1 | #ifndef Py_CPYTHON_UNICODEOBJECT_H |
2 | # error "this header file must not be included directly" |
3 | #endif |
4 | |
5 | /* Py_UNICODE was the native Unicode storage format (code unit) used by |
6 | Python and represents a single Unicode element in the Unicode type. |
7 | With PEP 393, Py_UNICODE is deprecated and replaced with a |
8 | typedef to wchar_t. */ |
9 | #define PY_UNICODE_TYPE wchar_t |
10 | /* Py_DEPRECATED(3.3) */ typedef wchar_t Py_UNICODE; |
11 | |
12 | /* --- Internal Unicode Operations ---------------------------------------- */ |
13 | |
14 | #ifndef USE_UNICODE_WCHAR_CACHE |
15 | # define USE_UNICODE_WCHAR_CACHE 1 |
16 | #endif /* USE_UNICODE_WCHAR_CACHE */ |
17 | |
18 | /* Since splitting on whitespace is an important use case, and |
19 | whitespace in most situations is solely ASCII whitespace, we |
20 | optimize for the common case by using a quick look-up table |
21 | _Py_ascii_whitespace (see below) with an inlined check. |
22 | |
23 | */ |
24 | #define Py_UNICODE_ISSPACE(ch) \ |
25 | ((Py_UCS4)(ch) < 128U ? _Py_ascii_whitespace[(ch)] : _PyUnicode_IsWhitespace(ch)) |
26 | |
27 | #define Py_UNICODE_ISLOWER(ch) _PyUnicode_IsLowercase(ch) |
28 | #define Py_UNICODE_ISUPPER(ch) _PyUnicode_IsUppercase(ch) |
29 | #define Py_UNICODE_ISTITLE(ch) _PyUnicode_IsTitlecase(ch) |
30 | #define Py_UNICODE_ISLINEBREAK(ch) _PyUnicode_IsLinebreak(ch) |
31 | |
32 | #define Py_UNICODE_TOLOWER(ch) _PyUnicode_ToLowercase(ch) |
33 | #define Py_UNICODE_TOUPPER(ch) _PyUnicode_ToUppercase(ch) |
34 | #define Py_UNICODE_TOTITLE(ch) _PyUnicode_ToTitlecase(ch) |
35 | |
36 | #define Py_UNICODE_ISDECIMAL(ch) _PyUnicode_IsDecimalDigit(ch) |
37 | #define Py_UNICODE_ISDIGIT(ch) _PyUnicode_IsDigit(ch) |
38 | #define Py_UNICODE_ISNUMERIC(ch) _PyUnicode_IsNumeric(ch) |
39 | #define Py_UNICODE_ISPRINTABLE(ch) _PyUnicode_IsPrintable(ch) |
40 | |
41 | #define Py_UNICODE_TODECIMAL(ch) _PyUnicode_ToDecimalDigit(ch) |
42 | #define Py_UNICODE_TODIGIT(ch) _PyUnicode_ToDigit(ch) |
43 | #define Py_UNICODE_TONUMERIC(ch) _PyUnicode_ToNumeric(ch) |
44 | |
45 | #define Py_UNICODE_ISALPHA(ch) _PyUnicode_IsAlpha(ch) |
46 | |
47 | #define Py_UNICODE_ISALNUM(ch) \ |
48 | (Py_UNICODE_ISALPHA(ch) || \ |
49 | Py_UNICODE_ISDECIMAL(ch) || \ |
50 | Py_UNICODE_ISDIGIT(ch) || \ |
51 | Py_UNICODE_ISNUMERIC(ch)) |
52 | |
53 | Py_DEPRECATED(3.3) static inline void |
54 | Py_UNICODE_COPY(Py_UNICODE *target, const Py_UNICODE *source, Py_ssize_t length) { |
55 | memcpy(dest: target, src: source, n: (size_t)(length) * sizeof(Py_UNICODE)); |
56 | } |
57 | |
58 | Py_DEPRECATED(3.3) static inline void |
59 | Py_UNICODE_FILL(Py_UNICODE *target, Py_UNICODE value, Py_ssize_t length) { |
60 | Py_ssize_t i; |
61 | for (i = 0; i < length; i++) { |
62 | target[i] = value; |
63 | } |
64 | } |
65 | |
66 | /* macros to work with surrogates */ |
67 | #define Py_UNICODE_IS_SURROGATE(ch) (0xD800 <= (ch) && (ch) <= 0xDFFF) |
68 | #define Py_UNICODE_IS_HIGH_SURROGATE(ch) (0xD800 <= (ch) && (ch) <= 0xDBFF) |
69 | #define Py_UNICODE_IS_LOW_SURROGATE(ch) (0xDC00 <= (ch) && (ch) <= 0xDFFF) |
70 | /* Join two surrogate characters and return a single Py_UCS4 value. */ |
71 | #define Py_UNICODE_JOIN_SURROGATES(high, low) \ |
72 | (((((Py_UCS4)(high) & 0x03FF) << 10) | \ |
73 | ((Py_UCS4)(low) & 0x03FF)) + 0x10000) |
74 | /* high surrogate = top 10 bits added to D800 */ |
75 | #define Py_UNICODE_HIGH_SURROGATE(ch) (0xD800 - (0x10000 >> 10) + ((ch) >> 10)) |
76 | /* low surrogate = bottom 10 bits added to DC00 */ |
77 | #define Py_UNICODE_LOW_SURROGATE(ch) (0xDC00 + ((ch) & 0x3FF)) |
78 | |
79 | /* --- Unicode Type ------------------------------------------------------- */ |
80 | |
81 | /* ASCII-only strings created through PyUnicode_New use the PyASCIIObject |
82 | structure. state.ascii and state.compact are set, and the data |
83 | immediately follow the structure. utf8_length and wstr_length can be found |
84 | in the length field; the utf8 pointer is equal to the data pointer. */ |
85 | typedef struct { |
86 | /* There are 4 forms of Unicode strings: |
87 | |
88 | - compact ascii: |
89 | |
90 | * structure = PyASCIIObject |
91 | * test: PyUnicode_IS_COMPACT_ASCII(op) |
92 | * kind = PyUnicode_1BYTE_KIND |
93 | * compact = 1 |
94 | * ascii = 1 |
95 | * ready = 1 |
96 | * (length is the length of the utf8 and wstr strings) |
97 | * (data starts just after the structure) |
98 | * (since ASCII is decoded from UTF-8, the utf8 string are the data) |
99 | |
100 | - compact: |
101 | |
102 | * structure = PyCompactUnicodeObject |
103 | * test: PyUnicode_IS_COMPACT(op) && !PyUnicode_IS_ASCII(op) |
104 | * kind = PyUnicode_1BYTE_KIND, PyUnicode_2BYTE_KIND or |
105 | PyUnicode_4BYTE_KIND |
106 | * compact = 1 |
107 | * ready = 1 |
108 | * ascii = 0 |
109 | * utf8 is not shared with data |
110 | * utf8_length = 0 if utf8 is NULL |
111 | * wstr is shared with data and wstr_length=length |
112 | if kind=PyUnicode_2BYTE_KIND and sizeof(wchar_t)=2 |
113 | or if kind=PyUnicode_4BYTE_KIND and sizeof(wchar_t)=4 |
114 | * wstr_length = 0 if wstr is NULL |
115 | * (data starts just after the structure) |
116 | |
117 | - legacy string, not ready: |
118 | |
119 | * structure = PyUnicodeObject |
120 | * test: kind == PyUnicode_WCHAR_KIND |
121 | * length = 0 (use wstr_length) |
122 | * hash = -1 |
123 | * kind = PyUnicode_WCHAR_KIND |
124 | * compact = 0 |
125 | * ascii = 0 |
126 | * ready = 0 |
127 | * interned = SSTATE_NOT_INTERNED |
128 | * wstr is not NULL |
129 | * data.any is NULL |
130 | * utf8 is NULL |
131 | * utf8_length = 0 |
132 | |
133 | - legacy string, ready: |
134 | |
135 | * structure = PyUnicodeObject structure |
136 | * test: !PyUnicode_IS_COMPACT(op) && kind != PyUnicode_WCHAR_KIND |
137 | * kind = PyUnicode_1BYTE_KIND, PyUnicode_2BYTE_KIND or |
138 | PyUnicode_4BYTE_KIND |
139 | * compact = 0 |
140 | * ready = 1 |
141 | * data.any is not NULL |
142 | * utf8 is shared and utf8_length = length with data.any if ascii = 1 |
143 | * utf8_length = 0 if utf8 is NULL |
144 | * wstr is shared with data.any and wstr_length = length |
145 | if kind=PyUnicode_2BYTE_KIND and sizeof(wchar_t)=2 |
146 | or if kind=PyUnicode_4BYTE_KIND and sizeof(wchar_4)=4 |
147 | * wstr_length = 0 if wstr is NULL |
148 | |
149 | Compact strings use only one memory block (structure + characters), |
150 | whereas legacy strings use one block for the structure and one block |
151 | for characters. |
152 | |
153 | Legacy strings are created by PyUnicode_FromUnicode() and |
154 | PyUnicode_FromStringAndSize(NULL, size) functions. They become ready |
155 | when PyUnicode_READY() is called. |
156 | |
157 | See also _PyUnicode_CheckConsistency(). |
158 | */ |
159 | PyObject_HEAD |
160 | Py_ssize_t length; /* Number of code points in the string */ |
161 | Py_hash_t hash; /* Hash value; -1 if not set */ |
162 | struct { |
163 | /* |
164 | SSTATE_NOT_INTERNED (0) |
165 | SSTATE_INTERNED_MORTAL (1) |
166 | SSTATE_INTERNED_IMMORTAL (2) |
167 | |
168 | If interned != SSTATE_NOT_INTERNED, the two references from the |
169 | dictionary to this object are *not* counted in ob_refcnt. |
170 | */ |
171 | unsigned int interned:2; |
172 | /* Character size: |
173 | |
174 | - PyUnicode_WCHAR_KIND (0): |
175 | |
176 | * character type = wchar_t (16 or 32 bits, depending on the |
177 | platform) |
178 | |
179 | - PyUnicode_1BYTE_KIND (1): |
180 | |
181 | * character type = Py_UCS1 (8 bits, unsigned) |
182 | * all characters are in the range U+0000-U+00FF (latin1) |
183 | * if ascii is set, all characters are in the range U+0000-U+007F |
184 | (ASCII), otherwise at least one character is in the range |
185 | U+0080-U+00FF |
186 | |
187 | - PyUnicode_2BYTE_KIND (2): |
188 | |
189 | * character type = Py_UCS2 (16 bits, unsigned) |
190 | * all characters are in the range U+0000-U+FFFF (BMP) |
191 | * at least one character is in the range U+0100-U+FFFF |
192 | |
193 | - PyUnicode_4BYTE_KIND (4): |
194 | |
195 | * character type = Py_UCS4 (32 bits, unsigned) |
196 | * all characters are in the range U+0000-U+10FFFF |
197 | * at least one character is in the range U+10000-U+10FFFF |
198 | */ |
199 | unsigned int kind:3; |
200 | /* Compact is with respect to the allocation scheme. Compact unicode |
201 | objects only require one memory block while non-compact objects use |
202 | one block for the PyUnicodeObject struct and another for its data |
203 | buffer. */ |
204 | unsigned int compact:1; |
205 | /* The string only contains characters in the range U+0000-U+007F (ASCII) |
206 | and the kind is PyUnicode_1BYTE_KIND. If ascii is set and compact is |
207 | set, use the PyASCIIObject structure. */ |
208 | unsigned int ascii:1; |
209 | /* The ready flag indicates whether the object layout is initialized |
210 | completely. This means that this is either a compact object, or |
211 | the data pointer is filled out. The bit is redundant, and helps |
212 | to minimize the test in PyUnicode_IS_READY(). */ |
213 | unsigned int ready:1; |
214 | /* Padding to ensure that PyUnicode_DATA() is always aligned to |
215 | 4 bytes (see issue #19537 on m68k). */ |
216 | unsigned int :24; |
217 | } state; |
218 | wchar_t *wstr; /* wchar_t representation (null-terminated) */ |
219 | } PyASCIIObject; |
220 | |
221 | /* Non-ASCII strings allocated through PyUnicode_New use the |
222 | PyCompactUnicodeObject structure. state.compact is set, and the data |
223 | immediately follow the structure. */ |
224 | typedef struct { |
225 | PyASCIIObject _base; |
226 | Py_ssize_t utf8_length; /* Number of bytes in utf8, excluding the |
227 | * terminating \0. */ |
228 | char *utf8; /* UTF-8 representation (null-terminated) */ |
229 | Py_ssize_t wstr_length; /* Number of code points in wstr, possible |
230 | * surrogates count as two code points. */ |
231 | } PyCompactUnicodeObject; |
232 | |
233 | /* Strings allocated through PyUnicode_FromUnicode(NULL, len) use the |
234 | PyUnicodeObject structure. The actual string data is initially in the wstr |
235 | block, and copied into the data block using _PyUnicode_Ready. */ |
236 | typedef struct { |
237 | PyCompactUnicodeObject _base; |
238 | union { |
239 | void *any; |
240 | Py_UCS1 *latin1; |
241 | Py_UCS2 *ucs2; |
242 | Py_UCS4 *ucs4; |
243 | } data; /* Canonical, smallest-form Unicode buffer */ |
244 | } PyUnicodeObject; |
245 | |
246 | PyAPI_FUNC(int) _PyUnicode_CheckConsistency( |
247 | PyObject *op, |
248 | int check_content); |
249 | |
250 | /* Fast access macros */ |
251 | |
252 | /* Returns the deprecated Py_UNICODE representation's size in code units |
253 | (this includes surrogate pairs as 2 units). |
254 | If the Py_UNICODE representation is not available, it will be computed |
255 | on request. Use PyUnicode_GET_LENGTH() for the length in code points. */ |
256 | |
257 | /* Py_DEPRECATED(3.3) */ |
258 | #define PyUnicode_GET_SIZE(op) \ |
259 | (assert(PyUnicode_Check(op)), \ |
260 | (((PyASCIIObject *)(op))->wstr) ? \ |
261 | PyUnicode_WSTR_LENGTH(op) : \ |
262 | ((void)PyUnicode_AsUnicode(_PyObject_CAST(op)),\ |
263 | assert(((PyASCIIObject *)(op))->wstr), \ |
264 | PyUnicode_WSTR_LENGTH(op))) |
265 | |
266 | /* Py_DEPRECATED(3.3) */ |
267 | #define PyUnicode_GET_DATA_SIZE(op) \ |
268 | (PyUnicode_GET_SIZE(op) * Py_UNICODE_SIZE) |
269 | |
270 | /* Alias for PyUnicode_AsUnicode(). This will create a wchar_t/Py_UNICODE |
271 | representation on demand. Using this macro is very inefficient now, |
272 | try to port your code to use the new PyUnicode_*BYTE_DATA() macros or |
273 | use PyUnicode_WRITE() and PyUnicode_READ(). */ |
274 | |
275 | /* Py_DEPRECATED(3.3) */ |
276 | #define PyUnicode_AS_UNICODE(op) \ |
277 | (assert(PyUnicode_Check(op)), \ |
278 | (((PyASCIIObject *)(op))->wstr) ? (((PyASCIIObject *)(op))->wstr) : \ |
279 | PyUnicode_AsUnicode(_PyObject_CAST(op))) |
280 | |
281 | /* Py_DEPRECATED(3.3) */ |
282 | #define PyUnicode_AS_DATA(op) \ |
283 | ((const char *)(PyUnicode_AS_UNICODE(op))) |
284 | |
285 | |
286 | /* --- Flexible String Representation Helper Macros (PEP 393) -------------- */ |
287 | |
288 | /* Values for PyASCIIObject.state: */ |
289 | |
290 | /* Interning state. */ |
291 | #define SSTATE_NOT_INTERNED 0 |
292 | #define SSTATE_INTERNED_MORTAL 1 |
293 | #define SSTATE_INTERNED_IMMORTAL 2 |
294 | |
295 | /* Return true if the string contains only ASCII characters, or 0 if not. The |
296 | string may be compact (PyUnicode_IS_COMPACT_ASCII) or not, but must be |
297 | ready. */ |
298 | #define PyUnicode_IS_ASCII(op) \ |
299 | (assert(PyUnicode_Check(op)), \ |
300 | assert(PyUnicode_IS_READY(op)), \ |
301 | ((PyASCIIObject*)op)->state.ascii) |
302 | |
303 | /* Return true if the string is compact or 0 if not. |
304 | No type checks or Ready calls are performed. */ |
305 | #define PyUnicode_IS_COMPACT(op) \ |
306 | (((PyASCIIObject*)(op))->state.compact) |
307 | |
308 | /* Return true if the string is a compact ASCII string (use PyASCIIObject |
309 | structure), or 0 if not. No type checks or Ready calls are performed. */ |
310 | #define PyUnicode_IS_COMPACT_ASCII(op) \ |
311 | (((PyASCIIObject*)op)->state.ascii && PyUnicode_IS_COMPACT(op)) |
312 | |
313 | enum PyUnicode_Kind { |
314 | /* String contains only wstr byte characters. This is only possible |
315 | when the string was created with a legacy API and _PyUnicode_Ready() |
316 | has not been called yet. */ |
317 | PyUnicode_WCHAR_KIND = 0, |
318 | /* Return values of the PyUnicode_KIND() macro: */ |
319 | PyUnicode_1BYTE_KIND = 1, |
320 | PyUnicode_2BYTE_KIND = 2, |
321 | PyUnicode_4BYTE_KIND = 4 |
322 | }; |
323 | |
324 | /* Return pointers to the canonical representation cast to unsigned char, |
325 | Py_UCS2, or Py_UCS4 for direct character access. |
326 | No checks are performed, use PyUnicode_KIND() before to ensure |
327 | these will work correctly. */ |
328 | |
329 | #define PyUnicode_1BYTE_DATA(op) ((Py_UCS1*)PyUnicode_DATA(op)) |
330 | #define PyUnicode_2BYTE_DATA(op) ((Py_UCS2*)PyUnicode_DATA(op)) |
331 | #define PyUnicode_4BYTE_DATA(op) ((Py_UCS4*)PyUnicode_DATA(op)) |
332 | |
333 | /* Return one of the PyUnicode_*_KIND values defined above. */ |
334 | #define PyUnicode_KIND(op) \ |
335 | (assert(PyUnicode_Check(op)), \ |
336 | assert(PyUnicode_IS_READY(op)), \ |
337 | ((PyASCIIObject *)(op))->state.kind) |
338 | |
339 | /* Return a void pointer to the raw unicode buffer. */ |
340 | #define _PyUnicode_COMPACT_DATA(op) \ |
341 | (PyUnicode_IS_ASCII(op) ? \ |
342 | ((void*)((PyASCIIObject*)(op) + 1)) : \ |
343 | ((void*)((PyCompactUnicodeObject*)(op) + 1))) |
344 | |
345 | #define _PyUnicode_NONCOMPACT_DATA(op) \ |
346 | (assert(((PyUnicodeObject*)(op))->data.any), \ |
347 | ((((PyUnicodeObject *)(op))->data.any))) |
348 | |
349 | #define PyUnicode_DATA(op) \ |
350 | (assert(PyUnicode_Check(op)), \ |
351 | PyUnicode_IS_COMPACT(op) ? _PyUnicode_COMPACT_DATA(op) : \ |
352 | _PyUnicode_NONCOMPACT_DATA(op)) |
353 | |
354 | /* In the access macros below, "kind" may be evaluated more than once. |
355 | All other macro parameters are evaluated exactly once, so it is safe |
356 | to put side effects into them (such as increasing the index). */ |
357 | |
358 | /* Write into the canonical representation, this macro does not do any sanity |
359 | checks and is intended for usage in loops. The caller should cache the |
360 | kind and data pointers obtained from other macro calls. |
361 | index is the index in the string (starts at 0) and value is the new |
362 | code point value which should be written to that location. */ |
363 | #define PyUnicode_WRITE(kind, data, index, value) \ |
364 | do { \ |
365 | switch ((kind)) { \ |
366 | case PyUnicode_1BYTE_KIND: { \ |
367 | ((Py_UCS1 *)(data))[(index)] = (Py_UCS1)(value); \ |
368 | break; \ |
369 | } \ |
370 | case PyUnicode_2BYTE_KIND: { \ |
371 | ((Py_UCS2 *)(data))[(index)] = (Py_UCS2)(value); \ |
372 | break; \ |
373 | } \ |
374 | default: { \ |
375 | assert((kind) == PyUnicode_4BYTE_KIND); \ |
376 | ((Py_UCS4 *)(data))[(index)] = (Py_UCS4)(value); \ |
377 | } \ |
378 | } \ |
379 | } while (0) |
380 | |
381 | /* Read a code point from the string's canonical representation. No checks |
382 | or ready calls are performed. */ |
383 | #define PyUnicode_READ(kind, data, index) \ |
384 | ((Py_UCS4) \ |
385 | ((kind) == PyUnicode_1BYTE_KIND ? \ |
386 | ((const Py_UCS1 *)(data))[(index)] : \ |
387 | ((kind) == PyUnicode_2BYTE_KIND ? \ |
388 | ((const Py_UCS2 *)(data))[(index)] : \ |
389 | ((const Py_UCS4 *)(data))[(index)] \ |
390 | ) \ |
391 | )) |
392 | |
393 | /* PyUnicode_READ_CHAR() is less efficient than PyUnicode_READ() because it |
394 | calls PyUnicode_KIND() and might call it twice. For single reads, use |
395 | PyUnicode_READ_CHAR, for multiple consecutive reads callers should |
396 | cache kind and use PyUnicode_READ instead. */ |
397 | #define PyUnicode_READ_CHAR(unicode, index) \ |
398 | (assert(PyUnicode_Check(unicode)), \ |
399 | assert(PyUnicode_IS_READY(unicode)), \ |
400 | (Py_UCS4) \ |
401 | (PyUnicode_KIND((unicode)) == PyUnicode_1BYTE_KIND ? \ |
402 | ((const Py_UCS1 *)(PyUnicode_DATA((unicode))))[(index)] : \ |
403 | (PyUnicode_KIND((unicode)) == PyUnicode_2BYTE_KIND ? \ |
404 | ((const Py_UCS2 *)(PyUnicode_DATA((unicode))))[(index)] : \ |
405 | ((const Py_UCS4 *)(PyUnicode_DATA((unicode))))[(index)] \ |
406 | ) \ |
407 | )) |
408 | |
409 | /* Returns the length of the unicode string. The caller has to make sure that |
410 | the string has it's canonical representation set before calling |
411 | this macro. Call PyUnicode_(FAST_)Ready to ensure that. */ |
412 | #define PyUnicode_GET_LENGTH(op) \ |
413 | (assert(PyUnicode_Check(op)), \ |
414 | assert(PyUnicode_IS_READY(op)), \ |
415 | ((PyASCIIObject *)(op))->length) |
416 | |
417 | |
418 | /* Fast check to determine whether an object is ready. Equivalent to |
419 | PyUnicode_IS_COMPACT(op) || ((PyUnicodeObject*)(op))->data.any */ |
420 | |
421 | #define PyUnicode_IS_READY(op) (((PyASCIIObject*)op)->state.ready) |
422 | |
423 | /* PyUnicode_READY() does less work than _PyUnicode_Ready() in the best |
424 | case. If the canonical representation is not yet set, it will still call |
425 | _PyUnicode_Ready(). |
426 | Returns 0 on success and -1 on errors. */ |
427 | #define PyUnicode_READY(op) \ |
428 | (assert(PyUnicode_Check(op)), \ |
429 | (PyUnicode_IS_READY(op) ? \ |
430 | 0 : _PyUnicode_Ready(_PyObject_CAST(op)))) |
431 | |
432 | /* Return a maximum character value which is suitable for creating another |
433 | string based on op. This is always an approximation but more efficient |
434 | than iterating over the string. */ |
435 | #define PyUnicode_MAX_CHAR_VALUE(op) \ |
436 | (assert(PyUnicode_IS_READY(op)), \ |
437 | (PyUnicode_IS_ASCII(op) ? \ |
438 | (0x7f) : \ |
439 | (PyUnicode_KIND(op) == PyUnicode_1BYTE_KIND ? \ |
440 | (0xffU) : \ |
441 | (PyUnicode_KIND(op) == PyUnicode_2BYTE_KIND ? \ |
442 | (0xffffU) : \ |
443 | (0x10ffffU))))) |
444 | |
445 | Py_DEPRECATED(3.3) |
446 | static inline Py_ssize_t _PyUnicode_get_wstr_length(PyObject *op) { |
447 | return PyUnicode_IS_COMPACT_ASCII(op) ? |
448 | ((PyASCIIObject*)op)->length : |
449 | ((PyCompactUnicodeObject*)op)->wstr_length; |
450 | } |
451 | #define PyUnicode_WSTR_LENGTH(op) _PyUnicode_get_wstr_length((PyObject*)op) |
452 | |
453 | /* === Public API ========================================================= */ |
454 | |
455 | /* --- Plain Py_UNICODE --------------------------------------------------- */ |
456 | |
457 | /* With PEP 393, this is the recommended way to allocate a new unicode object. |
458 | This function will allocate the object and its buffer in a single memory |
459 | block. Objects created using this function are not resizable. */ |
460 | PyAPI_FUNC(PyObject*) PyUnicode_New( |
461 | Py_ssize_t size, /* Number of code points in the new string */ |
462 | Py_UCS4 maxchar /* maximum code point value in the string */ |
463 | ); |
464 | |
465 | /* Initializes the canonical string representation from the deprecated |
466 | wstr/Py_UNICODE representation. This function is used to convert Unicode |
467 | objects which were created using the old API to the new flexible format |
468 | introduced with PEP 393. |
469 | |
470 | Don't call this function directly, use the public PyUnicode_READY() macro |
471 | instead. */ |
472 | PyAPI_FUNC(int) _PyUnicode_Ready( |
473 | PyObject *unicode /* Unicode object */ |
474 | ); |
475 | |
476 | /* Get a copy of a Unicode string. */ |
477 | PyAPI_FUNC(PyObject*) _PyUnicode_Copy( |
478 | PyObject *unicode |
479 | ); |
480 | |
481 | /* Copy character from one unicode object into another, this function performs |
482 | character conversion when necessary and falls back to memcpy() if possible. |
483 | |
484 | Fail if to is too small (smaller than *how_many* or smaller than |
485 | len(from)-from_start), or if kind(from[from_start:from_start+how_many]) > |
486 | kind(to), or if *to* has more than 1 reference. |
487 | |
488 | Return the number of written character, or return -1 and raise an exception |
489 | on error. |
490 | |
491 | Pseudo-code: |
492 | |
493 | how_many = min(how_many, len(from) - from_start) |
494 | to[to_start:to_start+how_many] = from[from_start:from_start+how_many] |
495 | return how_many |
496 | |
497 | Note: The function doesn't write a terminating null character. |
498 | */ |
499 | PyAPI_FUNC(Py_ssize_t) PyUnicode_CopyCharacters( |
500 | PyObject *to, |
501 | Py_ssize_t to_start, |
502 | PyObject *from, |
503 | Py_ssize_t from_start, |
504 | Py_ssize_t how_many |
505 | ); |
506 | |
507 | /* Unsafe version of PyUnicode_CopyCharacters(): don't check arguments and so |
508 | may crash if parameters are invalid (e.g. if the output string |
509 | is too short). */ |
510 | PyAPI_FUNC(void) _PyUnicode_FastCopyCharacters( |
511 | PyObject *to, |
512 | Py_ssize_t to_start, |
513 | PyObject *from, |
514 | Py_ssize_t from_start, |
515 | Py_ssize_t how_many |
516 | ); |
517 | |
518 | /* Fill a string with a character: write fill_char into |
519 | unicode[start:start+length]. |
520 | |
521 | Fail if fill_char is bigger than the string maximum character, or if the |
522 | string has more than 1 reference. |
523 | |
524 | Return the number of written character, or return -1 and raise an exception |
525 | on error. */ |
526 | PyAPI_FUNC(Py_ssize_t) PyUnicode_Fill( |
527 | PyObject *unicode, |
528 | Py_ssize_t start, |
529 | Py_ssize_t length, |
530 | Py_UCS4 fill_char |
531 | ); |
532 | |
533 | /* Unsafe version of PyUnicode_Fill(): don't check arguments and so may crash |
534 | if parameters are invalid (e.g. if length is longer than the string). */ |
535 | PyAPI_FUNC(void) _PyUnicode_FastFill( |
536 | PyObject *unicode, |
537 | Py_ssize_t start, |
538 | Py_ssize_t length, |
539 | Py_UCS4 fill_char |
540 | ); |
541 | |
542 | /* Create a Unicode Object from the Py_UNICODE buffer u of the given |
543 | size. |
544 | |
545 | u may be NULL which causes the contents to be undefined. It is the |
546 | user's responsibility to fill in the needed data afterwards. Note |
547 | that modifying the Unicode object contents after construction is |
548 | only allowed if u was set to NULL. |
549 | |
550 | The buffer is copied into the new object. */ |
551 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_FromUnicode( |
552 | const Py_UNICODE *u, /* Unicode buffer */ |
553 | Py_ssize_t size /* size of buffer */ |
554 | ); |
555 | |
556 | /* Create a new string from a buffer of Py_UCS1, Py_UCS2 or Py_UCS4 characters. |
557 | Scan the string to find the maximum character. */ |
558 | PyAPI_FUNC(PyObject*) PyUnicode_FromKindAndData( |
559 | int kind, |
560 | const void *buffer, |
561 | Py_ssize_t size); |
562 | |
563 | /* Create a new string from a buffer of ASCII characters. |
564 | WARNING: Don't check if the string contains any non-ASCII character. */ |
565 | PyAPI_FUNC(PyObject*) _PyUnicode_FromASCII( |
566 | const char *buffer, |
567 | Py_ssize_t size); |
568 | |
569 | /* Compute the maximum character of the substring unicode[start:end]. |
570 | Return 127 for an empty string. */ |
571 | PyAPI_FUNC(Py_UCS4) _PyUnicode_FindMaxChar ( |
572 | PyObject *unicode, |
573 | Py_ssize_t start, |
574 | Py_ssize_t end); |
575 | |
576 | /* Return a read-only pointer to the Unicode object's internal |
577 | Py_UNICODE buffer. |
578 | If the wchar_t/Py_UNICODE representation is not yet available, this |
579 | function will calculate it. */ |
580 | Py_DEPRECATED(3.3) PyAPI_FUNC(Py_UNICODE *) PyUnicode_AsUnicode( |
581 | PyObject *unicode /* Unicode object */ |
582 | ); |
583 | |
584 | /* Similar to PyUnicode_AsUnicode(), but raises a ValueError if the string |
585 | contains null characters. */ |
586 | PyAPI_FUNC(const Py_UNICODE *) _PyUnicode_AsUnicode( |
587 | PyObject *unicode /* Unicode object */ |
588 | ); |
589 | |
590 | /* Return a read-only pointer to the Unicode object's internal |
591 | Py_UNICODE buffer and save the length at size. |
592 | If the wchar_t/Py_UNICODE representation is not yet available, this |
593 | function will calculate it. */ |
594 | |
595 | Py_DEPRECATED(3.3) PyAPI_FUNC(Py_UNICODE *) PyUnicode_AsUnicodeAndSize( |
596 | PyObject *unicode, /* Unicode object */ |
597 | Py_ssize_t *size /* location where to save the length */ |
598 | ); |
599 | |
600 | |
601 | /* --- _PyUnicodeWriter API ----------------------------------------------- */ |
602 | |
603 | typedef struct { |
604 | PyObject *buffer; |
605 | void *data; |
606 | enum PyUnicode_Kind kind; |
607 | Py_UCS4 maxchar; |
608 | Py_ssize_t size; |
609 | Py_ssize_t pos; |
610 | |
611 | /* minimum number of allocated characters (default: 0) */ |
612 | Py_ssize_t min_length; |
613 | |
614 | /* minimum character (default: 127, ASCII) */ |
615 | Py_UCS4 min_char; |
616 | |
617 | /* If non-zero, overallocate the buffer (default: 0). */ |
618 | unsigned char overallocate; |
619 | |
620 | /* If readonly is 1, buffer is a shared string (cannot be modified) |
621 | and size is set to 0. */ |
622 | unsigned char readonly; |
623 | } _PyUnicodeWriter ; |
624 | |
625 | /* Initialize a Unicode writer. |
626 | * |
627 | * By default, the minimum buffer size is 0 character and overallocation is |
628 | * disabled. Set min_length, min_char and overallocate attributes to control |
629 | * the allocation of the buffer. */ |
630 | PyAPI_FUNC(void) |
631 | _PyUnicodeWriter_Init(_PyUnicodeWriter *writer); |
632 | |
633 | /* Prepare the buffer to write 'length' characters |
634 | with the specified maximum character. |
635 | |
636 | Return 0 on success, raise an exception and return -1 on error. */ |
637 | #define _PyUnicodeWriter_Prepare(WRITER, LENGTH, MAXCHAR) \ |
638 | (((MAXCHAR) <= (WRITER)->maxchar \ |
639 | && (LENGTH) <= (WRITER)->size - (WRITER)->pos) \ |
640 | ? 0 \ |
641 | : (((LENGTH) == 0) \ |
642 | ? 0 \ |
643 | : _PyUnicodeWriter_PrepareInternal((WRITER), (LENGTH), (MAXCHAR)))) |
644 | |
645 | /* Don't call this function directly, use the _PyUnicodeWriter_Prepare() macro |
646 | instead. */ |
647 | PyAPI_FUNC(int) |
648 | _PyUnicodeWriter_PrepareInternal(_PyUnicodeWriter *writer, |
649 | Py_ssize_t length, Py_UCS4 maxchar); |
650 | |
651 | /* Prepare the buffer to have at least the kind KIND. |
652 | For example, kind=PyUnicode_2BYTE_KIND ensures that the writer will |
653 | support characters in range U+000-U+FFFF. |
654 | |
655 | Return 0 on success, raise an exception and return -1 on error. */ |
656 | #define _PyUnicodeWriter_PrepareKind(WRITER, KIND) \ |
657 | (assert((KIND) != PyUnicode_WCHAR_KIND), \ |
658 | (KIND) <= (WRITER)->kind \ |
659 | ? 0 \ |
660 | : _PyUnicodeWriter_PrepareKindInternal((WRITER), (KIND))) |
661 | |
662 | /* Don't call this function directly, use the _PyUnicodeWriter_PrepareKind() |
663 | macro instead. */ |
664 | PyAPI_FUNC(int) |
665 | _PyUnicodeWriter_PrepareKindInternal(_PyUnicodeWriter *writer, |
666 | enum PyUnicode_Kind kind); |
667 | |
668 | /* Append a Unicode character. |
669 | Return 0 on success, raise an exception and return -1 on error. */ |
670 | PyAPI_FUNC(int) |
671 | _PyUnicodeWriter_WriteChar(_PyUnicodeWriter *writer, |
672 | Py_UCS4 ch |
673 | ); |
674 | |
675 | /* Append a Unicode string. |
676 | Return 0 on success, raise an exception and return -1 on error. */ |
677 | PyAPI_FUNC(int) |
678 | _PyUnicodeWriter_WriteStr(_PyUnicodeWriter *writer, |
679 | PyObject *str /* Unicode string */ |
680 | ); |
681 | |
682 | /* Append a substring of a Unicode string. |
683 | Return 0 on success, raise an exception and return -1 on error. */ |
684 | PyAPI_FUNC(int) |
685 | _PyUnicodeWriter_WriteSubstring(_PyUnicodeWriter *writer, |
686 | PyObject *str, /* Unicode string */ |
687 | Py_ssize_t start, |
688 | Py_ssize_t end |
689 | ); |
690 | |
691 | /* Append an ASCII-encoded byte string. |
692 | Return 0 on success, raise an exception and return -1 on error. */ |
693 | PyAPI_FUNC(int) |
694 | _PyUnicodeWriter_WriteASCIIString(_PyUnicodeWriter *writer, |
695 | const char *str, /* ASCII-encoded byte string */ |
696 | Py_ssize_t len /* number of bytes, or -1 if unknown */ |
697 | ); |
698 | |
699 | /* Append a latin1-encoded byte string. |
700 | Return 0 on success, raise an exception and return -1 on error. */ |
701 | PyAPI_FUNC(int) |
702 | _PyUnicodeWriter_WriteLatin1String(_PyUnicodeWriter *writer, |
703 | const char *str, /* latin1-encoded byte string */ |
704 | Py_ssize_t len /* length in bytes */ |
705 | ); |
706 | |
707 | /* Get the value of the writer as a Unicode string. Clear the |
708 | buffer of the writer. Raise an exception and return NULL |
709 | on error. */ |
710 | PyAPI_FUNC(PyObject *) |
711 | _PyUnicodeWriter_Finish(_PyUnicodeWriter *writer); |
712 | |
713 | /* Deallocate memory of a writer (clear its internal buffer). */ |
714 | PyAPI_FUNC(void) |
715 | _PyUnicodeWriter_Dealloc(_PyUnicodeWriter *writer); |
716 | |
717 | |
718 | /* Format the object based on the format_spec, as defined in PEP 3101 |
719 | (Advanced String Formatting). */ |
720 | PyAPI_FUNC(int) _PyUnicode_FormatAdvancedWriter( |
721 | _PyUnicodeWriter *writer, |
722 | PyObject *obj, |
723 | PyObject *format_spec, |
724 | Py_ssize_t start, |
725 | Py_ssize_t end); |
726 | |
727 | /* --- Manage the default encoding ---------------------------------------- */ |
728 | |
729 | /* Returns a pointer to the default encoding (UTF-8) of the |
730 | Unicode object unicode. |
731 | |
732 | Like PyUnicode_AsUTF8AndSize(), this also caches the UTF-8 representation |
733 | in the unicodeobject. |
734 | |
735 | _PyUnicode_AsString is a #define for PyUnicode_AsUTF8 to |
736 | support the previous internal function with the same behaviour. |
737 | |
738 | Use of this API is DEPRECATED since no size information can be |
739 | extracted from the returned data. |
740 | */ |
741 | |
742 | PyAPI_FUNC(const char *) PyUnicode_AsUTF8(PyObject *unicode); |
743 | |
744 | #define _PyUnicode_AsString PyUnicode_AsUTF8 |
745 | |
746 | /* --- Generic Codecs ----------------------------------------------------- */ |
747 | |
748 | /* Encodes a Py_UNICODE buffer of the given size and returns a |
749 | Python string object. */ |
750 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_Encode( |
751 | const Py_UNICODE *s, /* Unicode char buffer */ |
752 | Py_ssize_t size, /* number of Py_UNICODE chars to encode */ |
753 | const char *encoding, /* encoding */ |
754 | const char *errors /* error handling */ |
755 | ); |
756 | |
757 | /* --- UTF-7 Codecs ------------------------------------------------------- */ |
758 | |
759 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF7( |
760 | const Py_UNICODE *data, /* Unicode char buffer */ |
761 | Py_ssize_t length, /* number of Py_UNICODE chars to encode */ |
762 | int base64SetO, /* Encode RFC2152 Set O characters in base64 */ |
763 | int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ |
764 | const char *errors /* error handling */ |
765 | ); |
766 | |
767 | PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF7( |
768 | PyObject *unicode, /* Unicode object */ |
769 | int base64SetO, /* Encode RFC2152 Set O characters in base64 */ |
770 | int base64WhiteSpace, /* Encode whitespace (sp, ht, nl, cr) in base64 */ |
771 | const char *errors /* error handling */ |
772 | ); |
773 | |
774 | /* --- UTF-8 Codecs ------------------------------------------------------- */ |
775 | |
776 | PyAPI_FUNC(PyObject*) _PyUnicode_AsUTF8String( |
777 | PyObject *unicode, |
778 | const char *errors); |
779 | |
780 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF8( |
781 | const Py_UNICODE *data, /* Unicode char buffer */ |
782 | Py_ssize_t length, /* number of Py_UNICODE chars to encode */ |
783 | const char *errors /* error handling */ |
784 | ); |
785 | |
786 | /* --- UTF-32 Codecs ------------------------------------------------------ */ |
787 | |
788 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF32( |
789 | const Py_UNICODE *data, /* Unicode char buffer */ |
790 | Py_ssize_t length, /* number of Py_UNICODE chars to encode */ |
791 | const char *errors, /* error handling */ |
792 | int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ |
793 | ); |
794 | |
795 | PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF32( |
796 | PyObject *object, /* Unicode object */ |
797 | const char *errors, /* error handling */ |
798 | int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ |
799 | ); |
800 | |
801 | /* --- UTF-16 Codecs ------------------------------------------------------ */ |
802 | |
803 | /* Returns a Python string object holding the UTF-16 encoded value of |
804 | the Unicode data. |
805 | |
806 | If byteorder is not 0, output is written according to the following |
807 | byte order: |
808 | |
809 | byteorder == -1: little endian |
810 | byteorder == 0: native byte order (writes a BOM mark) |
811 | byteorder == 1: big endian |
812 | |
813 | If byteorder is 0, the output string will always start with the |
814 | Unicode BOM mark (U+FEFF). In the other two modes, no BOM mark is |
815 | prepended. |
816 | |
817 | Note that Py_UNICODE data is being interpreted as UTF-16 reduced to |
818 | UCS-2. This trick makes it possible to add full UTF-16 capabilities |
819 | at a later point without compromising the APIs. |
820 | |
821 | */ |
822 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF16( |
823 | const Py_UNICODE *data, /* Unicode char buffer */ |
824 | Py_ssize_t length, /* number of Py_UNICODE chars to encode */ |
825 | const char *errors, /* error handling */ |
826 | int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ |
827 | ); |
828 | |
829 | PyAPI_FUNC(PyObject*) _PyUnicode_EncodeUTF16( |
830 | PyObject* unicode, /* Unicode object */ |
831 | const char *errors, /* error handling */ |
832 | int byteorder /* byteorder to use 0=BOM+native;-1=LE,1=BE */ |
833 | ); |
834 | |
835 | /* --- Unicode-Escape Codecs ---------------------------------------------- */ |
836 | |
837 | /* Variant of PyUnicode_DecodeUnicodeEscape that supports partial decoding. */ |
838 | PyAPI_FUNC(PyObject*) _PyUnicode_DecodeUnicodeEscapeStateful( |
839 | const char *string, /* Unicode-Escape encoded string */ |
840 | Py_ssize_t length, /* size of string */ |
841 | const char *errors, /* error handling */ |
842 | Py_ssize_t *consumed /* bytes consumed */ |
843 | ); |
844 | |
845 | /* Helper for PyUnicode_DecodeUnicodeEscape that detects invalid escape |
846 | chars. */ |
847 | PyAPI_FUNC(PyObject*) _PyUnicode_DecodeUnicodeEscapeInternal( |
848 | const char *string, /* Unicode-Escape encoded string */ |
849 | Py_ssize_t length, /* size of string */ |
850 | const char *errors, /* error handling */ |
851 | Py_ssize_t *consumed, /* bytes consumed */ |
852 | const char **first_invalid_escape /* on return, points to first |
853 | invalid escaped char in |
854 | string. */ |
855 | ); |
856 | |
857 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeUnicodeEscape( |
858 | const Py_UNICODE *data, /* Unicode char buffer */ |
859 | Py_ssize_t length /* Number of Py_UNICODE chars to encode */ |
860 | ); |
861 | |
862 | /* --- Raw-Unicode-Escape Codecs ------------------------------------------ */ |
863 | |
864 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeRawUnicodeEscape( |
865 | const Py_UNICODE *data, /* Unicode char buffer */ |
866 | Py_ssize_t length /* Number of Py_UNICODE chars to encode */ |
867 | ); |
868 | |
869 | /* Variant of PyUnicode_DecodeRawUnicodeEscape that supports partial decoding. */ |
870 | PyAPI_FUNC(PyObject*) _PyUnicode_DecodeRawUnicodeEscapeStateful( |
871 | const char *string, /* Unicode-Escape encoded string */ |
872 | Py_ssize_t length, /* size of string */ |
873 | const char *errors, /* error handling */ |
874 | Py_ssize_t *consumed /* bytes consumed */ |
875 | ); |
876 | |
877 | /* --- Latin-1 Codecs ----------------------------------------------------- */ |
878 | |
879 | PyAPI_FUNC(PyObject*) _PyUnicode_AsLatin1String( |
880 | PyObject* unicode, |
881 | const char* errors); |
882 | |
883 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeLatin1( |
884 | const Py_UNICODE *data, /* Unicode char buffer */ |
885 | Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ |
886 | const char *errors /* error handling */ |
887 | ); |
888 | |
889 | /* --- ASCII Codecs ------------------------------------------------------- */ |
890 | |
891 | PyAPI_FUNC(PyObject*) _PyUnicode_AsASCIIString( |
892 | PyObject* unicode, |
893 | const char* errors); |
894 | |
895 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeASCII( |
896 | const Py_UNICODE *data, /* Unicode char buffer */ |
897 | Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ |
898 | const char *errors /* error handling */ |
899 | ); |
900 | |
901 | /* --- Character Map Codecs ----------------------------------------------- */ |
902 | |
903 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeCharmap( |
904 | const Py_UNICODE *data, /* Unicode char buffer */ |
905 | Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ |
906 | PyObject *mapping, /* encoding mapping */ |
907 | const char *errors /* error handling */ |
908 | ); |
909 | |
910 | PyAPI_FUNC(PyObject*) _PyUnicode_EncodeCharmap( |
911 | PyObject *unicode, /* Unicode object */ |
912 | PyObject *mapping, /* encoding mapping */ |
913 | const char *errors /* error handling */ |
914 | ); |
915 | |
916 | /* Translate a Py_UNICODE buffer of the given length by applying a |
917 | character mapping table to it and return the resulting Unicode |
918 | object. |
919 | |
920 | The mapping table must map Unicode ordinal integers to Unicode strings, |
921 | Unicode ordinal integers or None (causing deletion of the character). |
922 | |
923 | Mapping tables may be dictionaries or sequences. Unmapped character |
924 | ordinals (ones which cause a LookupError) are left untouched and |
925 | are copied as-is. |
926 | |
927 | */ |
928 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject *) PyUnicode_TranslateCharmap( |
929 | const Py_UNICODE *data, /* Unicode char buffer */ |
930 | Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ |
931 | PyObject *table, /* Translate table */ |
932 | const char *errors /* error handling */ |
933 | ); |
934 | |
935 | /* --- MBCS codecs for Windows -------------------------------------------- */ |
936 | |
937 | #ifdef MS_WINDOWS |
938 | Py_DEPRECATED(3.3) PyAPI_FUNC(PyObject*) PyUnicode_EncodeMBCS( |
939 | const Py_UNICODE *data, /* Unicode char buffer */ |
940 | Py_ssize_t length, /* number of Py_UNICODE chars to encode */ |
941 | const char *errors /* error handling */ |
942 | ); |
943 | #endif |
944 | |
945 | /* --- Decimal Encoder ---------------------------------------------------- */ |
946 | |
947 | /* Takes a Unicode string holding a decimal value and writes it into |
948 | an output buffer using standard ASCII digit codes. |
949 | |
950 | The output buffer has to provide at least length+1 bytes of storage |
951 | area. The output string is 0-terminated. |
952 | |
953 | The encoder converts whitespace to ' ', decimal characters to their |
954 | corresponding ASCII digit and all other Latin-1 characters except |
955 | \0 as-is. Characters outside this range (Unicode ordinals 1-256) |
956 | are treated as errors. This includes embedded NULL bytes. |
957 | |
958 | Error handling is defined by the errors argument: |
959 | |
960 | NULL or "strict": raise a ValueError |
961 | "ignore": ignore the wrong characters (these are not copied to the |
962 | output buffer) |
963 | "replace": replaces illegal characters with '?' |
964 | |
965 | Returns 0 on success, -1 on failure. |
966 | |
967 | */ |
968 | |
969 | Py_DEPRECATED(3.3) PyAPI_FUNC(int) PyUnicode_EncodeDecimal( |
970 | Py_UNICODE *s, /* Unicode buffer */ |
971 | Py_ssize_t length, /* Number of Py_UNICODE chars to encode */ |
972 | char *output, /* Output buffer; must have size >= length */ |
973 | const char *errors /* error handling */ |
974 | ); |
975 | |
976 | /* Transforms code points that have decimal digit property to the |
977 | corresponding ASCII digit code points. |
978 | |
979 | Returns a new Unicode string on success, NULL on failure. |
980 | */ |
981 | |
982 | Py_DEPRECATED(3.3) |
983 | PyAPI_FUNC(PyObject*) PyUnicode_TransformDecimalToASCII( |
984 | Py_UNICODE *s, /* Unicode buffer */ |
985 | Py_ssize_t length /* Number of Py_UNICODE chars to transform */ |
986 | ); |
987 | |
988 | /* Coverts a Unicode object holding a decimal value to an ASCII string |
989 | for using in int, float and complex parsers. |
990 | Transforms code points that have decimal digit property to the |
991 | corresponding ASCII digit code points. Transforms spaces to ASCII. |
992 | Transforms code points starting from the first non-ASCII code point that |
993 | is neither a decimal digit nor a space to the end into '?'. */ |
994 | |
995 | PyAPI_FUNC(PyObject*) _PyUnicode_TransformDecimalAndSpaceToASCII( |
996 | PyObject *unicode /* Unicode object */ |
997 | ); |
998 | |
999 | /* --- Methods & Slots ---------------------------------------------------- */ |
1000 | |
1001 | PyAPI_FUNC(PyObject *) _PyUnicode_JoinArray( |
1002 | PyObject *separator, |
1003 | PyObject *const *items, |
1004 | Py_ssize_t seqlen |
1005 | ); |
1006 | |
1007 | /* Test whether a unicode is equal to ASCII identifier. Return 1 if true, |
1008 | 0 otherwise. The right argument must be ASCII identifier. |
1009 | Any error occurs inside will be cleared before return. */ |
1010 | PyAPI_FUNC(int) _PyUnicode_EqualToASCIIId( |
1011 | PyObject *left, /* Left string */ |
1012 | _Py_Identifier *right /* Right identifier */ |
1013 | ); |
1014 | |
1015 | /* Test whether a unicode is equal to ASCII string. Return 1 if true, |
1016 | 0 otherwise. The right argument must be ASCII-encoded string. |
1017 | Any error occurs inside will be cleared before return. */ |
1018 | PyAPI_FUNC(int) _PyUnicode_EqualToASCIIString( |
1019 | PyObject *left, |
1020 | const char *right /* ASCII-encoded string */ |
1021 | ); |
1022 | |
1023 | /* Externally visible for str.strip(unicode) */ |
1024 | PyAPI_FUNC(PyObject *) _PyUnicode_XStrip( |
1025 | PyObject *self, |
1026 | int striptype, |
1027 | PyObject *sepobj |
1028 | ); |
1029 | |
1030 | /* Using explicit passed-in values, insert the thousands grouping |
1031 | into the string pointed to by buffer. For the argument descriptions, |
1032 | see Objects/stringlib/localeutil.h */ |
1033 | PyAPI_FUNC(Py_ssize_t) _PyUnicode_InsertThousandsGrouping( |
1034 | _PyUnicodeWriter *writer, |
1035 | Py_ssize_t n_buffer, |
1036 | PyObject *digits, |
1037 | Py_ssize_t d_pos, |
1038 | Py_ssize_t n_digits, |
1039 | Py_ssize_t min_width, |
1040 | const char *grouping, |
1041 | PyObject *thousands_sep, |
1042 | Py_UCS4 *maxchar); |
1043 | |
1044 | /* === Characters Type APIs =============================================== */ |
1045 | |
1046 | /* Helper array used by Py_UNICODE_ISSPACE(). */ |
1047 | |
1048 | PyAPI_DATA(const unsigned char) _Py_ascii_whitespace[]; |
1049 | |
1050 | /* These should not be used directly. Use the Py_UNICODE_IS* and |
1051 | Py_UNICODE_TO* macros instead. |
1052 | |
1053 | These APIs are implemented in Objects/unicodectype.c. |
1054 | |
1055 | */ |
1056 | |
1057 | PyAPI_FUNC(int) _PyUnicode_IsLowercase( |
1058 | Py_UCS4 ch /* Unicode character */ |
1059 | ); |
1060 | |
1061 | PyAPI_FUNC(int) _PyUnicode_IsUppercase( |
1062 | Py_UCS4 ch /* Unicode character */ |
1063 | ); |
1064 | |
1065 | PyAPI_FUNC(int) _PyUnicode_IsTitlecase( |
1066 | Py_UCS4 ch /* Unicode character */ |
1067 | ); |
1068 | |
1069 | PyAPI_FUNC(int) _PyUnicode_IsXidStart( |
1070 | Py_UCS4 ch /* Unicode character */ |
1071 | ); |
1072 | |
1073 | PyAPI_FUNC(int) _PyUnicode_IsXidContinue( |
1074 | Py_UCS4 ch /* Unicode character */ |
1075 | ); |
1076 | |
1077 | PyAPI_FUNC(int) _PyUnicode_IsWhitespace( |
1078 | const Py_UCS4 ch /* Unicode character */ |
1079 | ); |
1080 | |
1081 | PyAPI_FUNC(int) _PyUnicode_IsLinebreak( |
1082 | const Py_UCS4 ch /* Unicode character */ |
1083 | ); |
1084 | |
1085 | /* Py_DEPRECATED(3.3) */ PyAPI_FUNC(Py_UCS4) _PyUnicode_ToLowercase( |
1086 | Py_UCS4 ch /* Unicode character */ |
1087 | ); |
1088 | |
1089 | /* Py_DEPRECATED(3.3) */ PyAPI_FUNC(Py_UCS4) _PyUnicode_ToUppercase( |
1090 | Py_UCS4 ch /* Unicode character */ |
1091 | ); |
1092 | |
1093 | Py_DEPRECATED(3.3) PyAPI_FUNC(Py_UCS4) _PyUnicode_ToTitlecase( |
1094 | Py_UCS4 ch /* Unicode character */ |
1095 | ); |
1096 | |
1097 | PyAPI_FUNC(int) _PyUnicode_ToLowerFull( |
1098 | Py_UCS4 ch, /* Unicode character */ |
1099 | Py_UCS4 *res |
1100 | ); |
1101 | |
1102 | PyAPI_FUNC(int) _PyUnicode_ToTitleFull( |
1103 | Py_UCS4 ch, /* Unicode character */ |
1104 | Py_UCS4 *res |
1105 | ); |
1106 | |
1107 | PyAPI_FUNC(int) _PyUnicode_ToUpperFull( |
1108 | Py_UCS4 ch, /* Unicode character */ |
1109 | Py_UCS4 *res |
1110 | ); |
1111 | |
1112 | PyAPI_FUNC(int) _PyUnicode_ToFoldedFull( |
1113 | Py_UCS4 ch, /* Unicode character */ |
1114 | Py_UCS4 *res |
1115 | ); |
1116 | |
1117 | PyAPI_FUNC(int) _PyUnicode_IsCaseIgnorable( |
1118 | Py_UCS4 ch /* Unicode character */ |
1119 | ); |
1120 | |
1121 | PyAPI_FUNC(int) _PyUnicode_IsCased( |
1122 | Py_UCS4 ch /* Unicode character */ |
1123 | ); |
1124 | |
1125 | PyAPI_FUNC(int) _PyUnicode_ToDecimalDigit( |
1126 | Py_UCS4 ch /* Unicode character */ |
1127 | ); |
1128 | |
1129 | PyAPI_FUNC(int) _PyUnicode_ToDigit( |
1130 | Py_UCS4 ch /* Unicode character */ |
1131 | ); |
1132 | |
1133 | PyAPI_FUNC(double) _PyUnicode_ToNumeric( |
1134 | Py_UCS4 ch /* Unicode character */ |
1135 | ); |
1136 | |
1137 | PyAPI_FUNC(int) _PyUnicode_IsDecimalDigit( |
1138 | Py_UCS4 ch /* Unicode character */ |
1139 | ); |
1140 | |
1141 | PyAPI_FUNC(int) _PyUnicode_IsDigit( |
1142 | Py_UCS4 ch /* Unicode character */ |
1143 | ); |
1144 | |
1145 | PyAPI_FUNC(int) _PyUnicode_IsNumeric( |
1146 | Py_UCS4 ch /* Unicode character */ |
1147 | ); |
1148 | |
1149 | PyAPI_FUNC(int) _PyUnicode_IsPrintable( |
1150 | Py_UCS4 ch /* Unicode character */ |
1151 | ); |
1152 | |
1153 | PyAPI_FUNC(int) _PyUnicode_IsAlpha( |
1154 | Py_UCS4 ch /* Unicode character */ |
1155 | ); |
1156 | |
1157 | PyAPI_FUNC(PyObject*) _PyUnicode_FormatLong(PyObject *, int, int, int); |
1158 | |
1159 | /* Return an interned Unicode object for an Identifier; may fail if there is no memory.*/ |
1160 | PyAPI_FUNC(PyObject*) _PyUnicode_FromId(_Py_Identifier*); |
1161 | |
1162 | /* Fast equality check when the inputs are known to be exact unicode types |
1163 | and where the hash values are equal (i.e. a very probable match) */ |
1164 | PyAPI_FUNC(int) _PyUnicode_EQ(PyObject *, PyObject *); |
1165 | |
1166 | PyAPI_FUNC(int) _PyUnicode_WideCharString_Converter(PyObject *, void *); |
1167 | PyAPI_FUNC(int) _PyUnicode_WideCharString_Opt_Converter(PyObject *, void *); |
1168 | |
1169 | PyAPI_FUNC(Py_ssize_t) _PyUnicode_ScanIdentifier(PyObject *); |
1170 | |