unicodeobject.h revision 7e9d1d1a1b8ec4db9b9b6789b448c4202ab84b48
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 ValueError 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 an 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/* Coerce obj to an Unicode object and return a reference with 848 *incremented* refcount. 849 850 Coercion is done in the following way: 851 852 1. bytes, bytearray and other bytes-like objects are decoded 853 under the assumptions that they contain data using the UTF-8 854 encoding. Decoding is done in "strict" mode. 855 856 2. All other objects (including Unicode objects) raise an 857 exception. 858 859 The API returns NULL in case of an error. The caller is responsible 860 for decref'ing the returned objects. 861 862*/ 863 864PyAPI_FUNC(PyObject*) PyUnicode_FromEncodedObject( 865 PyObject *obj, /* Object */ 866 const char *encoding, /* encoding */ 867 const char *errors /* error handling */ 868 ); 869 870/* Coerce obj to an Unicode object and return a reference with 871 *incremented* refcount. 872 873 Unicode objects are passed back as-is (subclasses are converted to 874 true Unicode objects), all other objects are delegated to 875 PyUnicode_FromEncodedObject(obj, NULL, "strict") which results in 876 using UTF-8 encoding as basis for decoding the object. 877 878 The API returns NULL in case of an error. The caller is responsible 879 for decref'ing the returned objects. 880 881*/ 882 883PyAPI_FUNC(PyObject*) PyUnicode_FromObject( 884 PyObject *obj /* Object */ 885 ); 886 887PyAPI_FUNC(PyObject *) PyUnicode_FromFormatV( 888 const char *format, /* ASCII-encoded string */ 889 va_list vargs 890 ); 891PyAPI_FUNC(PyObject *) PyUnicode_FromFormat( 892 const char *format, /* ASCII-encoded string */ 893 ... 894 ); 895 896#ifndef Py_LIMITED_API 897typedef struct { 898 PyObject *buffer; 899 void *data; 900 enum PyUnicode_Kind kind; 901 Py_UCS4 maxchar; 902 Py_ssize_t size; 903 Py_ssize_t pos; 904 905 /* minimum number of allocated characters (default: 0) */ 906 Py_ssize_t min_length; 907 908 /* minimum character (default: 127, ASCII) */ 909 Py_UCS4 min_char; 910 911 /* If non-zero, overallocate the buffer by 25% (default: 0). */ 912 unsigned char overallocate; 913 914 /* If readonly is 1, buffer is a shared string (cannot be modified) 915 and size is set to 0. */ 916 unsigned char readonly; 917} _PyUnicodeWriter ; 918 919/* Initialize a Unicode writer. 920 * 921 * By default, the minimum buffer size is 0 character and overallocation is 922 * disabled. Set min_length, min_char and overallocate attributes to control 923 * the allocation of the buffer. */ 924PyAPI_FUNC(void) 925_PyUnicodeWriter_Init(_PyUnicodeWriter *writer); 926 927/* Prepare the buffer to write 'length' characters 928 with the specified maximum character. 929 930 Return 0 on success, raise an exception and return -1 on error. */ 931#define _PyUnicodeWriter_Prepare(WRITER, LENGTH, MAXCHAR) \ 932 (((MAXCHAR) <= (WRITER)->maxchar \ 933 && (LENGTH) <= (WRITER)->size - (WRITER)->pos) \ 934 ? 0 \ 935 : (((LENGTH) == 0) \ 936 ? 0 \ 937 : _PyUnicodeWriter_PrepareInternal((WRITER), (LENGTH), (MAXCHAR)))) 938 939/* Don't call this function directly, use the _PyUnicodeWriter_Prepare() macro 940 instead. */ 941PyAPI_FUNC(int) 942_PyUnicodeWriter_PrepareInternal(_PyUnicodeWriter *writer, 943 Py_ssize_t length, Py_UCS4 maxchar); 944 945/* Append a Unicode character. 946 Return 0 on success, raise an exception and return -1 on error. */ 947PyAPI_FUNC(int) 948_PyUnicodeWriter_WriteChar(_PyUnicodeWriter *writer, 949 Py_UCS4 ch 950 ); 951 952/* Append a Unicode string. 953 Return 0 on success, raise an exception and return -1 on error. */ 954PyAPI_FUNC(int) 955_PyUnicodeWriter_WriteStr(_PyUnicodeWriter *writer, 956 PyObject *str /* Unicode string */ 957 ); 958 959/* Append a substring of a Unicode string. 960 Return 0 on success, raise an exception and return -1 on error. */ 961PyAPI_FUNC(int) 962_PyUnicodeWriter_WriteSubstring(_PyUnicodeWriter *writer, 963 PyObject *str, /* Unicode string */ 964 Py_ssize_t start, 965 Py_ssize_t end 966 ); 967 968/* Append a ASCII-encoded byte string. 969 Return 0 on success, raise an exception and return -1 on error. */ 970PyAPI_FUNC(int) 971_PyUnicodeWriter_WriteASCIIString(_PyUnicodeWriter *writer, 972 const char *str, /* ASCII-encoded byte string */ 973 Py_ssize_t len /* number of bytes, or -1 if unknown */ 974 ); 975 976/* Append a latin1-encoded byte string. 977 Return 0 on success, raise an exception and return -1 on error. */ 978PyAPI_FUNC(int) 979_PyUnicodeWriter_WriteLatin1String(_PyUnicodeWriter *writer, 980 const char *str, /* latin1-encoded byte string */ 981 Py_ssize_t len /* length in bytes */ 982 ); 983 984/* Get the value of the writer as an Unicode string. Clear the 985 buffer of the writer. Raise an exception and return NULL 986 on error. */ 987PyAPI_FUNC(PyObject *) 988_PyUnicodeWriter_Finish(_PyUnicodeWriter *writer); 989 990/* Deallocate memory of a writer (clear its internal buffer). */ 991PyAPI_FUNC(void) 992_PyUnicodeWriter_Dealloc(_PyUnicodeWriter *writer); 993#endif 994 995#ifndef Py_LIMITED_API 996/* Format the object based on the format_spec, as defined in PEP 3101 997 (Advanced String Formatting). */ 998PyAPI_FUNC(int) _PyUnicode_FormatAdvancedWriter( 999 _PyUnicodeWriter *writer, 1000 PyObject *obj, 1001 PyObject *format_spec, 1002 Py_ssize_t start, 1003 Py_ssize_t end); 1004#endif 1005 1006PyAPI_FUNC(void) PyUnicode_InternInPlace(PyObject **); 1007PyAPI_FUNC(void) PyUnicode_InternImmortal(PyObject **); 1008PyAPI_FUNC(PyObject *) PyUnicode_InternFromString( 1009 const char *u /* UTF-8 encoded string */ 1010 ); 1011#ifndef Py_LIMITED_API 1012PyAPI_FUNC(void) _Py_ReleaseInternedUnicodeStrings(void); 1013#endif 1014 1015/* Use only if you know it's a string */ 1016#define PyUnicode_CHECK_INTERNED(op) \ 1017 (((PyASCIIObject *)(op))->state.interned) 1018 1019/* --- wchar_t support for platforms which support it --------------------- */ 1020 1021#ifdef HAVE_WCHAR_H 1022 1023/* Create a Unicode Object from the wchar_t buffer w of the given 1024 size. 1025 1026 The buffer is copied into the new object. */ 1027 1028PyAPI_FUNC(PyObject*) PyUnicode_FromWideChar( 1029 const wchar_t *w, /* wchar_t buffer */ 1030 Py_ssize_t size /* size of buffer */ 1031 ); 1032 1033/* Copies the Unicode Object contents into the wchar_t buffer w. At 1034 most size wchar_t characters are copied. 1035 1036 Note that the resulting wchar_t string may or may not be 1037 0-terminated. It is the responsibility of the caller to make sure 1038 that the wchar_t string is 0-terminated in case this is required by 1039 the application. 1040 1041 Returns the number of wchar_t characters copied (excluding a 1042 possibly trailing 0-termination character) or -1 in case of an 1043 error. */ 1044 1045PyAPI_FUNC(Py_ssize_t) PyUnicode_AsWideChar( 1046 PyObject *unicode, /* Unicode object */ 1047 wchar_t *w, /* wchar_t buffer */ 1048 Py_ssize_t size /* size of buffer */ 1049 ); 1050 1051/* Convert the Unicode object to a wide character string. The output string 1052 always ends with a nul character. If size is not NULL, write the number of 1053 wide characters (excluding the null character) into *size. 1054 1055 Returns a buffer allocated by PyMem_Malloc() (use PyMem_Free() to free it) 1056 on success. On error, returns NULL, *size is undefined and raises a 1057 MemoryError. */ 1058 1059PyAPI_FUNC(wchar_t*) PyUnicode_AsWideCharString( 1060 PyObject *unicode, /* Unicode object */ 1061 Py_ssize_t *size /* number of characters of the result */ 1062 ); 1063 1064#ifndef Py_LIMITED_API 1065PyAPI_FUNC(void*) _PyUnicode_AsKind(PyObject *s, unsigned int kind); 1066#endif 1067 1068#endif 1069 1070/* --- Unicode ordinals --------------------------------------------------- */ 1071 1072/* Create a Unicode Object from the given Unicode code point ordinal. 1073 1074 The ordinal must be in range(0x110000). A ValueError is 1075 raised in case it is not. 1076 1077*/ 1078 1079PyAPI_FUNC(PyObject*) PyUnicode_FromOrdinal(int ordinal); 1080 1081/* --- Free-list management ----------------------------------------------- */ 1082 1083/* Clear the free list used by the Unicode implementation. 1084 1085 This can be used to release memory used for objects on the free 1086 list back to the Python memory allocator. 1087 1088*/ 1089 1090PyAPI_FUNC(int) PyUnicode_ClearFreeList(void); 1091 1092/* === Builtin Codecs ===================================================== 1093 1094 Many of these APIs take two arguments encoding and errors. These 1095 parameters encoding and errors have the same semantics as the ones 1096 of the builtin str() API. 1097 1098 Setting encoding to NULL causes the default encoding (UTF-8) to be used. 1099 1100 Error handling is set by errors which may also be set to NULL 1101 meaning to use the default handling defined for the codec. Default 1102 error handling for all builtin codecs is "strict" (ValueErrors are 1103 raised). 1104 1105 The codecs all use a similar interface. Only deviation from the 1106 generic ones are documented. 1107 1108*/ 1109 1110/* --- Manage the default encoding ---------------------------------------- */ 1111 1112/* Returns a pointer to the default encoding (UTF-8) of the 1113 Unicode object unicode and the size of the encoded representation 1114 in bytes stored in *size. 1115 1116 In case of an error, no *size is set. 1117 1118 This function caches the UTF-8 encoded string in the unicodeobject 1119 and subsequent calls will return the same string. The memory is released 1120 when the unicodeobject is deallocated. 1121 1122 _PyUnicode_AsStringAndSize is a #define for PyUnicode_AsUTF8AndSize to 1123 support the previous internal function with the same behaviour. 1124 1125 *** This API is for interpreter INTERNAL USE ONLY and will likely 1126 *** be removed or changed in the future. 1127 1128 *** If you need to access the Unicode object as UTF-8 bytes string, 1129 *** please use PyUnicode_AsUTF8String() instead. 1130*/ 1131 1132#ifndef Py_LIMITED_API 1133PyAPI_FUNC(char *) PyUnicode_AsUTF8AndSize( 1134 PyObject *unicode, 1135 Py_ssize_t *size); 1136#define _PyUnicode_AsStringAndSize PyUnicode_AsUTF8AndSize 1137#endif 1138 1139/* Returns a pointer to the default encoding (UTF-8) of the 1140 Unicode object unicode. 1141 1142 Like PyUnicode_AsUTF8AndSize(), this also caches the UTF-8 representation 1143 in the unicodeobject. 1144 1145 _PyUnicode_AsString is a #define for PyUnicode_AsUTF8 to 1146 support the previous internal function with the same behaviour. 1147 1148 Use of this API is DEPRECATED since no size information can be 1149 extracted from the returned data. 1150 1151 *** This API is for interpreter INTERNAL USE ONLY and will likely 1152 *** be removed or changed for Python 3.1. 1153 1154 *** If you need to access the Unicode object as UTF-8 bytes string, 1155 *** please use PyUnicode_AsUTF8String() instead. 1156 1157*/ 1158 1159#ifndef Py_LIMITED_API 1160PyAPI_FUNC(char *) PyUnicode_AsUTF8(PyObject *unicode); 1161#define _PyUnicode_AsString PyUnicode_AsUTF8 1162#endif 1163 1164/* Returns "utf-8". */ 1165 1166PyAPI_FUNC(const char*) PyUnicode_GetDefaultEncoding(void); 1167 1168/* --- Generic Codecs ----------------------------------------------------- */ 1169 1170/* Create a Unicode object by decoding the encoded string s of the 1171 given size. */ 1172 1173PyAPI_FUNC(PyObject*) PyUnicode_Decode( 1174 const char *s, /* encoded string */ 1175 Py_ssize_t size, /* size of buffer */ 1176 const char *encoding, /* encoding */ 1177 const char *errors /* error handling */ 1178 ); 1179 1180/* Decode a Unicode object unicode and return the result as Python 1181 object. */ 1182 1183PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedObject( 1184 PyObject *unicode, /* Unicode object */ 1185 const char *encoding, /* encoding */ 1186 const char *errors /* error handling */ 1187 ); 1188 1189/* Decode a Unicode object unicode and return the result as Unicode 1190 object. */ 1191 1192PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedUnicode( 1193 PyObject *unicode, /* Unicode object */ 1194 const char *encoding, /* encoding */ 1195 const char *errors /* error handling */ 1196 ); 1197 1198/* Encodes a Py_UNICODE buffer of the given size and returns a 1199 Python string object. */ 1200 1201#ifndef Py_LIMITED_API 1202PyAPI_FUNC(PyObject*) PyUnicode_Encode( 1203 const Py_UNICODE *s, /* Unicode char buffer */ 1204 Py_ssize_t size, /* number of Py_UNICODE chars to encode */ 1205 const char *encoding, /* encoding */ 1206 const char *errors /* error handling */ 1207 ); 1208#endif 1209 1210/* Encodes a Unicode object and returns the result as Python 1211 object. */ 1212 1213PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedObject( 1214 PyObject *unicode, /* Unicode object */ 1215 const char *encoding, /* encoding */ 1216 const char *errors /* error handling */ 1217 ); 1218 1219/* Encodes a Unicode object and returns the result as Python string 1220 object. */ 1221 1222PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedString( 1223 PyObject *unicode, /* Unicode object */ 1224 const char *encoding, /* encoding */ 1225 const char *errors /* error handling */ 1226 ); 1227 1228/* Encodes a Unicode object and returns the result as Unicode 1229 object. */ 1230 1231PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedUnicode( 1232 PyObject *unicode, /* Unicode object */ 1233 const char *encoding, /* encoding */ 1234 const char *errors /* error handling */ 1235 ); 1236 1237/* Build an encoding map. */ 1238 1239PyAPI_FUNC(PyObject*) PyUnicode_BuildEncodingMap( 1240 PyObject* string /* 256 character map */ 1241 ); 1242 1243/* --- UTF-7 Codecs ------------------------------------------------------- */ 1244 1245PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7( 1246 const char *string, /* UTF-7 encoded string */ 1247 Py_ssize_t length, /* size of string */ 1248 const char *errors /* error handling */ 1249 ); 1250 1251PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7Stateful( 1252 const char *string, /* UTF-7 encoded string */ 1253 Py_ssize_t length, /* size of string */ 1254 const char *errors, /* error handling */ 1255 Py_ssize_t *consumed /* bytes consumed */ 1256 ); 1257 1258#ifndef Py_LIMITED_API 1259PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF7( 1260 const Py_UNICODE *data, /* Unicode char buffer */ 1261 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1262 int base64SetO, /* Encode RFC2152 Set O characters in base64 */ 1263 int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ 1264 const char *errors /* error handling */ 1265 ); 1266PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF7( 1267 PyObject *unicode, /* Unicode object */ 1268 int base64SetO, /* Encode RFC2152 Set O characters in base64 */ 1269 int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ 1270 const char *errors /* error handling */ 1271 ); 1272#endif 1273 1274/* --- UTF-8 Codecs ------------------------------------------------------- */ 1275 1276PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8( 1277 const char *string, /* UTF-8 encoded string */ 1278 Py_ssize_t length, /* size of string */ 1279 const char *errors /* error handling */ 1280 ); 1281 1282PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF8Stateful( 1283 const char *string, /* UTF-8 encoded string */ 1284 Py_ssize_t length, /* size of string */ 1285 const char *errors, /* error handling */ 1286 Py_ssize_t *consumed /* bytes consumed */ 1287 ); 1288 1289PyAPI_FUNC(PyObject*) PyUnicode_AsUTF8String( 1290 PyObject *unicode /* Unicode object */ 1291 ); 1292 1293#ifndef Py_LIMITED_API 1294PyAPI_FUNC(PyObject*) _PyUnicode_AsUTF8String( 1295 PyObject *unicode, 1296 const char *errors); 1297 1298PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF8( 1299 const Py_UNICODE *data, /* Unicode char buffer */ 1300 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1301 const char *errors /* error handling */ 1302 ); 1303#endif 1304 1305/* --- UTF-32 Codecs ------------------------------------------------------ */ 1306 1307/* Decodes length bytes from a UTF-32 encoded buffer string and returns 1308 the corresponding Unicode object. 1309 1310 errors (if non-NULL) defines the error handling. It defaults 1311 to "strict". 1312 1313 If byteorder is non-NULL, the decoder starts decoding using the 1314 given byte order: 1315 1316 *byteorder == -1: little endian 1317 *byteorder == 0: native order 1318 *byteorder == 1: big endian 1319 1320 In native mode, the first four bytes of the stream are checked for a 1321 BOM mark. If found, the BOM mark is analysed, the byte order 1322 adjusted and the BOM skipped. In the other modes, no BOM mark 1323 interpretation is done. After completion, *byteorder is set to the 1324 current byte order at the end of input data. 1325 1326 If byteorder is NULL, the codec starts in native order mode. 1327 1328*/ 1329 1330PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32( 1331 const char *string, /* UTF-32 encoded string */ 1332 Py_ssize_t length, /* size of string */ 1333 const char *errors, /* error handling */ 1334 int *byteorder /* pointer to byteorder to use 1335 0=native;-1=LE,1=BE; updated on 1336 exit */ 1337 ); 1338 1339PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF32Stateful( 1340 const char *string, /* UTF-32 encoded string */ 1341 Py_ssize_t length, /* size of string */ 1342 const char *errors, /* error handling */ 1343 int *byteorder, /* pointer to byteorder to use 1344 0=native;-1=LE,1=BE; updated on 1345 exit */ 1346 Py_ssize_t *consumed /* bytes consumed */ 1347 ); 1348 1349/* Returns a Python string using the UTF-32 encoding in native byte 1350 order. The string always starts with a BOM mark. */ 1351 1352PyAPI_FUNC(PyObject*) PyUnicode_AsUTF32String( 1353 PyObject *unicode /* Unicode object */ 1354 ); 1355 1356/* Returns a Python string object holding the UTF-32 encoded value of 1357 the Unicode data. 1358 1359 If byteorder is not 0, output is written according to the following 1360 byte order: 1361 1362 byteorder == -1: little endian 1363 byteorder == 0: native byte order (writes a BOM mark) 1364 byteorder == 1: big endian 1365 1366 If byteorder is 0, the output string will always start with the 1367 Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is 1368 prepended. 1369 1370*/ 1371 1372#ifndef Py_LIMITED_API 1373PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF32( 1374 const Py_UNICODE *data, /* Unicode char buffer */ 1375 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1376 const char *errors, /* error handling */ 1377 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1378 ); 1379PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF32( 1380 PyObject *object, /* Unicode object */ 1381 const char *errors, /* error handling */ 1382 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1383 ); 1384#endif 1385 1386/* --- UTF-16 Codecs ------------------------------------------------------ */ 1387 1388/* Decodes length bytes from a UTF-16 encoded buffer string and returns 1389 the corresponding Unicode object. 1390 1391 errors (if non-NULL) defines the error handling. It defaults 1392 to "strict". 1393 1394 If byteorder is non-NULL, the decoder starts decoding using the 1395 given byte order: 1396 1397 *byteorder == -1: little endian 1398 *byteorder == 0: native order 1399 *byteorder == 1: big endian 1400 1401 In native mode, the first two bytes of the stream are checked for a 1402 BOM mark. If found, the BOM mark is analysed, the byte order 1403 adjusted and the BOM skipped. In the other modes, no BOM mark 1404 interpretation is done. After completion, *byteorder is set to the 1405 current byte order at the end of input data. 1406 1407 If byteorder is NULL, the codec starts in native order mode. 1408 1409*/ 1410 1411PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16( 1412 const char *string, /* UTF-16 encoded string */ 1413 Py_ssize_t length, /* size of string */ 1414 const char *errors, /* error handling */ 1415 int *byteorder /* pointer to byteorder to use 1416 0=native;-1=LE,1=BE; updated on 1417 exit */ 1418 ); 1419 1420PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF16Stateful( 1421 const char *string, /* UTF-16 encoded string */ 1422 Py_ssize_t length, /* size of string */ 1423 const char *errors, /* error handling */ 1424 int *byteorder, /* pointer to byteorder to use 1425 0=native;-1=LE,1=BE; updated on 1426 exit */ 1427 Py_ssize_t *consumed /* bytes consumed */ 1428 ); 1429 1430/* Returns a Python string using the UTF-16 encoding in native byte 1431 order. The string always starts with a BOM mark. */ 1432 1433PyAPI_FUNC(PyObject*) PyUnicode_AsUTF16String( 1434 PyObject *unicode /* Unicode object */ 1435 ); 1436 1437/* Returns a Python string object holding the UTF-16 encoded value of 1438 the Unicode data. 1439 1440 If byteorder is not 0, output is written according to the following 1441 byte order: 1442 1443 byteorder == -1: little endian 1444 byteorder == 0: native byte order (writes a BOM mark) 1445 byteorder == 1: big endian 1446 1447 If byteorder is 0, the output string will always start with the 1448 Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is 1449 prepended. 1450 1451 Note that Py_UNICODE data is being interpreted as UTF-16 reduced to 1452 UCS-2. This trick makes it possible to add full UTF-16 capabilities 1453 at a later point without compromising the APIs. 1454 1455*/ 1456 1457#ifndef Py_LIMITED_API 1458PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF16( 1459 const Py_UNICODE *data, /* Unicode char buffer */ 1460 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1461 const char *errors, /* error handling */ 1462 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1463 ); 1464PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF16( 1465 PyObject* unicode, /* Unicode object */ 1466 const char *errors, /* error handling */ 1467 int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ 1468 ); 1469#endif 1470 1471/* --- Unicode-Escape Codecs ---------------------------------------------- */ 1472 1473PyAPI_FUNC(PyObject*) PyUnicode_DecodeUnicodeEscape( 1474 const char *string, /* Unicode-Escape encoded string */ 1475 Py_ssize_t length, /* size of string */ 1476 const char *errors /* error handling */ 1477 ); 1478 1479PyAPI_FUNC(PyObject*) PyUnicode_AsUnicodeEscapeString( 1480 PyObject *unicode /* Unicode object */ 1481 ); 1482 1483#ifndef Py_LIMITED_API 1484PyAPI_FUNC(PyObject*) PyUnicode_EncodeUnicodeEscape( 1485 const Py_UNICODE *data, /* Unicode char buffer */ 1486 Py_ssize_t length /* Number of Py_UNICODE chars to encode */ 1487 ); 1488#endif 1489 1490/* --- Raw-Unicode-Escape Codecs ------------------------------------------ */ 1491 1492PyAPI_FUNC(PyObject*) PyUnicode_DecodeRawUnicodeEscape( 1493 const char *string, /* Raw-Unicode-Escape encoded string */ 1494 Py_ssize_t length, /* size of string */ 1495 const char *errors /* error handling */ 1496 ); 1497 1498PyAPI_FUNC(PyObject*) PyUnicode_AsRawUnicodeEscapeString( 1499 PyObject *unicode /* Unicode object */ 1500 ); 1501 1502#ifndef Py_LIMITED_API 1503PyAPI_FUNC(PyObject*) PyUnicode_EncodeRawUnicodeEscape( 1504 const Py_UNICODE *data, /* Unicode char buffer */ 1505 Py_ssize_t length /* Number of Py_UNICODE chars to encode */ 1506 ); 1507#endif 1508 1509/* --- Unicode Internal Codec --------------------------------------------- 1510 1511 Only for internal use in _codecsmodule.c */ 1512 1513#ifndef Py_LIMITED_API 1514PyObject *_PyUnicode_DecodeUnicodeInternal( 1515 const char *string, 1516 Py_ssize_t length, 1517 const char *errors 1518 ); 1519#endif 1520 1521/* --- Latin-1 Codecs ----------------------------------------------------- 1522 1523 Note: Latin-1 corresponds to the first 256 Unicode ordinals. 1524 1525*/ 1526 1527PyAPI_FUNC(PyObject*) PyUnicode_DecodeLatin1( 1528 const char *string, /* Latin-1 encoded string */ 1529 Py_ssize_t length, /* size of string */ 1530 const char *errors /* error handling */ 1531 ); 1532 1533PyAPI_FUNC(PyObject*) PyUnicode_AsLatin1String( 1534 PyObject *unicode /* Unicode object */ 1535 ); 1536 1537#ifndef Py_LIMITED_API 1538PyAPI_FUNC(PyObject*) _PyUnicode_AsLatin1String( 1539 PyObject* unicode, 1540 const char* errors); 1541 1542PyAPI_FUNC(PyObject*) PyUnicode_EncodeLatin1( 1543 const Py_UNICODE *data, /* Unicode char buffer */ 1544 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1545 const char *errors /* error handling */ 1546 ); 1547#endif 1548 1549/* --- ASCII Codecs ------------------------------------------------------- 1550 1551 Only 7-bit ASCII data is excepted. All other codes generate errors. 1552 1553*/ 1554 1555PyAPI_FUNC(PyObject*) PyUnicode_DecodeASCII( 1556 const char *string, /* ASCII encoded string */ 1557 Py_ssize_t length, /* size of string */ 1558 const char *errors /* error handling */ 1559 ); 1560 1561PyAPI_FUNC(PyObject*) PyUnicode_AsASCIIString( 1562 PyObject *unicode /* Unicode object */ 1563 ); 1564 1565#ifndef Py_LIMITED_API 1566PyAPI_FUNC(PyObject*) _PyUnicode_AsASCIIString( 1567 PyObject* unicode, 1568 const char* errors); 1569 1570PyAPI_FUNC(PyObject*) PyUnicode_EncodeASCII( 1571 const Py_UNICODE *data, /* Unicode char buffer */ 1572 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1573 const char *errors /* error handling */ 1574 ); 1575#endif 1576 1577/* --- Character Map Codecs ----------------------------------------------- 1578 1579 This codec uses mappings to encode and decode characters. 1580 1581 Decoding mappings must map single string characters to single 1582 Unicode characters, integers (which are then interpreted as Unicode 1583 ordinals) or None (meaning "undefined mapping" and causing an 1584 error). 1585 1586 Encoding mappings must map single Unicode characters to single 1587 string characters, integers (which are then interpreted as Latin-1 1588 ordinals) or None (meaning "undefined mapping" and causing an 1589 error). 1590 1591 If a character lookup fails with a LookupError, the character is 1592 copied as-is meaning that its ordinal value will be interpreted as 1593 Unicode or Latin-1 ordinal resp. Because of this mappings only need 1594 to contain those mappings which map characters to different code 1595 points. 1596 1597*/ 1598 1599PyAPI_FUNC(PyObject*) PyUnicode_DecodeCharmap( 1600 const char *string, /* Encoded string */ 1601 Py_ssize_t length, /* size of string */ 1602 PyObject *mapping, /* character mapping 1603 (char ordinal -> unicode ordinal) */ 1604 const char *errors /* error handling */ 1605 ); 1606 1607PyAPI_FUNC(PyObject*) PyUnicode_AsCharmapString( 1608 PyObject *unicode, /* Unicode object */ 1609 PyObject *mapping /* character mapping 1610 (unicode ordinal -> char ordinal) */ 1611 ); 1612 1613#ifndef Py_LIMITED_API 1614PyAPI_FUNC(PyObject*) PyUnicode_EncodeCharmap( 1615 const Py_UNICODE *data, /* Unicode char buffer */ 1616 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1617 PyObject *mapping, /* character mapping 1618 (unicode ordinal -> char ordinal) */ 1619 const char *errors /* error handling */ 1620 ); 1621PyAPI_FUNC(PyObject*) _PyUnicode_EncodeCharmap( 1622 PyObject *unicode, /* Unicode object */ 1623 PyObject *mapping, /* character mapping 1624 (unicode ordinal -> char ordinal) */ 1625 const char *errors /* error handling */ 1626 ); 1627#endif 1628 1629/* Translate a Py_UNICODE buffer of the given length by applying a 1630 character mapping table to it and return the resulting Unicode 1631 object. 1632 1633 The mapping table must map Unicode ordinal integers to Unicode 1634 ordinal integers or None (causing deletion of the character). 1635 1636 Mapping tables may be dictionaries or sequences. Unmapped character 1637 ordinals (ones which cause a LookupError) are left untouched and 1638 are copied as-is. 1639 1640*/ 1641 1642#ifndef Py_LIMITED_API 1643PyAPI_FUNC(PyObject *) PyUnicode_TranslateCharmap( 1644 const Py_UNICODE *data, /* Unicode char buffer */ 1645 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1646 PyObject *table, /* Translate table */ 1647 const char *errors /* error handling */ 1648 ); 1649#endif 1650 1651#ifdef HAVE_MBCS 1652 1653/* --- MBCS codecs for Windows -------------------------------------------- */ 1654 1655PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCS( 1656 const char *string, /* MBCS encoded string */ 1657 Py_ssize_t length, /* size of string */ 1658 const char *errors /* error handling */ 1659 ); 1660 1661PyAPI_FUNC(PyObject*) PyUnicode_DecodeMBCSStateful( 1662 const char *string, /* MBCS 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_DecodeCodePageStateful( 1669 int code_page, /* code page number */ 1670 const char *string, /* encoded string */ 1671 Py_ssize_t length, /* size of string */ 1672 const char *errors, /* error handling */ 1673 Py_ssize_t *consumed /* bytes consumed */ 1674 ); 1675 1676PyAPI_FUNC(PyObject*) PyUnicode_AsMBCSString( 1677 PyObject *unicode /* Unicode object */ 1678 ); 1679 1680#ifndef Py_LIMITED_API 1681PyAPI_FUNC(PyObject*) PyUnicode_EncodeMBCS( 1682 const Py_UNICODE *data, /* Unicode char buffer */ 1683 Py_ssize_t length, /* number of Py_UNICODE chars to encode */ 1684 const char *errors /* error handling */ 1685 ); 1686#endif 1687 1688PyAPI_FUNC(PyObject*) PyUnicode_EncodeCodePage( 1689 int code_page, /* code page number */ 1690 PyObject *unicode, /* Unicode object */ 1691 const char *errors /* error handling */ 1692 ); 1693 1694#endif /* HAVE_MBCS */ 1695 1696/* --- Decimal Encoder ---------------------------------------------------- */ 1697 1698/* Takes a Unicode string holding a decimal value and writes it into 1699 an output buffer using standard ASCII digit codes. 1700 1701 The output buffer has to provide at least length+1 bytes of storage 1702 area. The output string is 0-terminated. 1703 1704 The encoder converts whitespace to ' ', decimal characters to their 1705 corresponding ASCII digit and all other Latin-1 characters except 1706 \0 as-is. Characters outside this range (Unicode ordinals 1-256) 1707 are treated as errors. This includes embedded NULL bytes. 1708 1709 Error handling is defined by the errors argument: 1710 1711 NULL or "strict": raise a ValueError 1712 "ignore": ignore the wrong characters (these are not copied to the 1713 output buffer) 1714 "replace": replaces illegal characters with '?' 1715 1716 Returns 0 on success, -1 on failure. 1717 1718*/ 1719 1720#ifndef Py_LIMITED_API 1721PyAPI_FUNC(int) PyUnicode_EncodeDecimal( 1722 Py_UNICODE *s, /* Unicode buffer */ 1723 Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ 1724 char *output, /* Output buffer; must have size >= length */ 1725 const char *errors /* error handling */ 1726 ); 1727#endif 1728 1729/* Transforms code points that have decimal digit property to the 1730 corresponding ASCII digit code points. 1731 1732 Returns a new Unicode string on success, NULL on failure. 1733*/ 1734 1735#ifndef Py_LIMITED_API 1736PyAPI_FUNC(PyObject*) PyUnicode_TransformDecimalToASCII( 1737 Py_UNICODE *s, /* Unicode buffer */ 1738 Py_ssize_t length /* Number of Py_UNICODE chars to transform */ 1739 ); 1740#endif 1741 1742/* Similar to PyUnicode_TransformDecimalToASCII(), but takes a PyObject 1743 as argument instead of a raw buffer and length. This function additionally 1744 transforms spaces to ASCII because this is what the callers in longobject, 1745 floatobject, and complexobject did anyways. */ 1746 1747#ifndef Py_LIMITED_API 1748PyAPI_FUNC(PyObject*) _PyUnicode_TransformDecimalAndSpaceToASCII( 1749 PyObject *unicode /* Unicode object */ 1750 ); 1751#endif 1752 1753/* --- Locale encoding --------------------------------------------------- */ 1754 1755/* Decode a string from the current locale encoding. The decoder is strict if 1756 *surrogateescape* is equal to zero, otherwise it uses the 'surrogateescape' 1757 error handler (PEP 383) to escape undecodable bytes. If a byte sequence can 1758 be decoded as a surrogate character and *surrogateescape* is not equal to 1759 zero, the byte sequence is escaped using the 'surrogateescape' error handler 1760 instead of being decoded. *str* must end with a null character but cannot 1761 contain embedded null characters. */ 1762 1763PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocaleAndSize( 1764 const char *str, 1765 Py_ssize_t len, 1766 const char *errors); 1767 1768/* Similar to PyUnicode_DecodeLocaleAndSize(), but compute the string 1769 length using strlen(). */ 1770 1771PyAPI_FUNC(PyObject*) PyUnicode_DecodeLocale( 1772 const char *str, 1773 const char *errors); 1774 1775/* Encode a Unicode object to the current locale encoding. The encoder is 1776 strict is *surrogateescape* is equal to zero, otherwise the 1777 "surrogateescape" error handler is used. Return a bytes object. The string 1778 cannot contain embedded null characters. */ 1779 1780PyAPI_FUNC(PyObject*) PyUnicode_EncodeLocale( 1781 PyObject *unicode, 1782 const char *errors 1783 ); 1784 1785/* --- File system encoding ---------------------------------------------- */ 1786 1787/* ParseTuple converter: encode str objects to bytes using 1788 PyUnicode_EncodeFSDefault(); bytes objects are output as-is. */ 1789 1790PyAPI_FUNC(int) PyUnicode_FSConverter(PyObject*, void*); 1791 1792/* ParseTuple converter: decode bytes objects to unicode using 1793 PyUnicode_DecodeFSDefaultAndSize(); str objects are output as-is. */ 1794 1795PyAPI_FUNC(int) PyUnicode_FSDecoder(PyObject*, void*); 1796 1797/* Decode a null-terminated string using Py_FileSystemDefaultEncoding 1798 and the "surrogateescape" error handler. 1799 1800 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1801 encoding. 1802 1803 Use PyUnicode_DecodeFSDefaultAndSize() if the string length is known. 1804*/ 1805 1806PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefault( 1807 const char *s /* encoded string */ 1808 ); 1809 1810/* Decode a string using Py_FileSystemDefaultEncoding 1811 and the "surrogateescape" error handler. 1812 1813 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1814 encoding. 1815*/ 1816 1817PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefaultAndSize( 1818 const char *s, /* encoded string */ 1819 Py_ssize_t size /* size */ 1820 ); 1821 1822/* Encode a Unicode object to Py_FileSystemDefaultEncoding with the 1823 "surrogateescape" error handler, and return bytes. 1824 1825 If Py_FileSystemDefaultEncoding is not set, fall back to the locale 1826 encoding. 1827*/ 1828 1829PyAPI_FUNC(PyObject*) PyUnicode_EncodeFSDefault( 1830 PyObject *unicode 1831 ); 1832 1833/* --- Methods & Slots ---------------------------------------------------- 1834 1835 These are capable of handling Unicode objects and strings on input 1836 (we refer to them as strings in the descriptions) and return 1837 Unicode objects or integers as appropriate. */ 1838 1839/* Concat two strings giving a new Unicode string. */ 1840 1841PyAPI_FUNC(PyObject*) PyUnicode_Concat( 1842 PyObject *left, /* Left string */ 1843 PyObject *right /* Right string */ 1844 ); 1845 1846/* Concat two strings and put the result in *pleft 1847 (sets *pleft to NULL on error) */ 1848 1849PyAPI_FUNC(void) PyUnicode_Append( 1850 PyObject **pleft, /* Pointer to left string */ 1851 PyObject *right /* Right string */ 1852 ); 1853 1854/* Concat two strings, put the result in *pleft and drop the right object 1855 (sets *pleft to NULL on error) */ 1856 1857PyAPI_FUNC(void) PyUnicode_AppendAndDel( 1858 PyObject **pleft, /* Pointer to left string */ 1859 PyObject *right /* Right string */ 1860 ); 1861 1862/* Split a string giving a list of Unicode strings. 1863 1864 If sep is NULL, splitting will be done at all whitespace 1865 substrings. Otherwise, splits occur at the given separator. 1866 1867 At most maxsplit splits will be done. If negative, no limit is set. 1868 1869 Separators are not included in the resulting list. 1870 1871*/ 1872 1873PyAPI_FUNC(PyObject*) PyUnicode_Split( 1874 PyObject *s, /* String to split */ 1875 PyObject *sep, /* String separator */ 1876 Py_ssize_t maxsplit /* Maxsplit count */ 1877 ); 1878 1879/* Dito, but split at line breaks. 1880 1881 CRLF is considered to be one line break. Line breaks are not 1882 included in the resulting list. */ 1883 1884PyAPI_FUNC(PyObject*) PyUnicode_Splitlines( 1885 PyObject *s, /* String to split */ 1886 int keepends /* If true, line end markers are included */ 1887 ); 1888 1889/* Partition a string using a given separator. */ 1890 1891PyAPI_FUNC(PyObject*) PyUnicode_Partition( 1892 PyObject *s, /* String to partition */ 1893 PyObject *sep /* String separator */ 1894 ); 1895 1896/* Partition a string using a given separator, searching from the end of the 1897 string. */ 1898 1899PyAPI_FUNC(PyObject*) PyUnicode_RPartition( 1900 PyObject *s, /* String to partition */ 1901 PyObject *sep /* String separator */ 1902 ); 1903 1904/* Split a string giving a list of Unicode strings. 1905 1906 If sep is NULL, splitting will be done at all whitespace 1907 substrings. Otherwise, splits occur at the given separator. 1908 1909 At most maxsplit splits will be done. But unlike PyUnicode_Split 1910 PyUnicode_RSplit splits from the end of the string. If negative, 1911 no limit is set. 1912 1913 Separators are not included in the resulting list. 1914 1915*/ 1916 1917PyAPI_FUNC(PyObject*) PyUnicode_RSplit( 1918 PyObject *s, /* String to split */ 1919 PyObject *sep, /* String separator */ 1920 Py_ssize_t maxsplit /* Maxsplit count */ 1921 ); 1922 1923/* Translate a string by applying a character mapping table to it and 1924 return the resulting Unicode object. 1925 1926 The mapping table must map Unicode ordinal integers to Unicode 1927 ordinal integers or None (causing deletion of the character). 1928 1929 Mapping tables may be dictionaries or sequences. Unmapped character 1930 ordinals (ones which cause a LookupError) are left untouched and 1931 are copied as-is. 1932 1933*/ 1934 1935PyAPI_FUNC(PyObject *) PyUnicode_Translate( 1936 PyObject *str, /* String */ 1937 PyObject *table, /* Translate table */ 1938 const char *errors /* error handling */ 1939 ); 1940 1941/* Join a sequence of strings using the given separator and return 1942 the resulting Unicode string. */ 1943 1944PyAPI_FUNC(PyObject*) PyUnicode_Join( 1945 PyObject *separator, /* Separator string */ 1946 PyObject *seq /* Sequence object */ 1947 ); 1948 1949/* Return 1 if substr matches str[start:end] at the given tail end, 0 1950 otherwise. */ 1951 1952PyAPI_FUNC(Py_ssize_t) PyUnicode_Tailmatch( 1953 PyObject *str, /* String */ 1954 PyObject *substr, /* Prefix or Suffix string */ 1955 Py_ssize_t start, /* Start index */ 1956 Py_ssize_t end, /* Stop index */ 1957 int direction /* Tail end: -1 prefix, +1 suffix */ 1958 ); 1959 1960/* Return the first position of substr in str[start:end] using the 1961 given search direction or -1 if not found. -2 is returned in case 1962 an error occurred and an exception is set. */ 1963 1964PyAPI_FUNC(Py_ssize_t) PyUnicode_Find( 1965 PyObject *str, /* String */ 1966 PyObject *substr, /* Substring to find */ 1967 Py_ssize_t start, /* Start index */ 1968 Py_ssize_t end, /* Stop index */ 1969 int direction /* Find direction: +1 forward, -1 backward */ 1970 ); 1971 1972/* Like PyUnicode_Find, but search for single character only. */ 1973PyAPI_FUNC(Py_ssize_t) PyUnicode_FindChar( 1974 PyObject *str, 1975 Py_UCS4 ch, 1976 Py_ssize_t start, 1977 Py_ssize_t end, 1978 int direction 1979 ); 1980 1981/* Count the number of occurrences of substr in str[start:end]. */ 1982 1983PyAPI_FUNC(Py_ssize_t) PyUnicode_Count( 1984 PyObject *str, /* String */ 1985 PyObject *substr, /* Substring to count */ 1986 Py_ssize_t start, /* Start index */ 1987 Py_ssize_t end /* Stop index */ 1988 ); 1989 1990/* Replace at most maxcount occurrences of substr in str with replstr 1991 and return the resulting Unicode object. */ 1992 1993PyAPI_FUNC(PyObject *) PyUnicode_Replace( 1994 PyObject *str, /* String */ 1995 PyObject *substr, /* Substring to find */ 1996 PyObject *replstr, /* Substring to replace */ 1997 Py_ssize_t maxcount /* Max. number of replacements to apply; 1998 -1 = all */ 1999 ); 2000 2001/* Compare two strings and return -1, 0, 1 for less than, equal, 2002 greater than resp. 2003 Raise an exception and return -1 on error. */ 2004 2005PyAPI_FUNC(int) PyUnicode_Compare( 2006 PyObject *left, /* Left string */ 2007 PyObject *right /* Right string */ 2008 ); 2009 2010#ifndef Py_LIMITED_API 2011PyAPI_FUNC(int) _PyUnicode_CompareWithId( 2012 PyObject *left, /* Left string */ 2013 _Py_Identifier *right /* Right identifier */ 2014 ); 2015#endif 2016 2017PyAPI_FUNC(int) PyUnicode_CompareWithASCIIString( 2018 PyObject *left, 2019 const char *right /* ASCII-encoded string */ 2020 ); 2021 2022/* Rich compare two strings and return one of the following: 2023 2024 - NULL in case an exception was raised 2025 - Py_True or Py_False for successfully comparisons 2026 - Py_NotImplemented in case the type combination is unknown 2027 2028 Note that Py_EQ and Py_NE comparisons can cause a UnicodeWarning in 2029 case the conversion of the arguments to Unicode fails with a 2030 UnicodeDecodeError. 2031 2032 Possible values for op: 2033 2034 Py_GT, Py_GE, Py_EQ, Py_NE, Py_LT, Py_LE 2035 2036*/ 2037 2038PyAPI_FUNC(PyObject *) PyUnicode_RichCompare( 2039 PyObject *left, /* Left string */ 2040 PyObject *right, /* Right string */ 2041 int op /* Operation: Py_EQ, Py_NE, Py_GT, etc. */ 2042 ); 2043 2044/* Apply a argument tuple or dictionary to a format string and return 2045 the resulting Unicode string. */ 2046 2047PyAPI_FUNC(PyObject *) PyUnicode_Format( 2048 PyObject *format, /* Format string */ 2049 PyObject *args /* Argument tuple or dictionary */ 2050 ); 2051 2052/* Checks whether element is contained in container and return 1/0 2053 accordingly. 2054 2055 element has to coerce to an one element Unicode string. -1 is 2056 returned in case of an error. */ 2057 2058PyAPI_FUNC(int) PyUnicode_Contains( 2059 PyObject *container, /* Container string */ 2060 PyObject *element /* Element string */ 2061 ); 2062 2063/* Checks whether argument is a valid identifier. */ 2064 2065PyAPI_FUNC(int) PyUnicode_IsIdentifier(PyObject *s); 2066 2067#ifndef Py_LIMITED_API 2068/* Externally visible for str.strip(unicode) */ 2069PyAPI_FUNC(PyObject *) _PyUnicode_XStrip( 2070 PyObject *self, 2071 int striptype, 2072 PyObject *sepobj 2073 ); 2074#endif 2075 2076/* Using explicit passed-in values, insert the thousands grouping 2077 into the string pointed to by buffer. For the argument descriptions, 2078 see Objects/stringlib/localeutil.h */ 2079#ifndef Py_LIMITED_API 2080PyAPI_FUNC(Py_ssize_t) _PyUnicode_InsertThousandsGrouping( 2081 PyObject *unicode, 2082 Py_ssize_t index, 2083 Py_ssize_t n_buffer, 2084 void *digits, 2085 Py_ssize_t n_digits, 2086 Py_ssize_t min_width, 2087 const char *grouping, 2088 PyObject *thousands_sep, 2089 Py_UCS4 *maxchar); 2090#endif 2091/* === Characters Type APIs =============================================== */ 2092 2093/* Helper array used by Py_UNICODE_ISSPACE(). */ 2094 2095#ifndef Py_LIMITED_API 2096PyAPI_DATA(const unsigned char) _Py_ascii_whitespace[]; 2097 2098/* These should not be used directly. Use the Py_UNICODE_IS* and 2099 Py_UNICODE_TO* macros instead. 2100 2101 These APIs are implemented in Objects/unicodectype.c. 2102 2103*/ 2104 2105PyAPI_FUNC(int) _PyUnicode_IsLowercase( 2106 Py_UCS4 ch /* Unicode character */ 2107 ); 2108 2109PyAPI_FUNC(int) _PyUnicode_IsUppercase( 2110 Py_UCS4 ch /* Unicode character */ 2111 ); 2112 2113PyAPI_FUNC(int) _PyUnicode_IsTitlecase( 2114 Py_UCS4 ch /* Unicode character */ 2115 ); 2116 2117PyAPI_FUNC(int) _PyUnicode_IsXidStart( 2118 Py_UCS4 ch /* Unicode character */ 2119 ); 2120 2121PyAPI_FUNC(int) _PyUnicode_IsXidContinue( 2122 Py_UCS4 ch /* Unicode character */ 2123 ); 2124 2125PyAPI_FUNC(int) _PyUnicode_IsWhitespace( 2126 const Py_UCS4 ch /* Unicode character */ 2127 ); 2128 2129PyAPI_FUNC(int) _PyUnicode_IsLinebreak( 2130 const Py_UCS4 ch /* Unicode character */ 2131 ); 2132 2133PyAPI_FUNC(Py_UCS4) _PyUnicode_ToLowercase( 2134 Py_UCS4 ch /* Unicode character */ 2135 ); 2136 2137PyAPI_FUNC(Py_UCS4) _PyUnicode_ToUppercase( 2138 Py_UCS4 ch /* Unicode character */ 2139 ); 2140 2141PyAPI_FUNC(Py_UCS4) _PyUnicode_ToTitlecase( 2142 Py_UCS4 ch /* Unicode character */ 2143 ); 2144 2145PyAPI_FUNC(int) _PyUnicode_ToLowerFull( 2146 Py_UCS4 ch, /* Unicode character */ 2147 Py_UCS4 *res 2148 ); 2149 2150PyAPI_FUNC(int) _PyUnicode_ToTitleFull( 2151 Py_UCS4 ch, /* Unicode character */ 2152 Py_UCS4 *res 2153 ); 2154 2155PyAPI_FUNC(int) _PyUnicode_ToUpperFull( 2156 Py_UCS4 ch, /* Unicode character */ 2157 Py_UCS4 *res 2158 ); 2159 2160PyAPI_FUNC(int) _PyUnicode_ToFoldedFull( 2161 Py_UCS4 ch, /* Unicode character */ 2162 Py_UCS4 *res 2163 ); 2164 2165PyAPI_FUNC(int) _PyUnicode_IsCaseIgnorable( 2166 Py_UCS4 ch /* Unicode character */ 2167 ); 2168 2169PyAPI_FUNC(int) _PyUnicode_IsCased( 2170 Py_UCS4 ch /* Unicode character */ 2171 ); 2172 2173PyAPI_FUNC(int) _PyUnicode_ToDecimalDigit( 2174 Py_UCS4 ch /* Unicode character */ 2175 ); 2176 2177PyAPI_FUNC(int) _PyUnicode_ToDigit( 2178 Py_UCS4 ch /* Unicode character */ 2179 ); 2180 2181PyAPI_FUNC(double) _PyUnicode_ToNumeric( 2182 Py_UCS4 ch /* Unicode character */ 2183 ); 2184 2185PyAPI_FUNC(int) _PyUnicode_IsDecimalDigit( 2186 Py_UCS4 ch /* Unicode character */ 2187 ); 2188 2189PyAPI_FUNC(int) _PyUnicode_IsDigit( 2190 Py_UCS4 ch /* Unicode character */ 2191 ); 2192 2193PyAPI_FUNC(int) _PyUnicode_IsNumeric( 2194 Py_UCS4 ch /* Unicode character */ 2195 ); 2196 2197PyAPI_FUNC(int) _PyUnicode_IsPrintable( 2198 Py_UCS4 ch /* Unicode character */ 2199 ); 2200 2201PyAPI_FUNC(int) _PyUnicode_IsAlpha( 2202 Py_UCS4 ch /* Unicode character */ 2203 ); 2204 2205PyAPI_FUNC(size_t) Py_UNICODE_strlen( 2206 const Py_UNICODE *u 2207 ); 2208 2209PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strcpy( 2210 Py_UNICODE *s1, 2211 const Py_UNICODE *s2); 2212 2213PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strcat( 2214 Py_UNICODE *s1, const Py_UNICODE *s2); 2215 2216PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strncpy( 2217 Py_UNICODE *s1, 2218 const Py_UNICODE *s2, 2219 size_t n); 2220 2221PyAPI_FUNC(int) Py_UNICODE_strcmp( 2222 const Py_UNICODE *s1, 2223 const Py_UNICODE *s2 2224 ); 2225 2226PyAPI_FUNC(int) Py_UNICODE_strncmp( 2227 const Py_UNICODE *s1, 2228 const Py_UNICODE *s2, 2229 size_t n 2230 ); 2231 2232PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strchr( 2233 const Py_UNICODE *s, 2234 Py_UNICODE c 2235 ); 2236 2237PyAPI_FUNC(Py_UNICODE*) Py_UNICODE_strrchr( 2238 const Py_UNICODE *s, 2239 Py_UNICODE c 2240 ); 2241 2242PyAPI_FUNC(PyObject*) _PyUnicode_FormatLong(PyObject *, int, int, int); 2243 2244/* Create a copy of a unicode string ending with a nul character. Return NULL 2245 and raise a MemoryError exception on memory allocation failure, otherwise 2246 return a new allocated buffer (use PyMem_Free() to free the buffer). */ 2247 2248PyAPI_FUNC(Py_UNICODE*) PyUnicode_AsUnicodeCopy( 2249 PyObject *unicode 2250 ); 2251#endif /* Py_LIMITED_API */ 2252 2253#if defined(Py_DEBUG) && !defined(Py_LIMITED_API) 2254PyAPI_FUNC(int) _PyUnicode_CheckConsistency( 2255 PyObject *op, 2256 int check_content); 2257#endif 2258 2259/* Return an interned Unicode object for an Identifier; may fail if there is no memory.*/ 2260PyAPI_FUNC(PyObject*) _PyUnicode_FromId(_Py_Identifier*); 2261/* Clear all static strings. */ 2262PyAPI_FUNC(void) _PyUnicode_ClearStaticStrings(void); 2263 2264#ifdef __cplusplus 2265} 2266#endif 2267#endif /* !Py_UNICODEOBJECT_H */ 2268