|
1 /* ------------------------------------------------------------------------ |
|
2 |
|
3 _codecs -- Provides access to the codec registry and the builtin |
|
4 codecs. |
|
5 |
|
6 This module should never be imported directly. The standard library |
|
7 module "codecs" wraps this builtin module for use within Python. |
|
8 |
|
9 The codec registry is accessible via: |
|
10 |
|
11 register(search_function) -> None |
|
12 |
|
13 lookup(encoding) -> CodecInfo object |
|
14 |
|
15 The builtin Unicode codecs use the following interface: |
|
16 |
|
17 <encoding>_encode(Unicode_object[,errors='strict']) -> |
|
18 (string object, bytes consumed) |
|
19 |
|
20 <encoding>_decode(char_buffer_obj[,errors='strict']) -> |
|
21 (Unicode object, bytes consumed) |
|
22 |
|
23 <encoding>_encode() interfaces also accept non-Unicode object as |
|
24 input. The objects are then converted to Unicode using |
|
25 PyUnicode_FromObject() prior to applying the conversion. |
|
26 |
|
27 These <encoding>s are available: utf_8, unicode_escape, |
|
28 raw_unicode_escape, unicode_internal, latin_1, ascii (7-bit), |
|
29 mbcs (on win32). |
|
30 |
|
31 |
|
32 Written by Marc-Andre Lemburg (mal@lemburg.com). |
|
33 |
|
34 Copyright (c) Corporation for National Research Initiatives. |
|
35 |
|
36 ------------------------------------------------------------------------ */ |
|
37 |
|
38 #define PY_SSIZE_T_CLEAN |
|
39 #include "Python.h" |
|
40 |
|
41 /* --- Registry ----------------------------------------------------------- */ |
|
42 |
|
43 PyDoc_STRVAR(register__doc__, |
|
44 "register(search_function)\n\ |
|
45 \n\ |
|
46 Register a codec search function. Search functions are expected to take\n\ |
|
47 one argument, the encoding name in all lower case letters, and return\n\ |
|
48 a tuple of functions (encoder, decoder, stream_reader, stream_writer)\n\ |
|
49 (or a CodecInfo object)."); |
|
50 |
|
51 static |
|
52 PyObject *codec_register(PyObject *self, PyObject *search_function) |
|
53 { |
|
54 if (PyCodec_Register(search_function)) |
|
55 return NULL; |
|
56 |
|
57 Py_RETURN_NONE; |
|
58 } |
|
59 |
|
60 PyDoc_STRVAR(lookup__doc__, |
|
61 "lookup(encoding) -> CodecInfo\n\ |
|
62 \n\ |
|
63 Looks up a codec tuple in the Python codec registry and returns\n\ |
|
64 a tuple of function (or a CodecInfo object)."); |
|
65 |
|
66 static |
|
67 PyObject *codec_lookup(PyObject *self, PyObject *args) |
|
68 { |
|
69 char *encoding; |
|
70 |
|
71 if (!PyArg_ParseTuple(args, "s:lookup", &encoding)) |
|
72 return NULL; |
|
73 |
|
74 return _PyCodec_Lookup(encoding); |
|
75 } |
|
76 |
|
77 PyDoc_STRVAR(encode__doc__, |
|
78 "encode(obj, [encoding[,errors]]) -> object\n\ |
|
79 \n\ |
|
80 Encodes obj using the codec registered for encoding. encoding defaults\n\ |
|
81 to the default encoding. errors may be given to set a different error\n\ |
|
82 handling scheme. Default is 'strict' meaning that encoding errors raise\n\ |
|
83 a ValueError. Other possible values are 'ignore', 'replace' and\n\ |
|
84 'xmlcharrefreplace' as well as any other name registered with\n\ |
|
85 codecs.register_error that can handle ValueErrors."); |
|
86 |
|
87 static PyObject * |
|
88 codec_encode(PyObject *self, PyObject *args) |
|
89 { |
|
90 const char *encoding = NULL; |
|
91 const char *errors = NULL; |
|
92 PyObject *v; |
|
93 |
|
94 if (!PyArg_ParseTuple(args, "O|ss:encode", &v, &encoding, &errors)) |
|
95 return NULL; |
|
96 |
|
97 #ifdef Py_USING_UNICODE |
|
98 if (encoding == NULL) |
|
99 encoding = PyUnicode_GetDefaultEncoding(); |
|
100 #else |
|
101 if (encoding == NULL) { |
|
102 PyErr_SetString(PyExc_ValueError, "no encoding specified"); |
|
103 return NULL; |
|
104 } |
|
105 #endif |
|
106 |
|
107 /* Encode via the codec registry */ |
|
108 return PyCodec_Encode(v, encoding, errors); |
|
109 } |
|
110 |
|
111 PyDoc_STRVAR(decode__doc__, |
|
112 "decode(obj, [encoding[,errors]]) -> object\n\ |
|
113 \n\ |
|
114 Decodes obj using the codec registered for encoding. encoding defaults\n\ |
|
115 to the default encoding. errors may be given to set a different error\n\ |
|
116 handling scheme. Default is 'strict' meaning that encoding errors raise\n\ |
|
117 a ValueError. Other possible values are 'ignore' and 'replace'\n\ |
|
118 as well as any other name registered with codecs.register_error that is\n\ |
|
119 able to handle ValueErrors."); |
|
120 |
|
121 static PyObject * |
|
122 codec_decode(PyObject *self, PyObject *args) |
|
123 { |
|
124 const char *encoding = NULL; |
|
125 const char *errors = NULL; |
|
126 PyObject *v; |
|
127 |
|
128 if (!PyArg_ParseTuple(args, "O|ss:decode", &v, &encoding, &errors)) |
|
129 return NULL; |
|
130 |
|
131 #ifdef Py_USING_UNICODE |
|
132 if (encoding == NULL) |
|
133 encoding = PyUnicode_GetDefaultEncoding(); |
|
134 #else |
|
135 if (encoding == NULL) { |
|
136 PyErr_SetString(PyExc_ValueError, "no encoding specified"); |
|
137 return NULL; |
|
138 } |
|
139 #endif |
|
140 |
|
141 /* Decode via the codec registry */ |
|
142 return PyCodec_Decode(v, encoding, errors); |
|
143 } |
|
144 |
|
145 /* --- Helpers ------------------------------------------------------------ */ |
|
146 |
|
147 static |
|
148 PyObject *codec_tuple(PyObject *unicode, |
|
149 Py_ssize_t len) |
|
150 { |
|
151 PyObject *v; |
|
152 if (unicode == NULL) |
|
153 return NULL; |
|
154 v = Py_BuildValue("On", unicode, len); |
|
155 Py_DECREF(unicode); |
|
156 return v; |
|
157 } |
|
158 |
|
159 /* --- String codecs ------------------------------------------------------ */ |
|
160 static PyObject * |
|
161 escape_decode(PyObject *self, |
|
162 PyObject *args) |
|
163 { |
|
164 const char *errors = NULL; |
|
165 const char *data; |
|
166 Py_ssize_t size; |
|
167 |
|
168 if (!PyArg_ParseTuple(args, "s#|z:escape_decode", |
|
169 &data, &size, &errors)) |
|
170 return NULL; |
|
171 return codec_tuple(PyString_DecodeEscape(data, size, errors, 0, NULL), |
|
172 size); |
|
173 } |
|
174 |
|
175 static PyObject * |
|
176 escape_encode(PyObject *self, |
|
177 PyObject *args) |
|
178 { |
|
179 PyObject *str; |
|
180 const char *errors = NULL; |
|
181 char *buf; |
|
182 Py_ssize_t len; |
|
183 |
|
184 if (!PyArg_ParseTuple(args, "O!|z:escape_encode", |
|
185 &PyString_Type, &str, &errors)) |
|
186 return NULL; |
|
187 |
|
188 str = PyString_Repr(str, 0); |
|
189 if (!str) |
|
190 return NULL; |
|
191 |
|
192 /* The string will be quoted. Unquote, similar to unicode-escape. */ |
|
193 buf = PyString_AS_STRING (str); |
|
194 len = PyString_GET_SIZE (str); |
|
195 memmove(buf, buf+1, len-2); |
|
196 if (_PyString_Resize(&str, len-2) < 0) |
|
197 return NULL; |
|
198 |
|
199 return codec_tuple(str, PyString_Size(str)); |
|
200 } |
|
201 |
|
202 #ifdef Py_USING_UNICODE |
|
203 /* --- Decoder ------------------------------------------------------------ */ |
|
204 |
|
205 static PyObject * |
|
206 unicode_internal_decode(PyObject *self, |
|
207 PyObject *args) |
|
208 { |
|
209 PyObject *obj; |
|
210 const char *errors = NULL; |
|
211 const char *data; |
|
212 Py_ssize_t size; |
|
213 |
|
214 if (!PyArg_ParseTuple(args, "O|z:unicode_internal_decode", |
|
215 &obj, &errors)) |
|
216 return NULL; |
|
217 |
|
218 if (PyUnicode_Check(obj)) { |
|
219 Py_INCREF(obj); |
|
220 return codec_tuple(obj, PyUnicode_GET_SIZE(obj)); |
|
221 } |
|
222 else { |
|
223 if (PyObject_AsReadBuffer(obj, (const void **)&data, &size)) |
|
224 return NULL; |
|
225 |
|
226 return codec_tuple(_PyUnicode_DecodeUnicodeInternal(data, size, errors), |
|
227 size); |
|
228 } |
|
229 } |
|
230 |
|
231 static PyObject * |
|
232 utf_7_decode(PyObject *self, |
|
233 PyObject *args) |
|
234 { |
|
235 Py_buffer pbuf; |
|
236 const char *errors = NULL; |
|
237 int final = 0; |
|
238 Py_ssize_t consumed; |
|
239 PyObject *decoded = NULL; |
|
240 |
|
241 if (!PyArg_ParseTuple(args, "s*|zi:utf_7_decode", |
|
242 &pbuf, &errors, &final)) |
|
243 return NULL; |
|
244 consumed = pbuf.len; |
|
245 |
|
246 decoded = PyUnicode_DecodeUTF7Stateful(pbuf.buf, pbuf.len, errors, |
|
247 final ? NULL : &consumed); |
|
248 PyBuffer_Release(&pbuf); |
|
249 if (decoded == NULL) |
|
250 return NULL; |
|
251 return codec_tuple(decoded, consumed); |
|
252 } |
|
253 |
|
254 static PyObject * |
|
255 utf_8_decode(PyObject *self, |
|
256 PyObject *args) |
|
257 { |
|
258 Py_buffer pbuf; |
|
259 const char *errors = NULL; |
|
260 int final = 0; |
|
261 Py_ssize_t consumed; |
|
262 PyObject *decoded = NULL; |
|
263 |
|
264 if (!PyArg_ParseTuple(args, "s*|zi:utf_8_decode", |
|
265 &pbuf, &errors, &final)) |
|
266 return NULL; |
|
267 consumed = pbuf.len; |
|
268 |
|
269 decoded = PyUnicode_DecodeUTF8Stateful(pbuf.buf, pbuf.len, errors, |
|
270 final ? NULL : &consumed); |
|
271 PyBuffer_Release(&pbuf); |
|
272 if (decoded == NULL) |
|
273 return NULL; |
|
274 return codec_tuple(decoded, consumed); |
|
275 } |
|
276 |
|
277 static PyObject * |
|
278 utf_16_decode(PyObject *self, |
|
279 PyObject *args) |
|
280 { |
|
281 Py_buffer pbuf; |
|
282 const char *errors = NULL; |
|
283 int byteorder = 0; |
|
284 int final = 0; |
|
285 Py_ssize_t consumed; |
|
286 PyObject *decoded; |
|
287 |
|
288 if (!PyArg_ParseTuple(args, "s*|zi:utf_16_decode", |
|
289 &pbuf, &errors, &final)) |
|
290 return NULL; |
|
291 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
292 decoded = PyUnicode_DecodeUTF16Stateful(pbuf.buf, pbuf.len, errors, |
|
293 &byteorder, final ? NULL : &consumed); |
|
294 PyBuffer_Release(&pbuf); |
|
295 if (decoded == NULL) |
|
296 return NULL; |
|
297 return codec_tuple(decoded, consumed); |
|
298 } |
|
299 |
|
300 static PyObject * |
|
301 utf_16_le_decode(PyObject *self, |
|
302 PyObject *args) |
|
303 { |
|
304 Py_buffer pbuf; |
|
305 const char *errors = NULL; |
|
306 int byteorder = -1; |
|
307 int final = 0; |
|
308 Py_ssize_t consumed; |
|
309 PyObject *decoded = NULL; |
|
310 |
|
311 if (!PyArg_ParseTuple(args, "s*|zi:utf_16_le_decode", |
|
312 &pbuf, &errors, &final)) |
|
313 return NULL; |
|
314 |
|
315 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
316 decoded = PyUnicode_DecodeUTF16Stateful(pbuf.buf, pbuf.len, errors, |
|
317 &byteorder, final ? NULL : &consumed); |
|
318 PyBuffer_Release(&pbuf); |
|
319 if (decoded == NULL) |
|
320 return NULL; |
|
321 return codec_tuple(decoded, consumed); |
|
322 } |
|
323 |
|
324 static PyObject * |
|
325 utf_16_be_decode(PyObject *self, |
|
326 PyObject *args) |
|
327 { |
|
328 Py_buffer pbuf; |
|
329 const char *errors = NULL; |
|
330 int byteorder = 1; |
|
331 int final = 0; |
|
332 Py_ssize_t consumed; |
|
333 PyObject *decoded = NULL; |
|
334 |
|
335 if (!PyArg_ParseTuple(args, "s*|zi:utf_16_be_decode", |
|
336 &pbuf, &errors, &final)) |
|
337 return NULL; |
|
338 |
|
339 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
340 decoded = PyUnicode_DecodeUTF16Stateful(pbuf.buf, pbuf.len, errors, |
|
341 &byteorder, final ? NULL : &consumed); |
|
342 PyBuffer_Release(&pbuf); |
|
343 if (decoded == NULL) |
|
344 return NULL; |
|
345 return codec_tuple(decoded, consumed); |
|
346 } |
|
347 |
|
348 /* This non-standard version also provides access to the byteorder |
|
349 parameter of the builtin UTF-16 codec. |
|
350 |
|
351 It returns a tuple (unicode, bytesread, byteorder) with byteorder |
|
352 being the value in effect at the end of data. |
|
353 |
|
354 */ |
|
355 |
|
356 static PyObject * |
|
357 utf_16_ex_decode(PyObject *self, |
|
358 PyObject *args) |
|
359 { |
|
360 Py_buffer pbuf; |
|
361 const char *errors = NULL; |
|
362 int byteorder = 0; |
|
363 PyObject *unicode, *tuple; |
|
364 int final = 0; |
|
365 Py_ssize_t consumed; |
|
366 |
|
367 if (!PyArg_ParseTuple(args, "s*|zii:utf_16_ex_decode", |
|
368 &pbuf, &errors, &byteorder, &final)) |
|
369 return NULL; |
|
370 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
371 unicode = PyUnicode_DecodeUTF16Stateful(pbuf.buf, pbuf.len, errors, |
|
372 &byteorder, final ? NULL : &consumed); |
|
373 PyBuffer_Release(&pbuf); |
|
374 if (unicode == NULL) |
|
375 return NULL; |
|
376 tuple = Py_BuildValue("Oni", unicode, consumed, byteorder); |
|
377 Py_DECREF(unicode); |
|
378 return tuple; |
|
379 } |
|
380 |
|
381 static PyObject * |
|
382 utf_32_decode(PyObject *self, |
|
383 PyObject *args) |
|
384 { |
|
385 Py_buffer pbuf; |
|
386 const char *errors = NULL; |
|
387 int byteorder = 0; |
|
388 int final = 0; |
|
389 Py_ssize_t consumed; |
|
390 PyObject *decoded; |
|
391 |
|
392 if (!PyArg_ParseTuple(args, "s*|zi:utf_32_decode", |
|
393 &pbuf, &errors, &final)) |
|
394 return NULL; |
|
395 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
396 decoded = PyUnicode_DecodeUTF32Stateful(pbuf.buf, pbuf.len, errors, |
|
397 &byteorder, final ? NULL : &consumed); |
|
398 PyBuffer_Release(&pbuf); |
|
399 if (decoded == NULL) |
|
400 return NULL; |
|
401 return codec_tuple(decoded, consumed); |
|
402 } |
|
403 |
|
404 static PyObject * |
|
405 utf_32_le_decode(PyObject *self, |
|
406 PyObject *args) |
|
407 { |
|
408 Py_buffer pbuf; |
|
409 const char *errors = NULL; |
|
410 int byteorder = -1; |
|
411 int final = 0; |
|
412 Py_ssize_t consumed; |
|
413 PyObject *decoded; |
|
414 |
|
415 if (!PyArg_ParseTuple(args, "s*|zi:utf_32_le_decode", |
|
416 &pbuf, &errors, &final)) |
|
417 return NULL; |
|
418 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
419 decoded = PyUnicode_DecodeUTF32Stateful(pbuf.buf, pbuf.len, errors, |
|
420 &byteorder, final ? NULL : &consumed); |
|
421 PyBuffer_Release(&pbuf); |
|
422 if (decoded == NULL) |
|
423 return NULL; |
|
424 return codec_tuple(decoded, consumed); |
|
425 } |
|
426 |
|
427 static PyObject * |
|
428 utf_32_be_decode(PyObject *self, |
|
429 PyObject *args) |
|
430 { |
|
431 Py_buffer pbuf; |
|
432 const char *errors = NULL; |
|
433 int byteorder = 1; |
|
434 int final = 0; |
|
435 Py_ssize_t consumed; |
|
436 PyObject *decoded; |
|
437 |
|
438 if (!PyArg_ParseTuple(args, "s*|zi:utf_32_be_decode", |
|
439 &pbuf, &errors, &final)) |
|
440 return NULL; |
|
441 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
442 decoded = PyUnicode_DecodeUTF32Stateful(pbuf.buf, pbuf.len, errors, |
|
443 &byteorder, final ? NULL : &consumed); |
|
444 PyBuffer_Release(&pbuf); |
|
445 if (decoded == NULL) |
|
446 return NULL; |
|
447 return codec_tuple(decoded, consumed); |
|
448 } |
|
449 |
|
450 /* This non-standard version also provides access to the byteorder |
|
451 parameter of the builtin UTF-32 codec. |
|
452 |
|
453 It returns a tuple (unicode, bytesread, byteorder) with byteorder |
|
454 being the value in effect at the end of data. |
|
455 |
|
456 */ |
|
457 |
|
458 static PyObject * |
|
459 utf_32_ex_decode(PyObject *self, |
|
460 PyObject *args) |
|
461 { |
|
462 Py_buffer pbuf; |
|
463 const char *errors = NULL; |
|
464 int byteorder = 0; |
|
465 PyObject *unicode, *tuple; |
|
466 int final = 0; |
|
467 Py_ssize_t consumed; |
|
468 |
|
469 if (!PyArg_ParseTuple(args, "s*|zii:utf_32_ex_decode", |
|
470 &pbuf, &errors, &byteorder, &final)) |
|
471 return NULL; |
|
472 consumed = pbuf.len; /* This is overwritten unless final is true. */ |
|
473 unicode = PyUnicode_DecodeUTF32Stateful(pbuf.buf, pbuf.len, errors, |
|
474 &byteorder, final ? NULL : &consumed); |
|
475 PyBuffer_Release(&pbuf); |
|
476 if (unicode == NULL) |
|
477 return NULL; |
|
478 tuple = Py_BuildValue("Oni", unicode, consumed, byteorder); |
|
479 Py_DECREF(unicode); |
|
480 return tuple; |
|
481 } |
|
482 |
|
483 static PyObject * |
|
484 unicode_escape_decode(PyObject *self, |
|
485 PyObject *args) |
|
486 { |
|
487 Py_buffer pbuf; |
|
488 const char *errors = NULL; |
|
489 PyObject *unicode; |
|
490 |
|
491 if (!PyArg_ParseTuple(args, "s*|z:unicode_escape_decode", |
|
492 &pbuf, &errors)) |
|
493 return NULL; |
|
494 |
|
495 unicode = PyUnicode_DecodeUnicodeEscape(pbuf.buf, pbuf.len, errors); |
|
496 PyBuffer_Release(&pbuf); |
|
497 return codec_tuple(unicode, pbuf.len); |
|
498 } |
|
499 |
|
500 static PyObject * |
|
501 raw_unicode_escape_decode(PyObject *self, |
|
502 PyObject *args) |
|
503 { |
|
504 Py_buffer pbuf; |
|
505 const char *errors = NULL; |
|
506 PyObject *unicode; |
|
507 |
|
508 if (!PyArg_ParseTuple(args, "s*|z:raw_unicode_escape_decode", |
|
509 &pbuf, &errors)) |
|
510 return NULL; |
|
511 |
|
512 unicode = PyUnicode_DecodeRawUnicodeEscape(pbuf.buf, pbuf.len, errors); |
|
513 PyBuffer_Release(&pbuf); |
|
514 return codec_tuple(unicode, pbuf.len); |
|
515 } |
|
516 |
|
517 static PyObject * |
|
518 latin_1_decode(PyObject *self, |
|
519 PyObject *args) |
|
520 { |
|
521 Py_buffer pbuf; |
|
522 PyObject *unicode; |
|
523 const char *errors = NULL; |
|
524 |
|
525 if (!PyArg_ParseTuple(args, "s*|z:latin_1_decode", |
|
526 &pbuf, &errors)) |
|
527 return NULL; |
|
528 |
|
529 unicode = PyUnicode_DecodeLatin1(pbuf.buf, pbuf.len, errors); |
|
530 PyBuffer_Release(&pbuf); |
|
531 return codec_tuple(unicode, pbuf.len); |
|
532 } |
|
533 |
|
534 static PyObject * |
|
535 ascii_decode(PyObject *self, |
|
536 PyObject *args) |
|
537 { |
|
538 Py_buffer pbuf; |
|
539 PyObject *unicode; |
|
540 const char *errors = NULL; |
|
541 |
|
542 if (!PyArg_ParseTuple(args, "s*|z:ascii_decode", |
|
543 &pbuf, &errors)) |
|
544 return NULL; |
|
545 |
|
546 unicode = PyUnicode_DecodeASCII(pbuf.buf, pbuf.len, errors); |
|
547 PyBuffer_Release(&pbuf); |
|
548 return codec_tuple(unicode, pbuf.len); |
|
549 } |
|
550 |
|
551 static PyObject * |
|
552 charmap_decode(PyObject *self, |
|
553 PyObject *args) |
|
554 { |
|
555 Py_buffer pbuf; |
|
556 PyObject *unicode; |
|
557 const char *errors = NULL; |
|
558 PyObject *mapping = NULL; |
|
559 |
|
560 if (!PyArg_ParseTuple(args, "s*|zO:charmap_decode", |
|
561 &pbuf, &errors, &mapping)) |
|
562 return NULL; |
|
563 if (mapping == Py_None) |
|
564 mapping = NULL; |
|
565 |
|
566 unicode = PyUnicode_DecodeCharmap(pbuf.buf, pbuf.len, mapping, errors); |
|
567 PyBuffer_Release(&pbuf); |
|
568 return codec_tuple(unicode, pbuf.len); |
|
569 } |
|
570 |
|
571 #if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T) |
|
572 |
|
573 static PyObject * |
|
574 mbcs_decode(PyObject *self, |
|
575 PyObject *args) |
|
576 { |
|
577 Py_buffer pbuf; |
|
578 const char *errors = NULL; |
|
579 int final = 0; |
|
580 Py_ssize_t consumed; |
|
581 PyObject *decoded = NULL; |
|
582 |
|
583 if (!PyArg_ParseTuple(args, "s*|zi:mbcs_decode", |
|
584 &pbuf, &errors, &final)) |
|
585 return NULL; |
|
586 consumed = pbuf.len; |
|
587 |
|
588 decoded = PyUnicode_DecodeMBCSStateful(pbuf.buf, pbuf.len, errors, |
|
589 final ? NULL : &consumed); |
|
590 PyBuffer_Release(&pbuf); |
|
591 if (decoded == NULL) |
|
592 return NULL; |
|
593 return codec_tuple(decoded, consumed); |
|
594 } |
|
595 |
|
596 #endif /* MS_WINDOWS */ |
|
597 |
|
598 /* --- Encoder ------------------------------------------------------------ */ |
|
599 |
|
600 static PyObject * |
|
601 readbuffer_encode(PyObject *self, |
|
602 PyObject *args) |
|
603 { |
|
604 const char *data; |
|
605 Py_ssize_t size; |
|
606 const char *errors = NULL; |
|
607 |
|
608 if (!PyArg_ParseTuple(args, "s#|z:readbuffer_encode", |
|
609 &data, &size, &errors)) |
|
610 return NULL; |
|
611 |
|
612 return codec_tuple(PyString_FromStringAndSize(data, size), |
|
613 size); |
|
614 } |
|
615 |
|
616 static PyObject * |
|
617 charbuffer_encode(PyObject *self, |
|
618 PyObject *args) |
|
619 { |
|
620 const char *data; |
|
621 Py_ssize_t size; |
|
622 const char *errors = NULL; |
|
623 |
|
624 if (!PyArg_ParseTuple(args, "t#|z:charbuffer_encode", |
|
625 &data, &size, &errors)) |
|
626 return NULL; |
|
627 |
|
628 return codec_tuple(PyString_FromStringAndSize(data, size), |
|
629 size); |
|
630 } |
|
631 |
|
632 static PyObject * |
|
633 unicode_internal_encode(PyObject *self, |
|
634 PyObject *args) |
|
635 { |
|
636 PyObject *obj; |
|
637 const char *errors = NULL; |
|
638 const char *data; |
|
639 Py_ssize_t size; |
|
640 |
|
641 if (!PyArg_ParseTuple(args, "O|z:unicode_internal_encode", |
|
642 &obj, &errors)) |
|
643 return NULL; |
|
644 |
|
645 if (PyUnicode_Check(obj)) { |
|
646 data = PyUnicode_AS_DATA(obj); |
|
647 size = PyUnicode_GET_DATA_SIZE(obj); |
|
648 return codec_tuple(PyString_FromStringAndSize(data, size), |
|
649 size); |
|
650 } |
|
651 else { |
|
652 if (PyObject_AsReadBuffer(obj, (const void **)&data, &size)) |
|
653 return NULL; |
|
654 return codec_tuple(PyString_FromStringAndSize(data, size), |
|
655 size); |
|
656 } |
|
657 } |
|
658 |
|
659 static PyObject * |
|
660 utf_7_encode(PyObject *self, |
|
661 PyObject *args) |
|
662 { |
|
663 PyObject *str, *v; |
|
664 const char *errors = NULL; |
|
665 |
|
666 if (!PyArg_ParseTuple(args, "O|z:utf_7_encode", |
|
667 &str, &errors)) |
|
668 return NULL; |
|
669 |
|
670 str = PyUnicode_FromObject(str); |
|
671 if (str == NULL) |
|
672 return NULL; |
|
673 v = codec_tuple(PyUnicode_EncodeUTF7(PyUnicode_AS_UNICODE(str), |
|
674 PyUnicode_GET_SIZE(str), |
|
675 0, |
|
676 0, |
|
677 errors), |
|
678 PyUnicode_GET_SIZE(str)); |
|
679 Py_DECREF(str); |
|
680 return v; |
|
681 } |
|
682 |
|
683 static PyObject * |
|
684 utf_8_encode(PyObject *self, |
|
685 PyObject *args) |
|
686 { |
|
687 PyObject *str, *v; |
|
688 const char *errors = NULL; |
|
689 |
|
690 if (!PyArg_ParseTuple(args, "O|z:utf_8_encode", |
|
691 &str, &errors)) |
|
692 return NULL; |
|
693 |
|
694 str = PyUnicode_FromObject(str); |
|
695 if (str == NULL) |
|
696 return NULL; |
|
697 v = codec_tuple(PyUnicode_EncodeUTF8(PyUnicode_AS_UNICODE(str), |
|
698 PyUnicode_GET_SIZE(str), |
|
699 errors), |
|
700 PyUnicode_GET_SIZE(str)); |
|
701 Py_DECREF(str); |
|
702 return v; |
|
703 } |
|
704 |
|
705 /* This version provides access to the byteorder parameter of the |
|
706 builtin UTF-16 codecs as optional third argument. It defaults to 0 |
|
707 which means: use the native byte order and prepend the data with a |
|
708 BOM mark. |
|
709 |
|
710 */ |
|
711 |
|
712 static PyObject * |
|
713 utf_16_encode(PyObject *self, |
|
714 PyObject *args) |
|
715 { |
|
716 PyObject *str, *v; |
|
717 const char *errors = NULL; |
|
718 int byteorder = 0; |
|
719 |
|
720 if (!PyArg_ParseTuple(args, "O|zi:utf_16_encode", |
|
721 &str, &errors, &byteorder)) |
|
722 return NULL; |
|
723 |
|
724 str = PyUnicode_FromObject(str); |
|
725 if (str == NULL) |
|
726 return NULL; |
|
727 v = codec_tuple(PyUnicode_EncodeUTF16(PyUnicode_AS_UNICODE(str), |
|
728 PyUnicode_GET_SIZE(str), |
|
729 errors, |
|
730 byteorder), |
|
731 PyUnicode_GET_SIZE(str)); |
|
732 Py_DECREF(str); |
|
733 return v; |
|
734 } |
|
735 |
|
736 static PyObject * |
|
737 utf_16_le_encode(PyObject *self, |
|
738 PyObject *args) |
|
739 { |
|
740 PyObject *str, *v; |
|
741 const char *errors = NULL; |
|
742 |
|
743 if (!PyArg_ParseTuple(args, "O|z:utf_16_le_encode", |
|
744 &str, &errors)) |
|
745 return NULL; |
|
746 |
|
747 str = PyUnicode_FromObject(str); |
|
748 if (str == NULL) |
|
749 return NULL; |
|
750 v = codec_tuple(PyUnicode_EncodeUTF16(PyUnicode_AS_UNICODE(str), |
|
751 PyUnicode_GET_SIZE(str), |
|
752 errors, |
|
753 -1), |
|
754 PyUnicode_GET_SIZE(str)); |
|
755 Py_DECREF(str); |
|
756 return v; |
|
757 } |
|
758 |
|
759 static PyObject * |
|
760 utf_16_be_encode(PyObject *self, |
|
761 PyObject *args) |
|
762 { |
|
763 PyObject *str, *v; |
|
764 const char *errors = NULL; |
|
765 |
|
766 if (!PyArg_ParseTuple(args, "O|z:utf_16_be_encode", |
|
767 &str, &errors)) |
|
768 return NULL; |
|
769 |
|
770 str = PyUnicode_FromObject(str); |
|
771 if (str == NULL) |
|
772 return NULL; |
|
773 v = codec_tuple(PyUnicode_EncodeUTF16(PyUnicode_AS_UNICODE(str), |
|
774 PyUnicode_GET_SIZE(str), |
|
775 errors, |
|
776 +1), |
|
777 PyUnicode_GET_SIZE(str)); |
|
778 Py_DECREF(str); |
|
779 return v; |
|
780 } |
|
781 |
|
782 /* This version provides access to the byteorder parameter of the |
|
783 builtin UTF-32 codecs as optional third argument. It defaults to 0 |
|
784 which means: use the native byte order and prepend the data with a |
|
785 BOM mark. |
|
786 |
|
787 */ |
|
788 |
|
789 static PyObject * |
|
790 utf_32_encode(PyObject *self, |
|
791 PyObject *args) |
|
792 { |
|
793 PyObject *str, *v; |
|
794 const char *errors = NULL; |
|
795 int byteorder = 0; |
|
796 |
|
797 if (!PyArg_ParseTuple(args, "O|zi:utf_32_encode", |
|
798 &str, &errors, &byteorder)) |
|
799 return NULL; |
|
800 |
|
801 str = PyUnicode_FromObject(str); |
|
802 if (str == NULL) |
|
803 return NULL; |
|
804 v = codec_tuple(PyUnicode_EncodeUTF32(PyUnicode_AS_UNICODE(str), |
|
805 PyUnicode_GET_SIZE(str), |
|
806 errors, |
|
807 byteorder), |
|
808 PyUnicode_GET_SIZE(str)); |
|
809 Py_DECREF(str); |
|
810 return v; |
|
811 } |
|
812 |
|
813 static PyObject * |
|
814 utf_32_le_encode(PyObject *self, |
|
815 PyObject *args) |
|
816 { |
|
817 PyObject *str, *v; |
|
818 const char *errors = NULL; |
|
819 |
|
820 if (!PyArg_ParseTuple(args, "O|z:utf_32_le_encode", |
|
821 &str, &errors)) |
|
822 return NULL; |
|
823 |
|
824 str = PyUnicode_FromObject(str); |
|
825 if (str == NULL) |
|
826 return NULL; |
|
827 v = codec_tuple(PyUnicode_EncodeUTF32(PyUnicode_AS_UNICODE(str), |
|
828 PyUnicode_GET_SIZE(str), |
|
829 errors, |
|
830 -1), |
|
831 PyUnicode_GET_SIZE(str)); |
|
832 Py_DECREF(str); |
|
833 return v; |
|
834 } |
|
835 |
|
836 static PyObject * |
|
837 utf_32_be_encode(PyObject *self, |
|
838 PyObject *args) |
|
839 { |
|
840 PyObject *str, *v; |
|
841 const char *errors = NULL; |
|
842 |
|
843 if (!PyArg_ParseTuple(args, "O|z:utf_32_be_encode", |
|
844 &str, &errors)) |
|
845 return NULL; |
|
846 |
|
847 str = PyUnicode_FromObject(str); |
|
848 if (str == NULL) |
|
849 return NULL; |
|
850 v = codec_tuple(PyUnicode_EncodeUTF32(PyUnicode_AS_UNICODE(str), |
|
851 PyUnicode_GET_SIZE(str), |
|
852 errors, |
|
853 +1), |
|
854 PyUnicode_GET_SIZE(str)); |
|
855 Py_DECREF(str); |
|
856 return v; |
|
857 } |
|
858 |
|
859 static PyObject * |
|
860 unicode_escape_encode(PyObject *self, |
|
861 PyObject *args) |
|
862 { |
|
863 PyObject *str, *v; |
|
864 const char *errors = NULL; |
|
865 |
|
866 if (!PyArg_ParseTuple(args, "O|z:unicode_escape_encode", |
|
867 &str, &errors)) |
|
868 return NULL; |
|
869 |
|
870 str = PyUnicode_FromObject(str); |
|
871 if (str == NULL) |
|
872 return NULL; |
|
873 v = codec_tuple(PyUnicode_EncodeUnicodeEscape(PyUnicode_AS_UNICODE(str), |
|
874 PyUnicode_GET_SIZE(str)), |
|
875 PyUnicode_GET_SIZE(str)); |
|
876 Py_DECREF(str); |
|
877 return v; |
|
878 } |
|
879 |
|
880 static PyObject * |
|
881 raw_unicode_escape_encode(PyObject *self, |
|
882 PyObject *args) |
|
883 { |
|
884 PyObject *str, *v; |
|
885 const char *errors = NULL; |
|
886 |
|
887 if (!PyArg_ParseTuple(args, "O|z:raw_unicode_escape_encode", |
|
888 &str, &errors)) |
|
889 return NULL; |
|
890 |
|
891 str = PyUnicode_FromObject(str); |
|
892 if (str == NULL) |
|
893 return NULL; |
|
894 v = codec_tuple(PyUnicode_EncodeRawUnicodeEscape( |
|
895 PyUnicode_AS_UNICODE(str), |
|
896 PyUnicode_GET_SIZE(str)), |
|
897 PyUnicode_GET_SIZE(str)); |
|
898 Py_DECREF(str); |
|
899 return v; |
|
900 } |
|
901 |
|
902 static PyObject * |
|
903 latin_1_encode(PyObject *self, |
|
904 PyObject *args) |
|
905 { |
|
906 PyObject *str, *v; |
|
907 const char *errors = NULL; |
|
908 |
|
909 if (!PyArg_ParseTuple(args, "O|z:latin_1_encode", |
|
910 &str, &errors)) |
|
911 return NULL; |
|
912 |
|
913 str = PyUnicode_FromObject(str); |
|
914 if (str == NULL) |
|
915 return NULL; |
|
916 v = codec_tuple(PyUnicode_EncodeLatin1( |
|
917 PyUnicode_AS_UNICODE(str), |
|
918 PyUnicode_GET_SIZE(str), |
|
919 errors), |
|
920 PyUnicode_GET_SIZE(str)); |
|
921 Py_DECREF(str); |
|
922 return v; |
|
923 } |
|
924 |
|
925 static PyObject * |
|
926 ascii_encode(PyObject *self, |
|
927 PyObject *args) |
|
928 { |
|
929 PyObject *str, *v; |
|
930 const char *errors = NULL; |
|
931 |
|
932 if (!PyArg_ParseTuple(args, "O|z:ascii_encode", |
|
933 &str, &errors)) |
|
934 return NULL; |
|
935 |
|
936 str = PyUnicode_FromObject(str); |
|
937 if (str == NULL) |
|
938 return NULL; |
|
939 v = codec_tuple(PyUnicode_EncodeASCII( |
|
940 PyUnicode_AS_UNICODE(str), |
|
941 PyUnicode_GET_SIZE(str), |
|
942 errors), |
|
943 PyUnicode_GET_SIZE(str)); |
|
944 Py_DECREF(str); |
|
945 return v; |
|
946 } |
|
947 |
|
948 static PyObject * |
|
949 charmap_encode(PyObject *self, |
|
950 PyObject *args) |
|
951 { |
|
952 PyObject *str, *v; |
|
953 const char *errors = NULL; |
|
954 PyObject *mapping = NULL; |
|
955 |
|
956 if (!PyArg_ParseTuple(args, "O|zO:charmap_encode", |
|
957 &str, &errors, &mapping)) |
|
958 return NULL; |
|
959 if (mapping == Py_None) |
|
960 mapping = NULL; |
|
961 |
|
962 str = PyUnicode_FromObject(str); |
|
963 if (str == NULL) |
|
964 return NULL; |
|
965 v = codec_tuple(PyUnicode_EncodeCharmap( |
|
966 PyUnicode_AS_UNICODE(str), |
|
967 PyUnicode_GET_SIZE(str), |
|
968 mapping, |
|
969 errors), |
|
970 PyUnicode_GET_SIZE(str)); |
|
971 Py_DECREF(str); |
|
972 return v; |
|
973 } |
|
974 |
|
975 static PyObject* |
|
976 charmap_build(PyObject *self, PyObject *args) |
|
977 { |
|
978 PyObject *map; |
|
979 if (!PyArg_ParseTuple(args, "U:charmap_build", &map)) |
|
980 return NULL; |
|
981 return PyUnicode_BuildEncodingMap(map); |
|
982 } |
|
983 |
|
984 #if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T) |
|
985 |
|
986 static PyObject * |
|
987 mbcs_encode(PyObject *self, |
|
988 PyObject *args) |
|
989 { |
|
990 PyObject *str, *v; |
|
991 const char *errors = NULL; |
|
992 |
|
993 if (!PyArg_ParseTuple(args, "O|z:mbcs_encode", |
|
994 &str, &errors)) |
|
995 return NULL; |
|
996 |
|
997 str = PyUnicode_FromObject(str); |
|
998 if (str == NULL) |
|
999 return NULL; |
|
1000 v = codec_tuple(PyUnicode_EncodeMBCS( |
|
1001 PyUnicode_AS_UNICODE(str), |
|
1002 PyUnicode_GET_SIZE(str), |
|
1003 errors), |
|
1004 PyUnicode_GET_SIZE(str)); |
|
1005 Py_DECREF(str); |
|
1006 return v; |
|
1007 } |
|
1008 |
|
1009 #endif /* MS_WINDOWS */ |
|
1010 #endif /* Py_USING_UNICODE */ |
|
1011 |
|
1012 /* --- Error handler registry --------------------------------------------- */ |
|
1013 |
|
1014 PyDoc_STRVAR(register_error__doc__, |
|
1015 "register_error(errors, handler)\n\ |
|
1016 \n\ |
|
1017 Register the specified error handler under the name\n\ |
|
1018 errors. handler must be a callable object, that\n\ |
|
1019 will be called with an exception instance containing\n\ |
|
1020 information about the location of the encoding/decoding\n\ |
|
1021 error and must return a (replacement, new position) tuple."); |
|
1022 |
|
1023 static PyObject *register_error(PyObject *self, PyObject *args) |
|
1024 { |
|
1025 const char *name; |
|
1026 PyObject *handler; |
|
1027 |
|
1028 if (!PyArg_ParseTuple(args, "sO:register_error", |
|
1029 &name, &handler)) |
|
1030 return NULL; |
|
1031 if (PyCodec_RegisterError(name, handler)) |
|
1032 return NULL; |
|
1033 Py_RETURN_NONE; |
|
1034 } |
|
1035 |
|
1036 PyDoc_STRVAR(lookup_error__doc__, |
|
1037 "lookup_error(errors) -> handler\n\ |
|
1038 \n\ |
|
1039 Return the error handler for the specified error handling name\n\ |
|
1040 or raise a LookupError, if no handler exists under this name."); |
|
1041 |
|
1042 static PyObject *lookup_error(PyObject *self, PyObject *args) |
|
1043 { |
|
1044 const char *name; |
|
1045 |
|
1046 if (!PyArg_ParseTuple(args, "s:lookup_error", |
|
1047 &name)) |
|
1048 return NULL; |
|
1049 return PyCodec_LookupError(name); |
|
1050 } |
|
1051 |
|
1052 /* --- Module API --------------------------------------------------------- */ |
|
1053 |
|
1054 static PyMethodDef _codecs_functions[] = { |
|
1055 {"register", codec_register, METH_O, |
|
1056 register__doc__}, |
|
1057 {"lookup", codec_lookup, METH_VARARGS, |
|
1058 lookup__doc__}, |
|
1059 {"encode", codec_encode, METH_VARARGS, |
|
1060 encode__doc__}, |
|
1061 {"decode", codec_decode, METH_VARARGS, |
|
1062 decode__doc__}, |
|
1063 {"escape_encode", escape_encode, METH_VARARGS}, |
|
1064 {"escape_decode", escape_decode, METH_VARARGS}, |
|
1065 #ifdef Py_USING_UNICODE |
|
1066 {"utf_8_encode", utf_8_encode, METH_VARARGS}, |
|
1067 {"utf_8_decode", utf_8_decode, METH_VARARGS}, |
|
1068 {"utf_7_encode", utf_7_encode, METH_VARARGS}, |
|
1069 {"utf_7_decode", utf_7_decode, METH_VARARGS}, |
|
1070 {"utf_16_encode", utf_16_encode, METH_VARARGS}, |
|
1071 {"utf_16_le_encode", utf_16_le_encode, METH_VARARGS}, |
|
1072 {"utf_16_be_encode", utf_16_be_encode, METH_VARARGS}, |
|
1073 {"utf_16_decode", utf_16_decode, METH_VARARGS}, |
|
1074 {"utf_16_le_decode", utf_16_le_decode, METH_VARARGS}, |
|
1075 {"utf_16_be_decode", utf_16_be_decode, METH_VARARGS}, |
|
1076 {"utf_16_ex_decode", utf_16_ex_decode, METH_VARARGS}, |
|
1077 {"utf_32_encode", utf_32_encode, METH_VARARGS}, |
|
1078 {"utf_32_le_encode", utf_32_le_encode, METH_VARARGS}, |
|
1079 {"utf_32_be_encode", utf_32_be_encode, METH_VARARGS}, |
|
1080 {"utf_32_decode", utf_32_decode, METH_VARARGS}, |
|
1081 {"utf_32_le_decode", utf_32_le_decode, METH_VARARGS}, |
|
1082 {"utf_32_be_decode", utf_32_be_decode, METH_VARARGS}, |
|
1083 {"utf_32_ex_decode", utf_32_ex_decode, METH_VARARGS}, |
|
1084 {"unicode_escape_encode", unicode_escape_encode, METH_VARARGS}, |
|
1085 {"unicode_escape_decode", unicode_escape_decode, METH_VARARGS}, |
|
1086 {"unicode_internal_encode", unicode_internal_encode, METH_VARARGS}, |
|
1087 {"unicode_internal_decode", unicode_internal_decode, METH_VARARGS}, |
|
1088 {"raw_unicode_escape_encode", raw_unicode_escape_encode, METH_VARARGS}, |
|
1089 {"raw_unicode_escape_decode", raw_unicode_escape_decode, METH_VARARGS}, |
|
1090 {"latin_1_encode", latin_1_encode, METH_VARARGS}, |
|
1091 {"latin_1_decode", latin_1_decode, METH_VARARGS}, |
|
1092 {"ascii_encode", ascii_encode, METH_VARARGS}, |
|
1093 {"ascii_decode", ascii_decode, METH_VARARGS}, |
|
1094 {"charmap_encode", charmap_encode, METH_VARARGS}, |
|
1095 {"charmap_decode", charmap_decode, METH_VARARGS}, |
|
1096 {"charmap_build", charmap_build, METH_VARARGS}, |
|
1097 {"readbuffer_encode", readbuffer_encode, METH_VARARGS}, |
|
1098 {"charbuffer_encode", charbuffer_encode, METH_VARARGS}, |
|
1099 #if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T) |
|
1100 {"mbcs_encode", mbcs_encode, METH_VARARGS}, |
|
1101 {"mbcs_decode", mbcs_decode, METH_VARARGS}, |
|
1102 #endif |
|
1103 #endif /* Py_USING_UNICODE */ |
|
1104 {"register_error", register_error, METH_VARARGS, |
|
1105 register_error__doc__}, |
|
1106 {"lookup_error", lookup_error, METH_VARARGS, |
|
1107 lookup_error__doc__}, |
|
1108 {NULL, NULL} /* sentinel */ |
|
1109 }; |
|
1110 |
|
1111 PyMODINIT_FUNC |
|
1112 init_codecs(void) |
|
1113 { |
|
1114 Py_InitModule("_codecs", _codecs_functions); |
|
1115 } |