has more than 1 reference. 634 635 Return the number of written character, or return -1 and raise an exception 636 on error. 637 638 Pseudo-code: 639 640 how_many = min(how_many, len(from) - from_start) 641 to[to_start:to_start+how_many] = from[from_start:from_start+how_many] 642 return how_many 643 644 Note: The function doesn't write a terminating null character. 645 */ 646#ifndef Py_LIMITED_API 647PyAPI_FUNC(Py_ssize_t) PyUnicode_CopyCharacters( 648 PyObject *to, 649 Py_ssize_t to_start, 650 PyObject *from, 651 Py_ssize_t from_start, 652 Py_ssize_t how_many 653 ); 654 655/* Unsafe version of PyUnicode_CopyCharacters(): don't check arguments and so 656 may crash if parameters are invalid (e.g. if the output string 657 is too short). */ 658PyAPI_FUNC(void) _PyUnicode_FastCopyCharacters( 659 PyObject *to, 660 Py_ssize_t to_start, 661 PyObject *from, 662 Py_ssize_t from_start, 663 Py_ssize_t how_many 664 ); 665#endif 666 667#ifndef Py_LIMITED_API 668/* Fill a string with a character: write fill_char into 669 unicode[start:start+length]. 670 671 Fail if fill_char is bigger than the string maximum character, or if the 672 string has more than 1 reference. 673 674 Return the number of written character, or return -1 and raise an exception 675 on error. */ 676PyAPI_FUNC(Py_ssize_t) PyUnicode_Fill( 677 PyObject *unicode, 678 Py_ssize_t start, 679 Py_ssize_t length, 680 Py_UCS4 fill_char 681 ); 682 683/* Unsafe version of PyUnicode_Fill(): don't check arguments and so may crash 684 if parameters are invalid (e.g. if length is longer than the string). */ 685PyAPI_FUNC(void) _PyUnicode_FastFill( 686 PyObject *unicode, 687 Py_ssize_t start, 688 Py_ssize_t length, 689 Py_UCS4 fill_char 690 ); 691#endif 692 693/* Create a Unicode Object from the Py_UNICODE buffer u of the given 694 size. 695 696 u may be NULL which causes the contents to be undefined. It is the 697 user's responsibility to fill in the needed data afterwards. Note 698 that modifying the Unicode object contents after construction is 699 only allowed if u was set to NULL. 700 701 The buffer is copied into the new object. */ 702 703#ifndef Py_LIMITED_API 704PyAPI_FUNC(PyObject*) PyUnicode_FromUnicode( 705 const Py_UNICODE *u, /* Unicode buffer */ 706 Py_ssize_t size /* size of buffer */ 707 ); 708#endif 709 710/* Similar to PyUnicode_FromUnicode(), but u points to UTF-8 encoded bytes */ 711PyAPI_FUNC(PyObject*) PyUnicode_FromStringAndSize( 712 const char *u, /* UTF-8 encoded string */ 713 Py_ssize_t size /* size of buffer */ 714 ); 715 716/* Similar to PyUnicode_FromUnicode(), but u points to null-terminated 717 UTF-8 encoded bytes. The size is determined with strlen(). */ 718PyAPI_FUNC(PyObject*) PyUnicode_FromString( 719 const char *u /* UTF-8 encoded string */ 720 ); 721 722#ifndef Py_LIMITED_API 723/* Create a new string from a buffer of Py_UCS1, Py_UCS2 or Py_UCS4 characters. 724 Scan the string to find the maximum character. */ 725PyAPI_FUNC(PyObject*) PyUnicode_FromKindAndData( 726 int kind, 727 const void *buffer, 728 Py_ssize_t size); 729 730/* Create a new string from a buffer of ASCII characters. 731 WARNING: Don't check if the string contains any non-ASCII character. */ 732PyAPI_FUNC(PyObject*) _PyUnicode_FromASCII( 733 const char *buffer, 734 Py_ssize_t size); 735#endif 736 737PyAPI_FUNC(PyObject*) PyUnicode_Substring( 738 PyObject *str, 739 Py_ssize_t start, 740 Py_ssize_t end); 741 742#ifndef Py_LIMITED_API 743/* Compute the maximum character of the substring unicode[start:end]. 744 Return 127 for an empty string. */ 745PyAPI_FUNC(Py_UCS4) _PyUnicode_FindMaxChar ( 746 PyObject *unicode, 747 Py_ssize_t start, 748 Py_ssize_t end); 749#endif 750 751/* Copy the string into a UCS4 buffer including the null character if copy_null 752 is set. Return NULL and raise an exception on error. Raise a SystemError if 753 the buffer is smaller than the string. Return buffer on success. 754 755 buflen is the length of the buffer in (Py_UCS4) characters. */ 756PyAPI_FUNC(Py_UCS4*) PyUnicode_AsUCS4( 757 PyObject *unicode, 758 Py_UCS4* buffer, 759 Py_ssize_t buflen, 760 int copy_null); 761 762/* Copy the string into a UCS4 buffer. A new buffer is allocated using 763 * PyMem_Malloc; if this fails, NULL is returned with a memory error 764 exception set. */ 765PyAPI_FUNC(Py_UCS4*) PyUnicode_AsUCS4Copy(PyObject *unicode); 766 767/* Return a read-only pointer to the Unicode object's internal 768 Py_UNICODE buffer. 769 If the wchar_t/Py_UNICODE representation is not yet available, this 770 function will calculate it. */ 771 772#ifndef Py_LIMITED_API 773PyAPI_FUNC(Py_UNICODE *) PyUnicode_AsUnicode( 774 PyObject *unicode /* Unicode object */ 775 ); 776#endif 777 778/* Return a read-only pointer to the Unicode object's internal 779 Py_UNICODE buffer and save the length at size. 780 If the wchar_t/Py_UNICODE representation is not yet available, this 781 function will calculate it. */ 782 783#ifndef Py_LIMITED_API 784PyAPI_FUNC(Py_UNICODE *) PyUnicode_AsUnicodeAndSize( 785 PyObject *unicode, /* Unicode object */ 786 Py_ssize_t *size /* location where to save the length */ 787 ); 788#endif 789 790/* Get the length of the Unicode object. */ 791 792PyAPI_FUNC(Py_ssize_t) PyUnicode_GetLength( 793 PyObject *unicode 794); 795 796/* Get the number of Py_UNICODE units in the 797 string representation. */ 798 799PyAPI_FUNC(Py_ssize_t) PyUnicode_GetSize( 800 PyObject *unicode /* Unicode object */ 801 ); 802 803/* Read a character from the string. */ 804 805PyAPI_FUNC(Py_UCS4) PyUnicode_ReadChar( 806 PyObject *unicode, 807 Py_ssize_t index 808 ); 809 810/* Write a character to the string. The string must have been created through 811 PyUnicode_New, must not be shared, and must not have been hashed yet. 812 813 Return 0 on success, -1 on error. */ 814 815PyAPI_FUNC(int) PyUnicode_WriteChar( 816 PyObject *unicode, 817 Py_ssize_t index, 818 Py_UCS4 character 819 ); 820 821#ifndef Py_LIMITED_API 822/* Get the maximum ordinal for a Unicode character. */ 823PyAPI_FUNC(Py_UNICODE) PyUnicode_GetMax(void); 824#endif 825 826/* Resize a Unicode object. The length is the number of characters, except 827 if the kind of the string is PyUnicode_WCHAR_KIND: in this case, the length 828 is the number of Py_UNICODE characters. 829 830 *unicode is modified to point to the new (resized) object and 0 831 returned on success. 832 833 Try to resize the string in place (which is usually faster than allocating 834 a new string and copy characters), or create a new string. 835 836 Error handling is implemented as follows: an exception is set, -1 837 is returned and *unicode left untouched. 838 839 WARNING: The function doesn't check string content, the result may not be a 840 string in canonical representation. */ 841 842PyAPI_FUNC(int) PyUnicode_Resize( 843 PyObject **unicode, /* Pointer to the Unicode object */ 844 Py_ssize_t length /* New length */ 845 ); 846 847/* Decode obj to a Unicode object. 848 849 bytes, bytearray and other bytes-like objects are decoded according to the 850 given encoding and error handler. The encoding and error handler can be 851 NULL to have the interface use UTF-8 and "strict". 852 853 All other objects (including Unicode objects) raise an exception. 854 855 The API returns NULL in case of an error. The caller is responsible 856 for decref'ing the returned objects. 857 858*/ 859 860PyAPI_FUNC(PyObject*) PyUnicode_FromEncodedObject( 861 PyObject *obj, /* Object */ 862 const char *encoding, /* encoding */ 863 const char *errors /* error handling */ 864 ); 865 866/* Copy an instance of a Unicode subtype to a new true Unicode object if 867 necessary. If obj is already a true Unicode object (not a subtype), return 868 the reference with *incremented* refcount. 869 870 The API returns NULL in case of an error. The caller is responsible 871 for decref'ing the returned objects. 872 873*/ 874 875PyAPI_FUNC(PyObject*) PyUnicode_FromObject( 876 PyObject *obj /* Object */ 877 ); 878 879PyAPI_FUNC(PyObject *) PyUnicode_FromFormatV( 880 const char *format, /* ASCII-encoded string */ 881 va_list vargs 882 ); 883PyAPI_FUNC(PyObject *) PyUnicode_FromFormat( 884 const char *format, /* ASCII-encoded string */ 885 ... 886 ); 887 888#ifndef Py_LIMITED_API 889typedef struct { 890 PyObject *buffer; 891 void *data; 892 enum PyUnicode_Kind kind; 893 Py_UCS4 maxchar; 894 Py_ssize_t size; 895 Py_ssize_t pos; 896 897 /* minimum number of allocated characters (default: 0) */ 898 Py_ssize_t min_length; 899 900 /* minimum character (default: 127, ASCII) */ 901 Py_UCS4 min_char; 902 903 /* If non-zero, overallocate the buffer by 25% (default: 0). */ 904 unsigned char overallocate; 905 906 /* If readonly is 1, buffer is a shared string (cannot be modified) 907 and size is set to 0. */ 908 unsigned char readonly; 909} _PyUnicodeWriter ; 910 911/* Initialize a Unicode writer. 912 * 913 * By default, the minimum buffer size is 0 character and overallocation is 914 * disabled. Set min_length, min_char and overallocate attributes to control 915 * the allocation of the buffer. */ 916PyAPI_FUNC(void) 917_PyUnicodeWriter_Init(_PyUnicodeWriter *writer); 918 919/* Prepare the buffer to write 'length' characters 920 with the specified maximum character. 921 922 Return 0 on success, raise an exception and return -1 on error. */ 923#define _PyUnicodeWriter_Prepare(WRITER, LENGTH, MAXCHAR) \ 924 (((MAXCHAR) <= (WRITER)->maxchar \ 925 && (LENGTH) <= (WRITER)->size - (WRITER)->pos) \ 926 ? 0 \ 927 : (((LENGTH) == 0) \ 928 ? 0 \ 929 : _PyUnicodeWriter_PrepareInternal((WRITER), (LENGTH), (MAXCHAR)))) 930 931/* Don't call this function directly, use the _PyUnicodeWriter_Prepare() macro 932 instead. */ 933PyAPI_FUNC(int) 934_PyUnicodeWriter_PrepareInternal(_PyUnicodeWriter *writer, 935 Py_ssize_t length, Py_UCS4 maxchar); 936 937/* Append a Unicode character. 938 Return 0 on success, raise an exception and return -1 on error. */ 939PyAPI_FUNC(int) 940_PyUnicodeWriter_WriteChar(_PyUnicodeWriter *writer, 941 Py_UCS4 ch 942 ); 943 944/* Append a Unicode string. 945 Return 0 on success, raise an exception and return -1 on error. */ 946PyAPI_FUNC(int) 947_PyUnicodeWriter_WriteStr(_PyUnicodeWriter *writer, 948 PyObject *str /* Unicode string */ 949 ); 950 951/* Append a substring of a Unicode string. 952 Return 0 on success, raise an exception and return -1 on error. */ 953PyAPI_FUNC(int) 954_PyUnicodeWriter_WriteSubstring(_PyUnicodeWriter *writer, 955 PyObject *str, /* Unicode string */ 956 Py_ssize_t start, 957 Py_ssize_t end 958 ); 959 960/* Append an ASCII-encoded byte string. 961 Return 0 on success, raise an exception and return -1 on error. */ 962PyAPI_FUNC(int) 963_PyUnicodeWriter_WriteASCIIString(_PyUnicodeWriter *writer, 964 const char *str, /* ASCII-encoded byte string */ 965 Py_ssize_t len /* number of bytes, or -1 if unknown */ 966 ); 967 968/* Append a latin1-encoded byte string. 969 Return 0 on success, raise an exception and return -1 on error. */ 970PyAPI_FUNC(int) 971_PyUnicodeWriter_WriteLatin1String(_PyUnicodeWriter *writer, 972 const char *str, /* latin1-encoded byte string */ 973 Py_ssize_t len /* length in bytes */ 974 ); 975 976/* Get the value of the writer as a Unicode string. Clear the 977 buffer of the writer. Raise an exception and return NULL 978 on error. */ 979PyAPI_FUNC(PyObject *) 980_PyUnicodeWriter_Finish(_PyUnicodeWriter *writer); 981 982/* Deallocate memory of a writer (clear its internal buffer). */ 983PyAPI_FUNC(void) 984_PyUnicodeWriter_Dealloc(_PyUnicodeWriter *writer); 985#endif 986 987#ifndef Py_LIMITED_API 988/* Format the object based on the format_spec, as defined in PEP 3101 989 (Advanced String Formatting). */ 990PyAPI_FUNC(int) _PyUnicode_FormatAdvancedWriter( 991 _PyUnicodeWriter *writer, 992 PyObject *obj, 993 PyObject *format_spec, 994 Py_ssize_t start, 995 Py_ssize_t end); 996#endif 997 998PyAPI_FUNC(void) PyUnicode_InternInPlace(PyObject **); 999PyAPI_FUNC(void) PyUnicode_InternImmortal(PyObject **); 1000PyAPI_FUNC(PyObject *) PyUnicode_InternFromString( 1001 const char *u /* UTF-8 encoded string */ 1002 ); 1003#ifndef Py_LIMITED_API 1004PyAPI_FUNC(void) _Py_ReleaseInternedUnicodeStrings(void); 1005#endif 1006 1007/* Use only if you know it's a string */ 1008#define PyUnicode_CHECK_INTERNED(op) \ 1009 (((PyASCIIObject *)(op))->state.interned) 1010 1011/* --- wchar_t support for platforms which support it --------------------- */ 1012 1013#ifdef HAVE_WCHAR_H 1014 1015/* Create a Unicode Object from the wchar_t buffer w of the given 1016 size. 1017 1018 The buffer is copied into the new object. */ 1019 1020PyAPI_FUNC(PyObject*) PyUnicode_FromWideChar( 1021 const wchar_t *w, /* wchar_t buffer */ 1022 Py_ssize_t size /* size of buffer */ 1023 ); 1024 1025/* Copies the Unicode Object contents into the wchar_t buffer w. At 1026 most size wchar_t characters are copied. 1027 1028 Note that the resulting wchar_t string may or may not be 1029 0-terminated. It is the responsibility of the caller to make sure 1030 that the wchar_t string is 0-terminated in case this is required by 1031 the application. 1032 1033 Returns the number of wchar_t characters copied (excluding a 1034 possibly trailing 0-termination character) or -1 in case of an 1035 error. */ 1036 1037PyAPI_FUNC(Py_ssize_t) PyUnicode_AsWideChar( 1038 PyObject *unicode, /* Unicode object */ 1039 wchar_t *w, /* wchar_t buffer */ 1040 Py_ssize_t size /* size of buffer */ 1041 ); 1042 1043/* Convert the Unicode object to a wide character string. The output string 1044 always ends with a nul character. If size is not NULL, write the number of 1045 wide characters (excluding the null character) into *size. 1046 1047 Returns a buffer allocated by PyMem_Malloc() (use PyMem_Free() to free it) 1048 on success. On error, returns NULL, *size is undefined and raises a 1049 MemoryError. */ 1050 1051PyAPI_FUNC(wchar_t*) PyUnicode_AsWideCharString( 1052 PyObject *unicode, /* Unicode object */ 1053 Py_ssize_t *size /* number of characters of the result */ 1054 ); 1055 1056#ifndef Py_LIMITED_API 1057PyAPI_FUNC(void*) _PyUnicode_AsKind(PyObject *s, unsigned int kind); 1058#endif 1059 1060#endif 1061 1062/* --- Unicode ordinals --------------------------------------------------- */ 1063 1064/* Create a Unicode Object from the given Unicode code point ordinal. 1065 1066 The ordinal must be in range(0x110000). A ValueError is 1067 raised in case it is not. 1068 1069*/ 1070 1071PyAPI_FUNC(PyObject*) PyUnicode_FromOrdinal(int ordinal); 1072 1073/* --- Free-list management ----------------------------------------------- */ 1074 1075/* Clear the free list used by the Unicode implementation. 1076 1077 This can be used to release memory used for objects on the free 1078 list back to the Python memory allocator. 1079 1080*/ 1081 1082PyAPI_FUNC(int) PyUnicode_ClearFreeList(void); 1083 1084/* === Builtin Codecs ===================================================== 1085 1086 Many of these APIs take two arguments encoding and errors. These 1087 parameters encoding and errors have the same semantics as the ones 1088 of the builtin str() API. 1089 1090 Setting encoding to NULL causes the default encoding (UTF-8) to be used. 1091 1092 Error handling is set by errors which may also be set to NULL 1093 meaning to use the default handling defined for the codec. Default 1094 error handling for all builtin codecs is "strict" (ValueErrors are 1095 raised). 1096 1097 The codecs all use a similar interface. Only deviation from the 1098 generic ones are documented. 1099 1100*/ 1101 1102/* --- Manage the default encoding ---------------------------------------- */ 1103 1104/* Returns a pointer to the default encoding (UTF-8) of the 1105 Unicode object unicode and the size of the encoded representation 1106 in bytes stored in *size. 1107 1108 In case of an error, no *size is set. 1109 1110 This function caches the UTF-8 encoded string in the unicodeobject 1111 and subsequent calls will return the same string. The memory is released 1112 when the unicodeobject is deallocated. 1113 1114 _PyUnicode_AsStringAndSize is a #define for PyUnicode_AsUTF8AndSize to 1115 support the previous internal function with the same behaviour. 1116 1117 *** This API is for interpreter INTERNAL USE ONLY and will likely 1118 *** be removed or changed in the future. 1119 1120 *** If you need to access the Unicode object as UTF-8 bytes string, 1121 *** please use PyUnicode_AsUTF8String() instead. 1122*/ 1123 1124#ifndef Py_LIMITED_API 1125PyAPI_FUNC(char *) PyUnicode_AsUTF8AndSize( 1126 PyObject *unicode, 1127 Py_ssize_t *size); 1128#define _PyUnicode_AsStringAndSize PyUnicode_AsUTF8AndSize 1129#endif 1130 1131/* Returns a pointer to the default encoding (UTF-8) of the 1132 Unicode object unicode. 1133 1134 Like PyUnicode_AsUTF8AndSize(), this also caches the UTF-8 representation 1135 in the unicodeobject. 1136 1137 _PyUnicode_AsString is a #define for PyUnicode_AsUTF8 to 1138 support the previous internal function with the same behaviour. 1139 1140 Use of this API is DEPRECATED since no size information can be 1141 extracted from the returned data. 1142 1143 *** This API is for interpreter INTERNAL USE ONLY and will likely 1144 *** be removed or changed for Python 3.1. 1145 1146 *** If you need to access the Unicode object as UTF-8 bytes string, 1147 *** please use PyUnicode_AsUTF8String() instead. 1148 1149*/ 1150 1151#ifndef Py_LIMITED_API 1152PyAPI_FUNC(char *) PyUnicode_AsUTF8(PyObject *unicode); 1153#define _PyUnicode_AsString PyUnicode_AsUTF8 1154#endif 1155 1156/* Returns "utf-8". */ 1157 1158PyAPI_FUNC(const char*) PyUnicode_GetDefaultEncoding(void); 1159 1160/* --- Generic Codecs ----------------------------------------------------- */ 1161 1162/* Create a Unicode object by decoding the encoded string s of the 1163 given size. */ 1164 1165PyAPI_FUNC(PyObject*) PyUnicode_Decode( 1166 const char *s, /* encoded string */ 1167 Py_ssize_t size, /* size of buffer */ 1168 const char *encoding, /* encoding */ 1169 const char *errors /* error handling */ 1170 ); 1171 1172/* Decode a Unicode object unicode and return the result as Python 1173 object. */ 1174 1175PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedObject( 1176 PyObject *unicode, /* Unicode object */ 1177 const char *encoding, /* encoding */ 1178 const char *errors /* error handling */ 1179 ); 1180 1181/* Decode a Unicode object unicode and return the result as Unicode 1182 object. */ 1183 1184PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedUnicode( 1185 PyObject *unicode, /* Unicode object */ 1186 const char *encoding, /* encoding */ 1187 const char *errors /* error handling */ 1188 ); 1189 1190/* Encodes a Py_UNICODE buffer of the given size and returns a 1191 Python string object. */ 1192 1193#ifndef Py_LIMITED_API 1194PyAPI_FUNC(PyObject*) PyUnicode_Encode( 1195 const Py_UNICODE *s, /* Unicode char buffer */ 1196 Py_ssize_t size, /* number of Py_UNICODE chars to encode */ 1197 const char *encoding, /* encoding */ 1198 const char *errors /* error handling */ 1199 ); 1200#endif 1201 1202/* Encodes a Unicode object and returns the result as Python 1203 object. */ 1204 1205PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedObject( 1206 PyObject *unicode, /* Unicode object */ 1207 const char *encoding, /* encoding */ 1208 const char *errors /* error handling */ 1209 ); 1210 1211/* Encodes a Unicode object and returns the result as Python string 1212 object. */ 1213 1214PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedString( 1215 PyObject *unicode, /* Unicode object */ 1216 const char *encoding, /* encoding */ 1217 const char *errors /* error handling */ 1218 ); 1219 1220/* Encodes a Unicode object and returns the result as Unicode 1221 object. */ 1222 1223PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedUnicode( 1224 PyObject *unicode, /* Unicode object */ 1225 const char *encoding, /* encoding */ 1226 const char *errors /* error handling */ 1227 ); 1228 1229/* Build an encoding map. */ 1230 1231PyAPI_FUNC(PyObject*) PyUnicode_BuildEncodingMap( 1232 PyObject* string /* 256 character map */ 1233 ); 1234 1235/* --- UTF-7 Codecs ------------------------------------------------------- */ 1236 1237PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7( 1238 const char *string, /* UTF-7 encoded string */ 1239 Py_ssize_t length, /* size of string */ 1240 const char *errors /* error handling */ 1241 ); 1242 1243PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7Stateful( 1244 const char *string, /* UTF-7 encoded string */ 1245 Py_ssize_t length, /* size of string */ 1246 const char *errors, /* error handling */ 1247 Py_ssize_t *consumed /* bytes consumed */ 1248 ); 1249 1250#ifndef Py_LIMITED_API 1251PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF7( 1252 const Py_UNICODE *data, /* Unicode char buffer */ 1253 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1254 int base64SetO, /* Encode RFC2152 Set O characters in base64 */ 1255 int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ 1256 const char *errors /* error handling */ 1257 ); 1258PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF7( 1259 PyObject *unicode, /* Unicode object */ 1260 int base64SetO, /* Encode RFC2152 Set O characters in base64 */ 1261 int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ 1262 const char *errors /* error handling */ 1263 ); 1264#endif 1265 1266/* --- UTF-8 Codecs ------------------------------------------------------- */ 1267 1268PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8( 1269 const char *string, /* UTF-8 encoded string */ 1270 Py_ssize_t length, /* size of string */ 1271 const char *errors /* error handling */ 1272 ); 1273 1274PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8Stateful( 1275 const char *string, /* UTF-8 encoded string */ 1276 Py_ssize_t length, /* size of string */ 1277 const char *errors, /* error handling */ 1278 Py_ssize_t *consumed /* bytes consumed */ 1279 ); 1280 1281PyAPI_FUNC(PyObject*) PyUnicode_AsUTF8String( 1282 PyObject *unicode /* Unicode object */ 1283 ); 1284 1285#ifndef Py_LIMITED_API 1286PyAPI_FUNC(PyObject*) _PyUnicode_AsUTF8String( 1287 PyObject *unicode, 1288 const char *errors); 1289 1290PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF8( 1291 const Py_UNICODE *data, /* Unicode char buffer */ 1292 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1293 const char *errors /* error handling */ 1294 ); 1295#endif 1296 1297/* --- UTF-32 Codecs ------------------------------------------------------ */ 1298 1299/* Decodes length bytes from a UTF-32 encoded buffer string and returns 1300 the corresponding Unicode object. 1301 1302 errors (if non-NULL) defines the error handling. It defaults 1303 to "strict". 1304 1305 If byteorder is non-NULL, the decoder starts decoding using the 1306 given byte order: 1307 1308 *byteorder == -1: little endian 1309 *byteorder == 0: native order 1310 *byteorder == 1: big endian 1311 1312 In native mode, the first four bytes of the stream are checked for a 1313 BOM mark. If found, the BOM mark is analysed, the byte order 1314 adjusted and the BOM skipped. In the other modes, no BOM mark 1315 interpretation is done. After completion, *byteorder is set to the 1316 current byte order at the end of input data. 1317 1318 If byteorder is NULL, the codec starts in native order mode. 1319 1320*/ 1321 1322PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32( 1323 const char *string, /* UTF-32 encoded string */ 1324 Py_ssize_t length, /* size of string */ 1325 const char *errors, /* error handling */ 1326 int *byteorder /* pointer to byteorder to use 1327 0=native;-1=LE,1=BE; updated on 1328 exit */ 1329 ); 1330 1331PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32Stateful( 1332 const char *string, /* UTF-32 encoded string */ 1333 Py_ssize_t length, /* size of string */ 1334 const char *errors, /* error handling */ 1335 int *byteorder, /* pointer to byteorder to use 1336 0=native;-1=LE,1=BE; updated on 1337 exit */ 1338 Py_ssize_t *consumed /* bytes consumed */ 1339 ); 1340 1341/* Returns a Python string using the UTF-32 encoding in native byte 1342 order. The string always starts with a BOM mark. */ 1343 1344PyAPI_FUNC(PyObject*) PyUnicode_AsUTF32String( 1345 PyObject *unicode /* Unicode object */ 1346 ); 1347 1348/* Returns a Python string object holding the UTF-32 encoded value of 1349 the Unicode data. 1350 1351 If byteorder is not 0, output is written according to the following 1352 byte order: 1353 1354 byteorder == -1: little endian 1355 byteorder == 0: native byte order (writes a BOM mark) 1356 byteorder == 1: big endian 1357 1358 If byteorder is 0, the output string will always start with the 1359 Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is 1360 prepended. 1361 1362*/ 1363 1364#ifndef Py_LIMITED_API 1365PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF32( 1366 const Py_UNICODE *data, /* Unicode char buffer */ 1367 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1368 const char *errors, /* error handling */ 1369 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1370 ); 1371PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF32( 1372 PyObject *object, /* Unicode object */ 1373 const char *errors, /* error handling */ 1374 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1375 ); 1376#endif 1377 1378/* --- UTF-16 Codecs ------------------------------------------------------ */ 1379 1380/* Decodes length bytes from a UTF-16 encoded buffer string and returns 1381 the corresponding Unicode object. 1382 1383 errors (if non-NULL) defines the error handling. It defaults 1384 to "strict". 1385 1386 If byteorder is non-NULL, the decoder starts decoding using the 1387 given byte order: 1388 1389 *byteorder == -1: little endian 1390 *byteorder == 0: native order 1391 *byteorder == 1: big endian 1392 1393 In native mode, the first two bytes of the stream are checked for a 1394 BOM mark. If found, the BOM mark is analysed, the byte order 1395 adjusted and the BOM skipped. In the other modes, no BOM mark 1396 interpretation is done. After completion, *byteorder is set to the 1397 current byte order at the end of input data. 1398 1399 If byteorder is NULL, the codec starts in native order mode. 1400 1401*/ 1402 1403PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16( 1404 const char *string, /* UTF-16 encoded string */ 1405 Py_ssize_t length, /* size of string */ 1406 const char *errors, /* error handling */ 1407 int *byteorder /* pointer to byteorder to use 1408 0=native;-1=LE,1=BE; updated on 1409 exit */ 1410 ); 1411 1412PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16Stateful( 1413 const char *string, /* UTF-16 encoded string */ 1414 Py_ssize_t length, /* size of string */ 1415 const char *errors, /* error handling */ 1416 int *byteorder, /* pointer to byteorder to use 1417 0=native;-1=LE,1=BE; updated on 1418 exit */ 1419 Py_ssize_t *consumed /* bytes consumed */ 1420 ); 1421 1422/* Returns a Python string using the UTF-16 encoding in native byte 1423 order. The string always starts with a BOM mark. */ 1424 1425PyAPI_FUNC(PyObject*) PyUnicode_AsUTF16String( 1426 PyObject *unicode /* Unicode object */ 1427 ); 1428 1429/* Returns a Python string object holding the UTF-16 encoded value of 1430 the Unicode data. 1431 1432 If byteorder is not 0, output is written according to the following 1433 byte order: 1434 1435 byteorder == -1: little endian 1436 byteorder == 0: native byte order (writes a BOM mark) 1437 byteorder == 1: big endian 1438 1439 If byteorder is 0, the output string will always start with the 1440 Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is 1441 prepended. 1442 1443 Note that Py_UNICODE data is being interpreted as UTF-16 reduced to 1444 UCS-2. This trick makes it possible to add full UTF-16 capabilities 1445 at a later point without compromising the APIs. 1446 1447*/ 1448 1449#ifndef Py_LIMITED_API 1450PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF16( 1451 const Py_UNICODE *data, /* Unicode char buffer */ 1452 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1453 const char *errors, /* error handling */ 1454 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1455 ); 1456PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF16( 1457 PyObject* unicode, /* Unicode object */ 1458 const char *errors, /* error handling */ 1459 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1460 ); 1461#endif 1462 1463/* --- Unicode-Escape Codecs ---------------------------------------------- */ 1464 1465PyAPI_FUNC(PyObject*) PyUnicode_DecodeUnicodeEscape( 1466 const char *string, /* Unicode-Escape encoded string */ 1467 Py_ssize_t length, /* size of string */ 1468 const char *errors /* error handling */ 1469 ); 1470 1471PyAPI_FUNC(PyObject*) PyUnicode_AsUnicodeEscapeString( 1472 PyObject *unicode /* Unicode object */ 1473 ); 1474 1475#ifndef Py_LIMITED_API 1476PyAPI_FUNC(PyObject*) PyUnicode_EncodeUnicodeEscape( 1477 const Py_UNICODE *data, /* Unicode char buffer */ 1478 Py_ssize_t length /* Number of Py_UNICODE chars to encode */ 1479 ); 1480#endif 1481 1482/* --- Raw-Unicode-Escape Codecs ------------------------------------------ */ 1483 1484PyAPI_FUNC(PyObject*) PyUnicode_DecodeRawUnicodeEscape( 1485 const char *string, /* Raw-Unicode-Escape encoded string */ 1486 Py_ssize_t length, /* size of string */ 1487 const char *errors /* error handling */ 1488 ); 1489 1490PyAPI_FUNC(PyObject*) PyUnicode_AsRawUnicodeEscapeString( 1491 PyObject *unicode /* Unicode object */ 1492 ); 1493 1494#ifndef Py_LIMITED_API 1495PyAPI_FUNC(PyObject*) PyUnicode_EncodeRawUnicodeEscape( 1496 const Py_UNICODE *data, /* Unicode char buffer */ 1497 Py_ssize_t length /* Number of Py_UNICODE chars to encode */ 1498 ); 1499#endif 1500 1501/* --- Unicode Internal Codec --------------------------------------------- 1502 1503 Only for internal use in _codecsmodule.c */ 1504 1505#ifndef Py_LIMITED_API 1506PyObject *_PyUnicode_DecodeUnicodeInternal( 1507 const char *string, 1508 Py_ssize_t length, 1509 const char *errors 1510 ); 1511#endif 1512 1513/* --- Latin-1 Codecs ----------------------------------------------------- 1514 1515 Note: Latin-1 corresponds to the first 256 Unicode ordinals. 1516 1517*/ 1518 1519PyAPI_FUNC(PyObject*) PyUnicode_DecodeLatin1( 1520 const char *string, /* Latin-1 encoded string */ 1521 Py_ssize_t length, /* size of string */ 1522 const char *errors /* error handling */ 1523 ); 1524 1525PyAPI_FUNC(PyObject*) PyUnicode_AsLatin1String( 1526 PyObject *unicode /* Unicode object */ 1527 ); 1528 1529#ifndef Py_LIMITED_API 1530PyAPI_FUNC(PyObject*) _PyUnicode_AsLatin1String( 1531 PyObject* unicode, 1532 const char* errors); 1533 1534PyAPI_FUNC(PyObject*) PyUnicode_EncodeLatin1( 1535 const Py_UNICODE *data, /* Unicode char buffer */ 1536 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1537 const char *errors /* error handling */ 1538 ); 1539#endif 1540 1541/* --- ASCII Codecs ------------------------------------------------------- 1542 1543 Only 7-bit ASCII data is excepted. All other codes generate errors. 1544 1545*/ 1546 1547PyAPI_FUNC(PyObject*) PyUnicode_DecodeASCII( 1548 const char *string, /* ASCII encoded string */ 1549 Py_ssize_t length, /* size of string */ 1550 const char *errors /* error handling */ 1551 ); 1552 1553PyAPI_FUNC(PyObject*) PyUnicode_AsASCIIString( 1554 PyObject *unicode /* Unicode object */ 1555 ); 1556 1557#ifndef Py_LIMITED_API 1558PyAPI_FUNC(PyObject*) _PyUnicode_AsASCIIString( 1559 PyObject* unicode, 1560 const char* errors); 1561 1562PyAPI_FUNC(PyObject*) PyUnicode_EncodeASCII( 1563 const Py_UNICODE *data, /* Unicode char buffer */ 1564 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1565 const char *errors /* error handling */ 1566 ); 1567#endif 1568 1569/* --- Character Map Codecs ----------------------------------------------- 1570 1571 This codec uses mappings to encode and decode characters. 1572 1573 Decoding mappings must map single string characters to single 1574 Unicode characters, integers (which are then interpreted as Unicode 1575 ordinals) or None (meaning "undefined mapping" and causing an 1576 error). 1577 1578 Encoding mappings must map single Unicode characters to single 1579 string characters, integers (which are then interpreted as Latin-1 1580 ordinals) or None (meaning "undefined mapping" and causing an 1581 error). 1582 1583 If a character lookup fails with a LookupError, the character is 1584 copied as-is meaning that its ordinal value will be interpreted as 1585 Unicode or Latin-1 ordinal resp. Because of this mappings only need 1586 to contain those mappings which map characters to different code 1587 points. 1588 1589*/ 1590 1591PyAPI_FUNC(PyObject*) PyUnicode_DecodeCharmap( 1592 const char *string, /* Encoded string */ 1593 Py_ssize_t length, /* size of string */ 1594 PyObject *mapping, /* character mapping 1595 (char ordinal -> unicode ordinal) */ 1596 const char *errors /* error handling */ 1597 ); 1598 1599PyAPI_FUNC(PyObject*) PyUnicode_AsCharmapString( 1600 PyObject *unicode, /* Unicode object */ 1601 PyObject *mapping /* character mapping 1602 (unicode ordinal -> char ordinal) */ 1603 ); 1604 1605#ifndef Py_LIMITED_API 1606PyAPI_FUNC(PyObject*) PyUnicode_EncodeCharmap( 1607 const Py_UNICODE *data, /* Unicode char buffer */ 1608 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1609 PyObject *mapping, /* character mapping 1610 (unicode ordinal -> char ordinal) */ 1611 const char *errors /* error handling */ 1612 ); 1613PyAPI_FUNC(PyObject*) _PyUnicode_EncodeCharmap( 1614 PyObject *unicode, /* Unicode object */ 1615 PyObject *mapping, /* character mapping 1616 (unicode ordinal -> char ordinal) */ 1617 const char *errors /* error handling */ 1618 ); 1619#endif 1620 1621/* Translate a Py_UNICODE buffer of the given length by applying a 1622 character mapping table to it and return the resulting Unicode 1623 object. 1624 1625 The mapping table must map Unicode ordinal integers to Unicode 1626 ordinal integers or None (causing deletion of the character). 1627 1628 Mapping tables may be dictionaries or sequences. Unmapped character 1629 ordinals (ones which cause a LookupError) are left untouched and 1630 are copied as-is. 1631 1632*/ 1633 1634#ifndef Py_LIMITED_API 1635PyAPI_FUNC(PyObject *) PyUnicode_TranslateCharmap( 1636 const Py_UNICODE *data, /* Unicode char buffer */ 1637 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1638 PyObject *table, /* Translate table */ 1639 const char *errors /* error handling */ 1640 ); 1641#endif 1642 1643#ifdef HAVE_MBCS 1644 1645/* --- MBCS codecs for Windows -------------------------------------------- */ 1646 1647PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCS( 1648 const char *string, /* MBCS encoded string */ 1649 Py_ssize_t length, /* size of string */ 1650 const char *errors /* error handling */ 1651 ); 1652 1653PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCSStateful( 1654 const char *string, /* MBCS encoded string */ 1655 Py_ssize_t length, /* size of string */ 1656 const char *errors, /* error handling */ 1657 Py_ssize_t *consumed /* bytes consumed */ 1658 ); 1659 1660PyAPI_FUNC(PyObject*) PyUnicode_DecodeCodePageStateful( 1661 int code_page, /* code page number */ 1662 const char *string, /* encoded string */ 1663 Py_ssize_t length, /* size of string */ 1664 const char *errors, /* error handling */ 1665 Py_ssize_t *consumed /* bytes consumed */ 1666 ); 1667 1668PyAPI_FUNC(PyObject*) PyUnicode_AsMBCSString( 1669 PyObject *unicode /* Unicode object */ 1670 ); 1671 1672#ifndef Py_LIMITED_API 1673PyAPI_FUNC(PyObject*) PyUnicode_EncodeMBCS( 1674 const Py_UNICODE *data, /* Unicode char buffer */ 1675 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1676 const char *errors /* error handling */ 1677 ); 1678#endif 1679 1680PyAPI_FUNC(PyObject*) PyUnicode_EncodeCodePage( 1681 int code_page, /* code page number */ 1682 PyObject *unicode, /* Unicode object */ 1683 const char *errors /* error handling */ 1684 ); 1685 1686#endif /* HAVE_MBCS */ 1687 1688/* --- Decimal Encoder ---------------------------------------------------- */ 1689 1690/* Takes a Unicode string holding a decimal value and writes it into 1691 an output buffer using standard ASCII digit codes. 1692 1693 The output buffer has to provide at least length+1 bytes of storage 1694 area. The output string is 0-terminated. 1695 1696 The encoder converts whitespace to ' ', decimal characters to their 1697 corresponding ASCII digit and all other Latin-1 characters except 1698 \0 as-is. Characters outside this range (Unicode ordinals 1-256) 1699 are treated as errors. This includes embedded NULL bytes. 1700 1701 Error handling is defined by the errors argument: 1702 1703 NULL or "strict": raise a ValueError 1704 "ignore": ignore the wrong characters (these are not copied to the 1705 output buffer) 1706 "replace": replaces illegal characters with '?' 1707 1708 Returns 0 on success, -1 on failure. 1709 1710*/ 1711 1712#ifndef Py_LIMITED_API 1713PyAPI_FUNC(int) PyUnicode_EncodeDecimal( 1714 Py_UNICODE *s, /* Unicode buffer */ 1715 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1716 char *output, /* Output buffer; must have size >= length */ 1717 const char *errors /* error handling */ 1718 ); 1719#endif 1720 1721/* Transforms code points that have decimal digit property to the 1722 corresponding ASCII digit code points. 1723 1724 Returns a new Unicode string on success, NULL on failure. 1725*/ 1726 1727#ifndef Py_LIMITED_API 1728PyAPI_FUNC(PyObject*) PyUnicode_TransformDecimalToASCII( 1729 Py_UNICODE *s, /* Unicode buffer */ 1730 Py_ssize_t length /* Number of Py_UNICODE chars to transform */ 1731 ); 1732#endif 1733 1734/* Similar to PyUnicode_TransformDecimalToASCII(), but takes a PyObject 1735 as argument instead of a raw buffer and length. This function additionally 1736 transforms spaces to ASCII because this is what the callers in longobject, 1737 floatobject, and complexobject did anyways. */ 1738 1739#ifndef Py_LIMITED_API 1740PyAPI_FUNC(PyObject*) _PyUnicode_TransformDecimalAndSpaceToASCII( 1741 PyObject *unicode /* Unicode object */ 1742 ); 1743#endif 1744 1745/* --- Locale encoding --------------------------------------------------- */ 1746 1747/* Decode a string from the current locale encoding. The decoder is strict if 1748 *surrogateescape* is equal to zero, otherwise it uses the 'surrogateescape' 1749 error handler (PEP 383) to escape undecodable bytes. If a byte sequence can 1750 be decoded as a surrogate character and *surrogateescape* is not equal to 1751 zero, the byte sequence is escaped using the 'surrogateescape' error handler 1752 instead of being decoded. *str* must end with a null character but cannot 1753 contain embedded null characters. */ 1754 1755PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocaleAndSize( 1756 const char *str, 1757 Py_ssize_t len, 1758 const char *errors); 1759 1760/* Similar to PyUnicode_DecodeLocaleAndSize(), but compute the string 1761 length using strlen(). */ 1762 1763PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocale( 1764 const char *str, 1765 const char *errors); 1766 1767/* Encode a Unicode object to the current locale encoding. The encoder is 1768 strict is *surrogateescape* is equal to zero, otherwise the 1769 "surrogateescape" error handler is used. Return a bytes object. The string 1770 cannot contain embedded null characters. */ 1771 1772PyAPI_FUNC(PyObject*) PyUnicode_EncodeLocale( 1773 PyObject *unicode, 1774 const char *errors 1775 ); 1776 1777/* --- File system encoding ---------------------------------------------- */ 1778 1779/* ParseTuple converter: encode str objects to bytes using 1780 PyUnicode_EncodeFSDefault(); bytes objects are output as-is. */ 1781 1782PyAPI_FUNC(int) PyUnicode_FSConverter(PyObject*, void*); 1783 1784/* ParseTuple converter: decode bytes objects to unicode using 1785 PyUnicode_DecodeFSDefaultAndSize(); str objects are output as-is. */ 1786 1787PyAPI_FUNC(int) PyUnicode_FSDecoder(PyObject*, void*); 1788 1789/* Decode a null-terminated string using Py_FileSystemDefaultEncoding 1790 and the "surrogateescape" error handler. 1791 1792 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1793 encoding. 1794 1795 Use PyUnicode_DecodeFSDefaultAndSize() if the string length is known. 1796*/ 1797 1798PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefault( 1799 const char *s /* encoded string */ 1800 ); 1801 1802/* Decode a string using Py_FileSystemDefaultEncoding 1803 and the "surrogateescape" error handler. 1804 1805 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1806 encoding. 1807*/ 1808 1809PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefaultAndSize( 1810 const char *s, /* encoded string */ 1811 Py_ssize_t size /* size */ 1812 ); 1813 1814/* Encode a Unicode object to Py_FileSystemDefaultEncoding with the 1815 "surrogateescape" error handler, and return bytes. 1816 1817 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1818 encoding. 1819*/ 1820 1821PyAPI_FUNC(PyObject*) PyUnicode_EncodeFSDefault( 1822 PyObject *unicode 1823 ); 1824 1825/* --- Methods & Slots ---------------------------------------------------- 1826 1827 These are capable of handling Unicode objects and strings on input 1828 (we refer to them as strings in the descriptions) and return 1829 Unicode objects or integers as appropriate. */ 1830 1831/* Concat two strings giving a new Unicode string. */ 1832 1833PyAPI_FUNC(PyObject*) PyUnicode_Concat( 1834 PyObject *left, /* Left string */ 1835 PyObject *right /* Right string */ 1836 ); 1837 1838/* Concat two strings and put the result in *pleft 1839 (sets *pleft to NULL on error) */ 1840 1841PyAPI_FUNC(void) PyUnicode_Append( 1842 PyObject **pleft, /* Pointer to left string */ 1843 PyObject *right /* Right string */ 1844 ); 1845 1846/* Concat two strings, put the result in *pleft and drop the right object 1847 (sets *pleft to NULL on error) */ 1848 1849PyAPI_FUNC(void) PyUnicode_AppendAndDel( 1850 PyObject **pleft, /* Pointer to left string */ 1851 PyObject *right /* Right string */ 1852 ); 1853 1854/* Split a string giving a list of Unicode strings. 1855 1856 If sep is NULL, splitting will be done at all whitespace 1857 substrings. Otherwise, splits occur at the given separator. 1858 1859 At most maxsplit splits will be done. If negative, no limit is set. 1860 1861 Separators are not included in the resulting list. 1862 1863*/ 1864 1865PyAPI_FUNC(PyObject*) PyUnicode_Split( 1866 PyObject *s, /* String to split */ 1867 PyObject *sep, /* String separator */ 1868 Py_ssize_t maxsplit /* Maxsplit count */ 1869 ); 1870 1871/* Dito, but split at line breaks. 1872 1873 CRLF is considered to be one line break. Line breaks are not 1874 included in the resulting list. */ 1875 1876PyAPI_FUNC(PyObject*) PyUnicode_Splitlines( 1877 PyObject *s, /* String to split */ 1878 int keepends /* If true, line end markers are included */ 1879 ); 1880 1881/* Partition a string using a given separator. */ 1882 1883PyAPI_FUNC(PyObject*) PyUnicode_Partition( 1884 PyObject *s, /* String to partition */ 1885 PyObject *sep /* String separator */ 1886 ); 1887 1888/* Partition a string using a given separator, searching from the end of the 1889 string. */ 1890 1891PyAPI_FUNC(PyObject*) PyUnicode_RPartition( 1892 PyObject *s, /* String to partition */ 1893 PyObject *sep /* String separator */ 1894 ); 1895 1896/* Split a string giving a list of Unicode strings. 1897 1898 If sep is NULL, splitting will be done at all whitespace 1899 substrings. Otherwise, splits occur at the given separator. 1900 1901 At most maxsplit splits will be done. But unlike PyUnicode_Split 1902 PyUnicode_RSplit splits from the end of the string. If negative, 1903 no limit is set. 1904 1905 Separators are not included in the resulting list. 1906 1907*/ 1908 1909PyAPI_FUNC(PyObject*) PyUnicode_RSplit( 1910 PyObject *s, /* String to split */ 1911 PyObject *sep, /* String separator */ 1912 Py_ssize_t maxsplit /* Maxsplit count */ 1913 ); 1914 1915/* Translate a string by applying a character mapping table to it and 1916 return the resulting Unicode object. 1917 1918 The mapping table must map Unicode ordinal integers to Unicode 1919 ordinal integers or None (causing deletion of the character). 1920 1921 Mapping tables may be dictionaries or sequences. Unmapped character 1922 ordinals (ones which cause a LookupError) are left untouched and 1923 are copied as-is. 1924 1925*/ 1926 1927PyAPI_FUNC(PyObject *) PyUnicode_Translate( 1928 PyObject *str, /* String */ 1929 PyObject *table, /* Translate table */ 1930 const char *errors /* error handling */ 1931 ); 1932 1933/* Join a sequence of strings using the given separator and return 1934 the resulting Unicode string. */ 1935 1936PyAPI_FUNC(PyObject*) PyUnicode_Join( 1937 PyObject *separator, /* Separator string */ 1938 PyObject *seq /* Sequence object */ 1939 ); 1940 1941/* Return 1 if substr matches str[start:end] at the given tail end, 0 1942 otherwise. */ 1943 1944PyAPI_FUNC(Py_ssize_t) PyUnicode_Tailmatch( 1945 PyObject *str, /* String */ 1946 PyObject *substr, /* Prefix or Suffix string */ 1947 Py_ssize_t start, /* Start index */ 1948 Py_ssize_t end, /* Stop index */ 1949 int direction /* Tail end: -1 prefix, +1 suffix */ 1950 ); 1951 1952/* Return the first position of substr in str[start:end] using the 1953 given search direction or -1 if not found. -2 is returned in case 1954 an error occurred and an exception is set. */ 1955 1956PyAPI_FUNC(Py_ssize_t) PyUnicode_Find( 1957 PyObject *str, /* String */ 1958 PyObject *substr, /* Substring to find */ 1959 Py_ssize_t start, /* Start index */ 1960 Py_ssize_t end, /* Stop index */ 1961 int direction /* Find direction: +1 forward, -1 backward */ 1962 ); 1963 1964/* Like PyUnicode_Find, but search for single character only. */ 1965PyAPI_FUNC(Py_ssize_t) PyUnicode_FindChar( 1966 PyObject *str, 1967 Py_UCS4 ch, 1968 Py_ssize_t start, 1969 Py_ssize_t end, 1970 int direction 1971 ); 1972 1973/* Count the number of occurrences of substr in str[start:end]. */ 1974 1975PyAPI_FUNC(Py_ssize_t) PyUnicode_Count( 1976 PyObject *str, /* String */ 1977 PyObject *substr, /* Substring to count */ 1978 Py_ssize_t start, /* Start index */ 1979 Py_ssize_t end /* Stop index */ 1980 ); 1981 1982/* Replace at most maxcount occurrences of substr in str with replstr 1983 and return the resulting Unicode object. */ 1984 1985PyAPI_FUNC(PyObject *) PyUnicode_Replace( 1986 PyObject *str, /* String */ 1987 PyObject *substr, /* Substring to find */ 1988 PyObject *replstr, /* Substring to replace */ 1989 Py_ssize_t maxcount /* Max. number of replacements to apply; 1990 -1 = all */ 1991 ); 1992 1993/* Compare two strings and return -1, 0, 1 for less than, equal, 1994 greater than resp. 1995 Raise an exception and return -1 on error. */ 1996 1997PyAPI_FUNC(int) PyUnicode_Compare( 1998 PyObject *left, /* Left string */ 1999 PyObject *right /* Right string */ 2000 ); 2001 2002#ifndef Py_LIMITED_API 2003/* Compare a string with an identifier and return -1, 0, 1 for less than, 2004 equal, and greater than, respectively. 2005 Raise an exception and return -1 on error. */ 2006 2007PyAPI_FUNC(int) _PyUnicode_CompareWithId( 2008 PyObject *left, /* Left string */ 2009 _Py_Identifier *right /* Right identifier */ 2010 ); 2011 2012/* Test whether a unicode is equal to ASCII identifier. Return 1 if true, 2013 0 otherwise. Return 0 if any argument contains non-ASCII characters. 2014 Any error occurs inside will be cleared before return. */ 2015 2016PyAPI_FUNC(int) _PyUnicode_EqualToASCIIId( 2017 PyObject *left, /* Left string */ 2018 _Py_Identifier *right /* Right identifier */ 2019 ); 2020#endif 2021 2022/* Compare a Unicode object with C string and return -1, 0, 1 for less than, 2023 equal, and greater than, respectively. It is best to pass only 2024 ASCII-encoded strings, but the function interprets the input string as 2025 ISO-8859-1 if it contains non-ASCII characters. 2026 This function does not raise exceptions. */ 2027 2028PyAPI_FUNC(int) PyUnicode_CompareWithASCIIString( 2029 PyObject *left, 2030 const char *right /* ASCII-encoded string */ 2031 ); 2032 2033#ifndef Py_LIMITED_API 2034/* Test whether a unicode is equal to ASCII string. Return 1 if true, 2035 0 otherwise. Return 0 if any argument contains non-ASCII characters. 2036 Any error occurs inside will be cleared before return. */ 2037 2038PyAPI_FUNC(int) _PyUnicode_EqualToASCIIString( 2039 PyObject *left, 2040 const char *right /* ASCII-encoded string */ 2041 ); 2042#endif 2043 2044/* Rich compare two strings and return one of the following: 2045 2046 - NULL in case an exception was raised 2047 - Py_True or Py_False for successful comparisons 2048 - Py_NotImplemented in case the type combination is unknown 2049 2050 Possible values for op: 2051 2052 Py_GT, Py_GE, Py_EQ, Py_NE, Py_LT, Py_LE 2053 2054*/ 2055 2056PyAPI_FUNC(PyObject *) PyUnicode_RichCompare( 2057 PyObject *left, /* Left string */ 2058 PyObject *right, /* Right string */ 2059 int op /* Operation: Py_EQ, Py_NE, Py_GT, etc. */ 2060 ); 2061 2062/* Apply an argument tuple or dictionary to a format string and return 2063 the resulting Unicode string. */ 2064 2065PyAPI_FUNC(PyObject *) PyUnicode_Format( 2066 PyObject *format, /* Format string */ 2067 PyObject *args /* Argument tuple or dictionary */ 2068 ); 2069 2070/* Checks whether element is contained in container and return 1/0 2071 accordingly. 2072 2073 element has to coerce to a one element Unicode string. -1 is 2074 returned in case of an error. */ 2075 2076PyAPI_FUNC(int) PyUnicode_Contains( 2077 PyObject *container, /* Container string */ 2078 PyObject *element /* Element string */ 2079 ); 2080 2081/* Checks whether argument is a valid identifier. */ 2082 2083PyAPI_FUNC(int) PyUnicode_IsIdentifier(PyObject *s); 2084 2085#ifndef Py_LIMITED_API 2086/* Externally visible for str.strip(unicode) */ 2087PyAPI_FUNC(PyObject *) _PyUnicode_XStrip( 2088 PyObject *self, 2089 int striptype, 2090 PyObject *sepobj 2091 ); 2092#endif 2093 2094/* Using explicit passed-in values, insert the thousands grouping 2095 into the string pointed to by buffer. For the argument descriptions, 2096 see Objects/stringlib/localeutil.h */ 2097#ifndef Py_LIMITED_API 2098PyAPI_FUNC(Py_ssize_t) _PyUnicode_InsertThousandsGrouping( 2099 PyObject *unicode, 2100 Py_ssize_t index, 2101 Py_ssize_t n_buffer, 2102 void *digits, 2103 Py_ssize_t n_digits, 2104 Py_ssize_t min_width, 2105 const char *grouping, 2106 PyObject *thousands_sep, 2107 Py_UCS4 *maxchar); 2108#endif 2109/* === Characters Type APIs =============================================== */ 2110 2111/* Helper array used by Py_UNICODE_ISSPACE(). */ 2112 2113#ifndef Py_LIMITED_API 2114PyAPI_DATA(const unsigned char) _Py_ascii_whitespace[]; 2115 2116/* These should not be used directly. Use the Py_UNICODE_IS* and 2117 Py_UNICODE_TO* macros instead. 2118 2119 These APIs are implemented in Objects/unicodectype.c. 2120 2121*/ 2122 2123PyAPI_FUNC(int) _PyUnicode_IsLowercase( 2124 Py_UCS4 ch /* Unicode character */ 2125 ); 2126 2127PyAPI_FUNC(int) _PyUnicode_IsUppercase( 2128 Py_UCS4 ch /* Unicode character */ 2129 ); 2130 2131PyAPI_FUNC(int) _PyUnicode_IsTitlecase( 2132 Py_UCS4 ch /* Unicode character */ 2133 ); 2134 2135PyAPI_FUNC(int) _PyUnicode_IsXidStart( 2136 Py_UCS4 ch /* Unicode character */ 2137 ); 2138 2139PyAPI_FUNC(int) _PyUnicode_IsXidContinue( 2140 Py_UCS4 ch /* Unicode character */ 2141 ); 2142 2143PyAPI_FUNC(int) _PyUnicode_IsWhitespace( 2144 const Py_UCS4 ch /* Unicode character */ 2145 ); 2146 2147PyAPI_FUNC(int) _PyUnicode_IsLinebreak( 2148 const Py_UCS4 ch /* Unicode character */ 2149 ); 2150 2151PyAPI_FUNC(Py_UCS4) _PyUnicode_ToLowercase( 2152 Py_UCS4 ch /* Unicode character */ 2153 ); 2154 2155PyAPI_FUNC(Py_UCS4) _PyUnicode_ToUppercase( 2156 Py_UCS4 ch /* Unicode character */ 2157 ); 2158 2159PyAPI_FUNC(Py_UCS4) _PyUnicode_ToTitlecase( 2160 Py_UCS4 ch /* Unicode character */ 2161 ); 2162 2163PyAPI_FUNC(int) _PyUnicode_ToLowerFull( 2164 Py_UCS4 ch, /* Unicode character */ 2165 Py_UCS4 *res 2166 ); 2167 2168PyAPI_FUNC(int) _PyUnicode_ToTitleFull( 2169 Py_UCS4 ch, /* Unicode character */ 2170 Py_UCS4 *res 2171 ); 2172 2173PyAPI_FUNC(int) _PyUnicode_ToUpperFull( 2174 Py_UCS4 ch, /* Unicode character */ 2175 Py_UCS4 *res 2176 ); 2177 2178PyAPI_FUNC(int) _PyUnicode_ToFoldedFull( 2179 Py_UCS4 ch, /* Unicode character */ 2180 Py_UCS4 *res 2181 ); 2182 2183PyAPI_FUNC(int) _PyUnicode_IsCaseIgnorable( 2184 Py_UCS4 ch /* Unicode character */ 2185 ); 2186 2187PyAPI_FUNC(int) _PyUnicode_IsCased( 2188 Py_UCS4 ch /* Unicode character */ 2189 ); 2190 2191PyAPI_FUNC(int) _PyUnicode_ToDecimalDigit( 2192 Py_UCS4 ch /* Unicode character */ 2193 ); 2194 2195PyAPI_FUNC(int) _PyUnicode_ToDigit( 2196 Py_UCS4 ch /* Unicode character */ 2197 ); 2198 2199PyAPI_FUNC(double) _PyUnicode_ToNumeric( 2200 Py_UCS4 ch /* Unicode character */ 2201 ); 2202 2203PyAPI_FUNC(int) _PyUnicode_IsDecimalDigit( 2204 Py_UCS4 ch /* Unicode character */ 2205 ); 2206 2207PyAPI_FUNC(int) _PyUnicode_IsDigit( 2208 Py_UCS4 ch /* Unicode character */ 2209 ); 2210 2211PyAPI_FUNC(int) _PyUnicode_IsNumeric( 2212 Py_UCS4 ch /* Unicode character */ 2213 ); 2214 2215PyAPI_FUNC(int) _PyUnicode_IsPrintable( 2216 Py_UCS4 ch /* Unicode character */ 2217 ); 2218 2219PyAPI_FUNC(int) _PyUnicode_IsAlpha( 2220 Py_UCS4 ch /* Unicode character */ 2221 ); 2222 2223PyAPI_FUNC(size_t) Py_UNICODE_strlen( 2224 const Py_UNICODE *u 2225 ); 2226 2227PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strcpy( 2228 Py_UNICODE *s1, 2229 const Py_UNICODE *s2); 2230 2231PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strcat( 2232 Py_UNICODE *s1, const Py_UNICODE *s2); 2233 2234PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strncpy( 2235 Py_UNICODE *s1, 2236 const Py_UNICODE *s2, 2237 size_t n); 2238 2239PyAPI_FUNC(int) Py_UNICODE_strcmp( 2240 const Py_UNICODE *s1, 2241 const Py_UNICODE *s2 2242 ); 2243 2244PyAPI_FUNC(int) Py_UNICODE_strncmp( 2245 const Py_UNICODE *s1, 2246 const Py_UNICODE *s2, 2247 size_t n 2248 ); 2249 2250PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strchr( 2251 const Py_UNICODE *s, 2252 Py_UNICODE c 2253 ); 2254 2255PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strrchr( 2256 const Py_UNICODE *s, 2257 Py_UNICODE c 2258 ); 2259 2260PyAPI_FUNC(PyObject*) _PyUnicode_FormatLong(PyObject *, int, int, int); 2261 2262/* Create a copy of a unicode string ending with a nul character. Return NULL 2263 and raise a MemoryError exception on memory allocation failure, otherwise 2264 return a new allocated buffer (use PyMem_Free() to free the buffer). */ 2265 2266PyAPI_FUNC(Py_UNICODE*) PyUnicode_AsUnicodeCopy( 2267 PyObject *unicode 2268 ); 2269#endif /* Py_LIMITED_API */ 2270 2271#if defined(Py_DEBUG) && !defined(Py_LIMITED_API) 2272PyAPI_FUNC(int) _PyUnicode_CheckConsistency( 2273 PyObject *op, 2274 int check_content); 2275#endif 2276 2277/* Return an interned Unicode object for an Identifier; may fail if there is no memory.*/ 2278PyAPI_FUNC(PyObject*) _PyUnicode_FromId(_Py_Identifier*); 2279/* Clear all static strings. */ 2280PyAPI_FUNC(void) _PyUnicode_ClearStaticStrings(void); 2281 2282#ifdef __cplusplus 2283} 2284#endif 2285#endif /* !Py_UNICODEOBJECT_H */ 2286