/home/mdboom/Work/builds/cpython/Objects/bytesobject.c
Line | Count | Source (jump to first uncovered line) |
1 | /* bytes object implementation */ |
2 | |
3 | #define PY_SSIZE_T_CLEAN |
4 | |
5 | #include "Python.h" |
6 | #include "pycore_abstract.h" // _PyIndex_Check() |
7 | #include "pycore_bytesobject.h" // _PyBytes_Find(), _PyBytes_Repeat() |
8 | #include "pycore_bytes_methods.h" // _Py_bytes_startswith() |
9 | #include "pycore_call.h" // _PyObject_CallNoArgs() |
10 | #include "pycore_format.h" // F_LJUST |
11 | #include "pycore_global_objects.h" // _Py_GET_GLOBAL_OBJECT() |
12 | #include "pycore_initconfig.h" // _PyStatus_OK() |
13 | #include "pycore_long.h" // _PyLong_DigitValue |
14 | #include "pycore_object.h" // _PyObject_GC_TRACK |
15 | #include "pycore_pymem.h" // PYMEM_CLEANBYTE |
16 | #include "pycore_strhex.h" // _Py_strhex_with_sep() |
17 | |
18 | #include <stddef.h> |
19 | |
20 | /*[clinic input] |
21 | class bytes "PyBytesObject *" "&PyBytes_Type" |
22 | [clinic start generated code]*/ |
23 | /*[clinic end generated code: output=da39a3ee5e6b4b0d input=7a238f965d64892b]*/ |
24 | |
25 | #include "clinic/bytesobject.c.h" |
26 | |
27 | /* PyBytesObject_SIZE gives the basic size of a bytes object; any memory allocation |
28 | for a bytes object of length n should request PyBytesObject_SIZE + n bytes. |
29 | |
30 | Using PyBytesObject_SIZE instead of sizeof(PyBytesObject) saves |
31 | 3 or 7 bytes per bytes object allocation on a typical system. |
32 | */ |
33 | #define PyBytesObject_SIZE (offsetof(PyBytesObject, ob_sval) + 1) |
34 | |
35 | /* Forward declaration */ |
36 | Py_LOCAL_INLINE(Py_ssize_t) _PyBytesWriter_GetSize(_PyBytesWriter *writer, |
37 | char *str); |
38 | |
39 | |
40 | #define CHARACTERS _Py_SINGLETON(bytes_characters) |
41 | #define CHARACTER(ch) \ |
42 | ((PyBytesObject *)&(CHARACTERS[ch])); |
43 | #define EMPTY (&_Py_SINGLETON(bytes_empty)) |
44 | |
45 | |
46 | // Return a borrowed reference to the empty bytes string singleton. |
47 | static inline PyObject* bytes_get_empty(void) |
48 | { |
49 | return &EMPTY->ob_base.ob_base; |
50 | } |
51 | |
52 | |
53 | // Return a strong reference to the empty bytes string singleton. |
54 | static inline PyObject* bytes_new_empty(void) |
55 | { |
56 | Py_INCREF(EMPTY); |
57 | return (PyObject *)EMPTY; |
58 | } |
59 | |
60 | |
61 | /* |
62 | For PyBytes_FromString(), the parameter `str' points to a null-terminated |
63 | string containing exactly `size' bytes. |
64 | |
65 | For PyBytes_FromStringAndSize(), the parameter `str' is |
66 | either NULL or else points to a string containing at least `size' bytes. |
67 | For PyBytes_FromStringAndSize(), the string in the `str' parameter does |
68 | not have to be null-terminated. (Therefore it is safe to construct a |
69 | substring by calling `PyBytes_FromStringAndSize(origstring, substrlen)'.) |
70 | If `str' is NULL then PyBytes_FromStringAndSize() will allocate `size+1' |
71 | bytes (setting the last byte to the null terminating character) and you can |
72 | fill in the data yourself. If `str' is non-NULL then the resulting |
73 | PyBytes object must be treated as immutable and you must not fill in nor |
74 | alter the data yourself, since the strings may be shared. |
75 | |
76 | The PyObject member `op->ob_size', which denotes the number of "extra |
77 | items" in a variable-size object, will contain the number of bytes |
78 | allocated for string data, not counting the null terminating character. |
79 | It is therefore equal to the `size' parameter (for |
80 | PyBytes_FromStringAndSize()) or the length of the string in the `str' |
81 | parameter (for PyBytes_FromString()). |
82 | */ |
83 | static PyObject * |
84 | _PyBytes_FromSize(Py_ssize_t size, int use_calloc) |
85 | { |
86 | PyBytesObject *op; |
87 | assert(size >= 0); |
88 | |
89 | if (size == 0) { Branch (89:9): [True: 3, False: 18.9M]
|
90 | return bytes_new_empty(); |
91 | } |
92 | |
93 | if ((size_t)size > (size_t)PY_SSIZE_T_MAX - PyBytesObject_SIZE) { Branch (93:9): [True: 0, False: 18.9M]
|
94 | PyErr_SetString(PyExc_OverflowError, |
95 | "byte string is too large"); |
96 | return NULL; |
97 | } |
98 | |
99 | /* Inline PyObject_NewVar */ |
100 | if (use_calloc) Branch (100:9): [True: 127, False: 18.9M]
|
101 | op = (PyBytesObject *)PyObject_Calloc(1, PyBytesObject_SIZE + size); |
102 | else |
103 | op = (PyBytesObject *)PyObject_Malloc(PyBytesObject_SIZE + size); |
104 | if (op == NULL) { Branch (104:9): [True: 0, False: 18.9M]
|
105 | return PyErr_NoMemory(); |
106 | } |
107 | _PyObject_InitVar((PyVarObject*)op, &PyBytes_Type, size); |
108 | _Py_COMP_DIAG_PUSH |
109 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
110 | op->ob_shash = -1; |
111 | _Py_COMP_DIAG_POP |
112 | if (!use_calloc) { Branch (112:9): [True: 18.9M, False: 127]
|
113 | op->ob_sval[size] = '\0'; |
114 | } |
115 | return (PyObject *) op; |
116 | } |
117 | |
118 | PyObject * |
119 | PyBytes_FromStringAndSize(const char *str, Py_ssize_t size) |
120 | { |
121 | PyBytesObject *op; |
122 | if (size < 0) { Branch (122:9): [True: 0, False: 36.0M]
|
123 | PyErr_SetString(PyExc_SystemError, |
124 | "Negative size passed to PyBytes_FromStringAndSize"); |
125 | return NULL; |
126 | } |
127 | if (size == 1 && str != NULL15.5M ) { Branch (127:9): [True: 15.5M, False: 20.4M]
Branch (127:22): [True: 15.4M, False: 116k]
|
128 | op = CHARACTER(*str & 255); |
129 | Py_INCREF(op); |
130 | return (PyObject *)op; |
131 | } |
132 | if (size == 0) { Branch (132:9): [True: 1.67M, False: 18.9M]
|
133 | return bytes_new_empty(); |
134 | } |
135 | |
136 | op = (PyBytesObject *)_PyBytes_FromSize(size, 0); |
137 | if (op == NULL) Branch (137:9): [True: 0, False: 18.9M]
|
138 | return NULL; |
139 | if (str == NULL) Branch (139:9): [True: 5.98M, False: 12.9M]
|
140 | return (PyObject *) op; |
141 | |
142 | memcpy(op->ob_sval, str, size); |
143 | return (PyObject *) op; |
144 | } |
145 | |
146 | PyObject * |
147 | PyBytes_FromString(const char *str) |
148 | { |
149 | size_t size; |
150 | PyBytesObject *op; |
151 | |
152 | assert(str != NULL); |
153 | size = strlen(str); |
154 | if (size > PY_SSIZE_T_MAX - PyBytesObject_SIZE) { Branch (154:9): [True: 0, False: 48.3k]
|
155 | PyErr_SetString(PyExc_OverflowError, |
156 | "byte string is too long"); |
157 | return NULL; |
158 | } |
159 | |
160 | if (size == 0) { Branch (160:9): [True: 2, False: 48.3k]
|
161 | return bytes_new_empty(); |
162 | } |
163 | else if (size == 1) { Branch (163:14): [True: 21, False: 48.3k]
|
164 | op = CHARACTER(*str & 255); |
165 | Py_INCREF(op); |
166 | return (PyObject *)op; |
167 | } |
168 | |
169 | /* Inline PyObject_NewVar */ |
170 | op = (PyBytesObject *)PyObject_Malloc(PyBytesObject_SIZE + size); |
171 | if (op == NULL) { Branch (171:9): [True: 0, False: 48.3k]
|
172 | return PyErr_NoMemory(); |
173 | } |
174 | _PyObject_InitVar((PyVarObject*)op, &PyBytes_Type, size); |
175 | _Py_COMP_DIAG_PUSH |
176 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
177 | op->ob_shash = -1; |
178 | _Py_COMP_DIAG_POP |
179 | memcpy(op->ob_sval, str, size+1); |
180 | return (PyObject *) op; |
181 | } |
182 | |
183 | PyObject * |
184 | PyBytes_FromFormatV(const char *format, va_list vargs) |
185 | { |
186 | char *s; |
187 | const char *f; |
188 | const char *p; |
189 | Py_ssize_t prec; |
190 | int longflag; |
191 | int size_tflag; |
192 | /* Longest 64-bit formatted numbers: |
193 | - "18446744073709551615\0" (21 bytes) |
194 | - "-9223372036854775808\0" (21 bytes) |
195 | Decimal takes the most space (it isn't enough for octal.) |
196 | |
197 | Longest 64-bit pointer representation: |
198 | "0xffffffffffffffff\0" (19 bytes). */ |
199 | char buffer[21]; |
200 | _PyBytesWriter writer; |
201 | |
202 | _PyBytesWriter_Init(&writer); |
203 | |
204 | s = _PyBytesWriter_Alloc(&writer, strlen(format)); |
205 | if (s == NULL) Branch (205:9): [True: 0, False: 5.83k]
|
206 | return NULL; |
207 | writer.overallocate = 1; |
208 | |
209 | #define WRITE_BYTES(str) \ |
210 | do 25 { \ |
211 | s = _PyBytesWriter_WriteBytes(&writer, s, (str), strlen(str)); \ |
212 | if (s == NULL) \ |
213 | goto error0 ; \ |
214 | } while (0) |
215 | |
216 | for (f = format; *f; f++17.5k ) { Branch (216:22): [True: 17.5k, False: 5.82k]
|
217 | if (*f != '%') { Branch (217:13): [True: 5.86k, False: 11.6k]
|
218 | *s++ = *f; |
219 | continue; |
220 | } |
221 | |
222 | p = f++; |
223 | |
224 | /* ignore the width (ex: 10 in "%10s") */ |
225 | while (Py_ISDIGIT(*f)) |
226 | f++; |
227 | |
228 | /* parse the precision (ex: 10 in "%.10s") */ |
229 | prec = 0; |
230 | if (*f == '.') { Branch (230:13): [True: 1, False: 11.6k]
|
231 | f++; |
232 | for (; Py_ISDIGIT(*f); f++1 ) { |
233 | prec = (prec * 10) + (*f - '0'); |
234 | } |
235 | } |
236 | |
237 | while (*f && *f != '%'11.6k && !11.6k Py_ISALPHA11.6k (*f)) Branch (237:16): [True: 11.6k, False: 2]
Branch (237:22): [True: 11.6k, False: 4]
Branch (237:35): [True: 0, False: 11.6k]
|
238 | f++; |
239 | |
240 | /* handle the long flag ('l'), but only for %ld and %lu. |
241 | others can be added when necessary. */ |
242 | longflag = 0; |
243 | if (*f == 'l' && (6 f[1] == 'd'6 || f[1] == 'u'2 )) { Branch (243:13): [True: 6, False: 11.6k]
Branch (243:27): [True: 4, False: 2]
Branch (243:42): [True: 2, False: 0]
|
244 | longflag = 1; |
245 | ++f; |
246 | } |
247 | |
248 | /* handle the size_t flag ('z'). */ |
249 | size_tflag = 0; |
250 | if (*f == 'z' && (6 f[1] == 'd'6 || f[1] == 'u'2 )) { Branch (250:13): [True: 6, False: 11.6k]
Branch (250:27): [True: 4, False: 2]
Branch (250:42): [True: 2, False: 0]
|
251 | size_tflag = 1; |
252 | ++f; |
253 | } |
254 | |
255 | /* subtract bytes preallocated for the format string |
256 | (ex: 2 for "%s") */ |
257 | writer.min_size -= (f - p + 1); |
258 | |
259 | switch (*f) { |
260 | case 'c': Branch (260:9): [True: 6, False: 11.6k]
|
261 | { |
262 | int c = va_arg(vargs, int); |
263 | if (c < 0 || c > 2555 ) { Branch (263:17): [True: 1, False: 5]
Branch (263:26): [True: 1, False: 4]
|
264 | PyErr_SetString(PyExc_OverflowError, |
265 | "PyBytes_FromFormatV(): %c format " |
266 | "expects an integer in range [0; 255]"); |
267 | goto error; |
268 | } |
269 | writer.min_size++; |
270 | *s++ = (unsigned char)c; |
271 | break; |
272 | } |
273 | |
274 | case 'd': Branch (274:9): [True: 12, False: 11.6k]
|
275 | if (longflag) { Branch (275:17): [True: 4, False: 8]
|
276 | sprintf(buffer, "%ld", va_arg(vargs, long)); |
277 | } |
278 | else if (size_tflag) { Branch (278:22): [True: 4, False: 4]
|
279 | sprintf(buffer, "%zd", va_arg(vargs, Py_ssize_t)); |
280 | } |
281 | else { |
282 | sprintf(buffer, "%d", va_arg(vargs, int)); |
283 | } |
284 | assert(strlen(buffer) < sizeof(buffer)); |
285 | WRITE_BYTES(buffer); |
286 | break; |
287 | |
288 | case 'u': Branch (288:9): [True: 5, False: 11.6k]
|
289 | if (longflag) { Branch (289:17): [True: 2, False: 3]
|
290 | sprintf(buffer, "%lu", va_arg(vargs, unsigned long)); |
291 | } |
292 | else if (size_tflag) { Branch (292:22): [True: 2, False: 1]
|
293 | sprintf(buffer, "%zu", va_arg(vargs, size_t)); |
294 | } |
295 | else { |
296 | sprintf(buffer, "%u", va_arg(vargs, unsigned int)); |
297 | } |
298 | assert(strlen(buffer) < sizeof(buffer)); |
299 | WRITE_BYTES(buffer); |
300 | break; |
301 | |
302 | case 'i': Branch (302:9): [True: 3, False: 11.6k]
|
303 | sprintf(buffer, "%i", va_arg(vargs, int)); |
304 | assert(strlen(buffer) < sizeof(buffer)); |
305 | WRITE_BYTES(buffer); |
306 | break; |
307 | |
308 | case 'x': Branch (308:9): [True: 1, False: 11.6k]
|
309 | sprintf(buffer, "%x", va_arg(vargs, int)); |
310 | assert(strlen(buffer) < sizeof(buffer)); |
311 | WRITE_BYTES(buffer); |
312 | break; |
313 | |
314 | case 's': Branch (314:9): [True: 11.6k, False: 35]
|
315 | { |
316 | Py_ssize_t i; |
317 | |
318 | p = va_arg(vargs, const char*); |
319 | if (prec <= 0) { Branch (319:17): [True: 11.6k, False: 1]
|
320 | i = strlen(p); |
321 | } |
322 | else { |
323 | i = 0; |
324 | while (i < prec && p[i]3 ) { Branch (324:24): [True: 3, False: 1]
Branch (324:36): [True: 3, False: 0]
|
325 | i++; |
326 | } |
327 | } |
328 | s = _PyBytesWriter_WriteBytes(&writer, s, p, i); |
329 | if (s == NULL) Branch (329:17): [True: 0, False: 11.6k]
|
330 | goto error; |
331 | break; |
332 | } |
333 | |
334 | case 'p': Branch (334:9): [True: 2, False: 11.6k]
|
335 | sprintf(buffer, "%p", va_arg(vargs, void*)); |
336 | assert(strlen(buffer) < sizeof(buffer)); |
337 | /* %p is ill-defined: ensure leading 0x. */ |
338 | if (buffer[1] == 'X') Branch (338:17): [True: 0, False: 2]
|
339 | buffer[1] = 'x'; |
340 | else if (buffer[1] != 'x') { Branch (340:22): [True: 0, False: 2]
|
341 | memmove(buffer+2, buffer, strlen(buffer)+1); |
342 | buffer[0] = '0'; |
343 | buffer[1] = 'x'; |
344 | } |
345 | WRITE_BYTES(buffer); |
346 | break; |
347 | |
348 | case '%': Branch (348:9): [True: 4, False: 11.6k]
|
349 | writer.min_size++; |
350 | *s++ = '%'; |
351 | break; |
352 | |
353 | default: Branch (353:9): [True: 2, False: 11.6k]
|
354 | if (*f == 0) { Branch (354:17): [True: 2, False: 0]
|
355 | /* fix min_size if we reached the end of the format string */ |
356 | writer.min_size++; |
357 | } |
358 | |
359 | /* invalid format string: copy unformatted string and exit */ |
360 | WRITE_BYTES(p); |
361 | return _PyBytesWriter_Finish(&writer, s); |
362 | } |
363 | } |
364 | |
365 | #undef WRITE_BYTES |
366 | |
367 | return _PyBytesWriter_Finish(&writer, s); |
368 | |
369 | error: |
370 | _PyBytesWriter_Dealloc(&writer); |
371 | return NULL; |
372 | } |
373 | |
374 | PyObject * |
375 | PyBytes_FromFormat(const char *format, ...) |
376 | { |
377 | PyObject* ret; |
378 | va_list vargs; |
379 | |
380 | va_start(vargs, format); |
381 | ret = PyBytes_FromFormatV(format, vargs); |
382 | va_end(vargs); |
383 | return ret; |
384 | } |
385 | |
386 | /* Helpers for formatstring */ |
387 | |
388 | Py_LOCAL_INLINE(PyObject *) |
389 | getnextarg(PyObject *args, Py_ssize_t arglen, Py_ssize_t *p_argidx) |
390 | { |
391 | Py_ssize_t argidx = *p_argidx; |
392 | if (argidx < arglen) { Branch (392:9): [True: 959, False: 0]
|
393 | (*p_argidx)++; |
394 | if (arglen < 0) Branch (394:13): [True: 609, False: 350]
|
395 | return args; |
396 | else |
397 | return PyTuple_GetItem(args, argidx); |
398 | } |
399 | PyErr_SetString(PyExc_TypeError, |
400 | "not enough arguments for format string"); |
401 | return NULL; |
402 | } |
403 | |
404 | /* Returns a new reference to a PyBytes object, or NULL on failure. */ |
405 | |
406 | static char* |
407 | formatfloat(PyObject *v, int flags, int prec, int type, |
408 | PyObject **p_result, _PyBytesWriter *writer, char *str) |
409 | { |
410 | char *p; |
411 | PyObject *result; |
412 | double x; |
413 | size_t len; |
414 | int dtoa_flags = 0; |
415 | |
416 | x = PyFloat_AsDouble(v); |
417 | if (x == -1.0 && PyErr_Occurred()2 ) { Branch (417:9): [True: 2, False: 22]
Branch (417:22): [True: 2, False: 0]
|
418 | PyErr_Format(PyExc_TypeError, "float argument required, " |
419 | "not %.200s", Py_TYPE(v)->tp_name); |
420 | return NULL; |
421 | } |
422 | |
423 | if (prec < 0) Branch (423:9): [True: 6, False: 16]
|
424 | prec = 6; |
425 | |
426 | if (flags & F_ALT) { Branch (426:9): [True: 16, False: 6]
|
427 | dtoa_flags |= Py_DTSF_ALT; |
428 | } |
429 | if (flags & F_NO_NEG_0) { Branch (429:9): [True: 0, False: 22]
|
430 | dtoa_flags |= Py_DTSF_NO_NEG_0; |
431 | } |
432 | p = PyOS_double_to_string(x, type, prec, dtoa_flags, NULL); |
433 | |
434 | if (p == NULL) Branch (434:9): [True: 0, False: 22]
|
435 | return NULL; |
436 | |
437 | len = strlen(p); |
438 | if (writer != NULL) { Branch (438:9): [True: 6, False: 16]
|
439 | str = _PyBytesWriter_Prepare(writer, str, len); |
440 | if (str == NULL) Branch (440:13): [True: 0, False: 6]
|
441 | return NULL; |
442 | memcpy(str, p, len); |
443 | PyMem_Free(p); |
444 | str += len; |
445 | return str; |
446 | } |
447 | |
448 | result = PyBytes_FromStringAndSize(p, len); |
449 | PyMem_Free(p); |
450 | *p_result = result; |
451 | return result != NULL ? str : NULL; Branch (451:12): [True: 16, False: 0]
|
452 | } |
453 | |
454 | static PyObject * |
455 | formatlong(PyObject *v, int flags, int prec, int type) |
456 | { |
457 | PyObject *result, *iobj; |
458 | if (type == 'i') Branch (458:9): [True: 0, False: 190]
|
459 | type = 'd'; |
460 | if (PyLong_Check(v)) |
461 | return _PyUnicode_FormatLong(v, flags & F_ALT, prec, type); |
462 | if (PyNumber_Check(v)) { Branch (462:9): [True: 5, False: 3]
|
463 | /* make sure number is a type of integer for o, x, and X */ |
464 | if (type == 'o' || type == 'x' || type == 'X'4 ) Branch (464:13): [True: 0, False: 5]
Branch (464:28): [True: 1, False: 4]
Branch (464:43): [True: 0, False: 4]
|
465 | iobj = _PyNumber_Index(v); |
466 | else |
467 | iobj = PyNumber_Long(v); |
468 | if (iobj != NULL) { Branch (468:13): [True: 4, False: 1]
|
469 | assert(PyLong_Check(iobj)); |
470 | result = _PyUnicode_FormatLong(iobj, flags & F_ALT, prec, type); |
471 | Py_DECREF(iobj); |
472 | return result; |
473 | } |
474 | if (!PyErr_ExceptionMatches(PyExc_TypeError)) Branch (474:13): [True: 0, False: 1]
|
475 | return NULL; |
476 | } |
477 | PyErr_Format(PyExc_TypeError, |
478 | "%%%c format: %s is required, not %.200s", type, |
479 | (type == 'o' || type == 'x' || type == 'X'2 ) ? "an integer"2 Branch (479:10): [True: 0, False: 4]
Branch (479:25): [True: 2, False: 2]
Branch (479:40): [True: 0, False: 2]
|
480 | : "a real number"2 , |
481 | Py_TYPE(v)->tp_name); |
482 | return NULL; |
483 | } |
484 | |
485 | static int |
486 | byte_converter(PyObject *arg, char *p) |
487 | { |
488 | if (PyBytes_Check(arg) && PyBytes_GET_SIZE3 (arg) == 13 ) { Branch (488:31): [True: 2, False: 1]
|
489 | *p = PyBytes_AS_STRING(arg)[0]; |
490 | return 1; |
491 | } |
492 | else if (PyByteArray_Check(arg) && PyByteArray_GET_SIZE2 (arg) == 12 ) { Branch (492:40): [True: 2, False: 0]
|
493 | *p = PyByteArray_AS_STRING(arg)[0]; |
494 | return 1; |
495 | } |
496 | else { |
497 | int overflow; |
498 | long ival = PyLong_AsLongAndOverflow(arg, &overflow); |
499 | if (ival == -1 && PyErr_Occurred()5 ) { Branch (499:13): [True: 5, False: 9]
Branch (499:27): [True: 3, False: 2]
|
500 | if (PyErr_ExceptionMatches(PyExc_TypeError)) { Branch (500:17): [True: 3, False: 0]
|
501 | goto onError; |
502 | } |
503 | return 0; |
504 | } |
505 | if (!(0 <= ival && ival <= 2559 )) { Branch (505:15): [True: 9, False: 2]
Branch (505:28): [True: 8, False: 1]
|
506 | /* this includes an overflow in converting to C long */ |
507 | PyErr_SetString(PyExc_OverflowError, |
508 | "%c arg not in range(256)"); |
509 | return 0; |
510 | } |
511 | *p = (char)ival; |
512 | return 1; |
513 | } |
514 | onError: |
515 | PyErr_SetString(PyExc_TypeError, |
516 | "%c requires an integer in range(256) or a single byte"); |
517 | return 0; |
518 | } |
519 | |
520 | static PyObject *_PyBytes_FromBuffer(PyObject *x); |
521 | |
522 | static PyObject * |
523 | format_obj(PyObject *v, const char **pbuf, Py_ssize_t *plen) |
524 | { |
525 | PyObject *func, *result; |
526 | /* is it a bytes object? */ |
527 | if (PyBytes_Check(v)) { |
528 | *pbuf = PyBytes_AS_STRING(v); |
529 | *plen = PyBytes_GET_SIZE(v); |
530 | Py_INCREF(v); |
531 | return v; |
532 | } |
533 | if (PyByteArray_Check(v)) { |
534 | *pbuf = PyByteArray_AS_STRING(v); |
535 | *plen = PyByteArray_GET_SIZE(v); |
536 | Py_INCREF(v); |
537 | return v; |
538 | } |
539 | /* does it support __bytes__? */ |
540 | func = _PyObject_LookupSpecial(v, &_Py_ID(__bytes__)); |
541 | if (func != NULL) { Branch (541:9): [True: 4, False: 6]
|
542 | result = _PyObject_CallNoArgs(func); |
543 | Py_DECREF(func); |
544 | if (result == NULL) Branch (544:13): [True: 0, False: 4]
|
545 | return NULL; |
546 | if (!PyBytes_Check(result)) { Branch (546:13): [True: 0, False: 4]
|
547 | PyErr_Format(PyExc_TypeError, |
548 | "__bytes__ returned non-bytes (type %.200s)", |
549 | Py_TYPE(result)->tp_name); |
550 | Py_DECREF(result); |
551 | return NULL; |
552 | } |
553 | *pbuf = PyBytes_AS_STRING(result); |
554 | *plen = PyBytes_GET_SIZE(result); |
555 | return result; |
556 | } |
557 | /* does it support buffer protocol? */ |
558 | if (PyObject_CheckBuffer(v)) { Branch (558:9): [True: 4, False: 2]
|
559 | /* maybe we can avoid making a copy of the buffer object here? */ |
560 | result = _PyBytes_FromBuffer(v); |
561 | if (result == NULL) Branch (561:13): [True: 0, False: 4]
|
562 | return NULL; |
563 | *pbuf = PyBytes_AS_STRING(result); |
564 | *plen = PyBytes_GET_SIZE(result); |
565 | return result; |
566 | } |
567 | PyErr_Format(PyExc_TypeError, |
568 | "%%b requires a bytes-like object, " |
569 | "or an object that implements __bytes__, not '%.100s'", |
570 | Py_TYPE(v)->tp_name); |
571 | return NULL; |
572 | } |
573 | |
574 | /* fmt%(v1,v2,...) is roughly equivalent to sprintf(fmt, v1, v2, ...) */ |
575 | |
576 | PyObject * |
577 | _PyBytes_FormatEx(const char *format, Py_ssize_t format_len, |
578 | PyObject *args, int use_bytearray) |
579 | { |
580 | const char *fmt; |
581 | char *res; |
582 | Py_ssize_t arglen, argidx; |
583 | Py_ssize_t fmtcnt; |
584 | int args_owned = 0; |
585 | PyObject *dict = NULL; |
586 | _PyBytesWriter writer; |
587 | |
588 | if (args == NULL) { Branch (588:9): [True: 0, False: 941]
|
589 | PyErr_BadInternalCall(); |
590 | return NULL; |
591 | } |
592 | fmt = format; |
593 | fmtcnt = format_len; |
594 | |
595 | _PyBytesWriter_Init(&writer); |
596 | writer.use_bytearray = use_bytearray; |
597 | |
598 | res = _PyBytesWriter_Alloc(&writer, fmtcnt); |
599 | if (res == NULL) Branch (599:9): [True: 0, False: 941]
|
600 | return NULL; |
601 | if (!use_bytearray) Branch (601:9): [True: 776, False: 165]
|
602 | writer.overallocate = 1; |
603 | |
604 | if (PyTuple_Check(args)) { |
605 | arglen = PyTuple_GET_SIZE(args); |
606 | argidx = 0; |
607 | } |
608 | else { |
609 | arglen = -1; |
610 | argidx = -2; |
611 | } |
612 | if (Py_TYPE(args)->tp_as_mapping && Py_TYPE357 (args)->tp_as_mapping->mp_subscript357 && Branch (612:9): [True: 357, False: 584]
Branch (612:41): [True: 357, False: 0]
|
613 | !357 PyTuple_Check(args) && !28 PyBytes_Check(args) && !7 PyUnicode_Check(args) && Branch (613:9): [True: 28, False: 329]
Branch (613:33): [True: 7, False: 21]
Branch (613:57): [True: 1, False: 6]
|
614 | !1 PyByteArray_Check1 (args)) { Branch (614:9): [True: 0, False: 1]
|
615 | dict = args; |
616 | } |
617 | |
618 | while (--fmtcnt >= 0) { Branch (618:12): [True: 1.59k, False: 923]
|
619 | if (*fmt != '%') { Branch (619:13): [True: 646, False: 950]
|
620 | Py_ssize_t len; |
621 | char *pos; |
622 | |
623 | pos = (char *)memchr(fmt + 1, '%', fmtcnt); |
624 | if (pos != NULL) Branch (624:17): [True: 582, False: 64]
|
625 | len = pos - fmt; |
626 | else |
627 | len = fmtcnt + 1; |
628 | assert(len != 0); |
629 | |
630 | memcpy(res, fmt, len); |
631 | res += len; |
632 | fmt += len; |
633 | fmtcnt -= (len - 1); |
634 | } |
635 | else { |
636 | /* Got a format specifier */ |
637 | int flags = 0; |
638 | Py_ssize_t width = -1; |
639 | int prec = -1; |
640 | int c = '\0'; |
641 | int fill; |
642 | PyObject *v = NULL; |
643 | PyObject *temp = NULL; |
644 | const char *pbuf = NULL; |
645 | int sign; |
646 | Py_ssize_t len = 0; |
647 | char onechar; /* For byte_converter() */ |
648 | Py_ssize_t alloc; |
649 | |
650 | fmt++; |
651 | if (*fmt == '%') { Branch (651:17): [True: 6, False: 944]
|
652 | *res++ = '%'; |
653 | fmt++; |
654 | fmtcnt--; |
655 | continue; |
656 | } |
657 | if (*fmt == '(') { Branch (657:17): [True: 0, False: 944]
|
658 | const char *keystart; |
659 | Py_ssize_t keylen; |
660 | PyObject *key; |
661 | int pcount = 1; |
662 |
|
663 | if (dict == NULL) { Branch (663:21): [True: 0, False: 0]
|
664 | PyErr_SetString(PyExc_TypeError, |
665 | "format requires a mapping"); |
666 | goto error; |
667 | } |
668 | ++fmt; |
669 | --fmtcnt; |
670 | keystart = fmt; |
671 | /* Skip over balanced parentheses */ |
672 | while (pcount > 0 && --fmtcnt >= 0) { Branch (672:24): [True: 0, False: 0]
Branch (672:38): [True: 0, False: 0]
|
673 | if (*fmt == ')') Branch (673:25): [True: 0, False: 0]
|
674 | --pcount; |
675 | else if (*fmt == '(') Branch (675:30): [True: 0, False: 0]
|
676 | ++pcount; |
677 | fmt++; |
678 | } |
679 | keylen = fmt - keystart - 1; |
680 | if (fmtcnt < 0 || pcount > 0) { Branch (680:21): [True: 0, False: 0]
Branch (680:35): [True: 0, False: 0]
|
681 | PyErr_SetString(PyExc_ValueError, |
682 | "incomplete format key"); |
683 | goto error; |
684 | } |
685 | key = PyBytes_FromStringAndSize(keystart, |
686 | keylen); |
687 | if (key == NULL) Branch (687:21): [True: 0, False: 0]
|
688 | goto error; |
689 | if (args_owned) { Branch (689:21): [True: 0, False: 0]
|
690 | Py_DECREF(args); |
691 | args_owned = 0; |
692 | } |
693 | args = PyObject_GetItem(dict, key); |
694 | Py_DECREF(key); |
695 | if (args == NULL) { Branch (695:21): [True: 0, False: 0]
|
696 | goto error; |
697 | } |
698 | args_owned = 1; |
699 | arglen = -1; |
700 | argidx = -2; |
701 | } |
702 | |
703 | /* Parse flags. Example: "%+i" => flags=F_SIGN. */ |
704 | while (944 --fmtcnt >= 0) { Branch (704:20): [True: 1.18k, False: 1]
|
705 | switch (c = *fmt++) { Branch (705:25): [True: 943, False: 237]
|
706 | case '-': flags |= F_LJUST; continue; Branch (706:17): [True: 34, False: 1.14k]
|
707 | case '+': flags |= F_SIGN; continue; Branch (707:17): [True: 30, False: 1.15k]
|
708 | case ' ': flags |= F_BLANK; continue; Branch (708:17): [True: 13, False: 1.16k]
|
709 | case '#': flags |= F_ALT; continue; Branch (709:17): [True: 110, False: 1.07k]
|
710 | case '0': flags |= F_ZERO; continue; Branch (710:17): [True: 50, False: 1.13k]
|
711 | case 'z': flags |= F_NO_NEG_0; continue; Branch (711:17): [True: 0, False: 1.18k]
|
712 | } |
713 | break; |
714 | } |
715 | |
716 | /* Parse width. Example: "%10s" => width=10 */ |
717 | if (c == '*') { Branch (717:17): [True: 0, False: 944]
|
718 | v = getnextarg(args, arglen, &argidx); |
719 | if (v == NULL) Branch (719:21): [True: 0, False: 0]
|
720 | goto error; |
721 | if (!PyLong_Check(v)) { Branch (721:21): [True: 0, False: 0]
|
722 | PyErr_SetString(PyExc_TypeError, |
723 | "* wants int"); |
724 | goto error; |
725 | } |
726 | width = PyLong_AsSsize_t(v); |
727 | if (width == -1 && PyErr_Occurred()) Branch (727:21): [True: 0, False: 0]
Branch (727:36): [True: 0, False: 0]
|
728 | goto error; |
729 | if (width < 0) { Branch (729:21): [True: 0, False: 0]
|
730 | flags |= F_LJUST; |
731 | width = -width; |
732 | } |
733 | if (--fmtcnt >= 0) Branch (733:21): [True: 0, False: 0]
|
734 | c = *fmt++; |
735 | } |
736 | else if (c >= 0 && isdigit(c)) { Branch (736:22): [True: 944, False: 0]
|
737 | width = c - '0'; |
738 | while (--fmtcnt >= 0) { Branch (738:24): [True: 250, False: 0]
|
739 | c = Py_CHARMASK(*fmt++); |
740 | if (!isdigit(c)) Branch (740:25): [True: 130, False: 120]
|
741 | break; |
742 | if (width > (PY_SSIZE_T_MAX - ((int)c - '0')) / 10) { Branch (742:25): [True: 0, False: 120]
|
743 | PyErr_SetString( |
744 | PyExc_ValueError, |
745 | "width too big"); |
746 | goto error; |
747 | } |
748 | width = width*10 + (c - '0'); |
749 | } |
750 | } |
751 | |
752 | /* Parse precision. Example: "%.3f" => prec=3 */ |
753 | if (c == '.') { Branch (753:17): [True: 114, False: 830]
|
754 | prec = 0; |
755 | if (--fmtcnt >= 0) Branch (755:21): [True: 114, False: 0]
|
756 | c = *fmt++; |
757 | if (c == '*') { Branch (757:21): [True: 18, False: 96]
|
758 | v = getnextarg(args, arglen, &argidx); |
759 | if (v == NULL) Branch (759:25): [True: 0, False: 18]
|
760 | goto error; |
761 | if (!PyLong_Check(v)) { Branch (761:25): [True: 0, False: 18]
|
762 | PyErr_SetString( |
763 | PyExc_TypeError, |
764 | "* wants int"); |
765 | goto error; |
766 | } |
767 | prec = _PyLong_AsInt(v); |
768 | if (prec == -1 && PyErr_Occurred()2 ) Branch (768:25): [True: 2, False: 16]
Branch (768:39): [True: 2, False: 0]
|
769 | goto error; |
770 | if (prec < 0) Branch (770:25): [True: 0, False: 16]
|
771 | prec = 0; |
772 | if (--fmtcnt >= 0) Branch (772:25): [True: 16, False: 0]
|
773 | c = *fmt++; |
774 | } |
775 | else if (c >= 0 && isdigit(c)) { Branch (775:26): [True: 96, False: 0]
|
776 | prec = c - '0'; |
777 | while (--fmtcnt >= 0) { Branch (777:28): [True: 190, False: 0]
|
778 | c = Py_CHARMASK(*fmt++); |
779 | if (!isdigit(c)) Branch (779:29): [True: 96, False: 94]
|
780 | break; |
781 | if (prec > (INT_MAX - ((int)c - '0')) / 10) { Branch (781:29): [True: 0, False: 94]
|
782 | PyErr_SetString( |
783 | PyExc_ValueError, |
784 | "prec too big"); |
785 | goto error; |
786 | } |
787 | prec = prec*10 + (c - '0'); |
788 | } |
789 | } |
790 | } /* prec */ |
791 | if (fmtcnt >= 0) { Branch (791:17): [True: 941, False: 1]
|
792 | if (c == 'h' || c == 'l' || c == 'L') { Branch (792:21): [True: 0, False: 941]
Branch (792:33): [True: 0, False: 941]
Branch (792:45): [True: 0, False: 941]
|
793 | if (--fmtcnt >= 0) Branch (793:25): [True: 0, False: 0]
|
794 | c = *fmt++; |
795 | } |
796 | } |
797 | if (fmtcnt < 0) { Branch (797:17): [True: 1, False: 941]
|
798 | PyErr_SetString(PyExc_ValueError, |
799 | "incomplete format"); |
800 | goto error; |
801 | } |
802 | v = getnextarg(args, arglen, &argidx); |
803 | if (v == NULL) Branch (803:17): [True: 0, False: 941]
|
804 | goto error; |
805 | |
806 | if (fmtcnt == 0) { Branch (806:17): [True: 867, False: 74]
|
807 | /* last write: disable writer overallocation */ |
808 | writer.overallocate = 0; |
809 | } |
810 | |
811 | sign = 0; |
812 | fill = ' '; |
813 | switch (c) { |
814 | case 'r': Branch (814:13): [True: 8, False: 933]
|
815 | // %r is only for 2/3 code; 3 only code should use %a |
816 | case 'a': Branch (816:13): [True: 8, False: 933]
|
817 | temp = PyObject_ASCII(v); |
818 | if (temp == NULL) Branch (818:21): [True: 0, False: 16]
|
819 | goto error; |
820 | assert(PyUnicode_IS_ASCII(temp)); |
821 | pbuf = (const char *)PyUnicode_1BYTE_DATA(temp); |
822 | len = PyUnicode_GET_LENGTH(temp); |
823 | if (prec >= 0 && len > prec0 ) Branch (823:21): [True: 0, False: 16]
Branch (823:34): [True: 0, False: 0]
|
824 | len = prec; |
825 | break; |
826 | |
827 | case 's': Branch (827:13): [True: 31, False: 910]
|
828 | // %s is only for 2/3 code; 3 only code should use %b |
829 | case 'b': Branch (829:13): [True: 17, False: 924]
|
830 | temp = format_obj(v, &pbuf, &len); |
831 | if (temp == NULL) Branch (831:21): [True: 2, False: 46]
|
832 | goto error; |
833 | if (prec >= 0 && len > prec0 ) Branch (833:21): [True: 0, False: 46]
Branch (833:34): [True: 0, False: 0]
|
834 | len = prec; |
835 | break; |
836 | |
837 | case 'i': Branch (837:13): [True: 0, False: 941]
|
838 | case 'd': Branch (838:13): [True: 128, False: 813]
|
839 | case 'u': Branch (839:13): [True: 0, False: 941]
|
840 | case 'o': Branch (840:13): [True: 608, False: 333]
|
841 | case 'x': Branch (841:13): [True: 82, False: 859]
|
842 | case 'X': Branch (842:13): [True: 16, False: 925]
|
843 | if (PyLong_CheckExact(v) |
844 | && width == -1826 && prec == -1702 Branch (844:24): [True: 702, False: 124]
Branch (844:39): [True: 652, False: 50]
|
845 | && !(flags & (652 F_SIGN652 | F_BLANK652 )) Branch (845:24): [True: 652, False: 0]
|
846 | && c != 'X'652 ) Branch (846:24): [True: 644, False: 8]
|
847 | { |
848 | /* Fast path */ |
849 | int alternate = flags & F_ALT; |
850 | int base; |
851 | |
852 | switch(c) |
853 | { |
854 | default: Branch (854:25): [True: 0, False: 644]
|
855 | Py_UNREACHABLE(); |
856 | case 'd': Branch (856:25): [True: 88, False: 556]
|
857 | case 'i': Branch (857:25): [True: 0, False: 644]
|
858 | case 'u': Branch (858:25): [True: 0, False: 644]
|
859 | base = 10; |
860 | break; |
861 | case 'o': Branch (861:25): [True: 536, False: 108]
|
862 | base = 8; |
863 | break; |
864 | case 'x': Branch (864:25): [True: 20, False: 624]
|
865 | case 'X': Branch (865:25): [True: 0, False: 644]
|
866 | base = 16; |
867 | break; |
868 | } |
869 | |
870 | /* Fast path */ |
871 | writer.min_size -= 2; /* size preallocated for "%d" */ |
872 | res = _PyLong_FormatBytesWriter(&writer, res, |
873 | v, base, alternate); |
874 | if (res == NULL) Branch (874:25): [True: 0, False: 644]
|
875 | goto error; |
876 | continue; |
877 | } |
878 | |
879 | temp = formatlong(v, flags, prec, c); |
880 | if (!temp) Branch (880:21): [True: 4, False: 186]
|
881 | goto error; |
882 | assert(PyUnicode_IS_ASCII(temp)); |
883 | pbuf = (const char *)PyUnicode_1BYTE_DATA(temp); |
884 | len = PyUnicode_GET_LENGTH(temp); |
885 | sign = 1; |
886 | if (flags & F_ZERO) Branch (886:21): [True: 50, False: 136]
|
887 | fill = '0'; |
888 | break; |
889 | |
890 | case 'e': Branch (890:13): [True: 0, False: 941]
|
891 | case 'E': Branch (891:13): [True: 0, False: 941]
|
892 | case 'f': Branch (892:13): [True: 6, False: 935]
|
893 | case 'F': Branch (893:13): [True: 2, False: 939]
|
894 | case 'g': Branch (894:13): [True: 14, False: 927]
|
895 | case 'G': Branch (895:13): [True: 2, False: 939]
|
896 | if (width == -1 && prec == -122 Branch (896:21): [True: 22, False: 2]
Branch (896:36): [True: 8, False: 14]
|
897 | && !(flags & (8 F_SIGN8 | F_BLANK8 ))) Branch (897:24): [True: 8, False: 0]
|
898 | { |
899 | /* Fast path */ |
900 | writer.min_size -= 2; /* size preallocated for "%f" */ |
901 | res = formatfloat(v, flags, prec, c, NULL, &writer, res); |
902 | if (res == NULL) Branch (902:25): [True: 2, False: 6]
|
903 | goto error; |
904 | continue; |
905 | } |
906 | |
907 | if (!formatfloat(v, flags, prec, c, &temp, NULL, res)) Branch (907:21): [True: 0, False: 16]
|
908 | goto error; |
909 | pbuf = PyBytes_AS_STRING(temp); |
910 | len = PyBytes_GET_SIZE(temp); |
911 | sign = 1; |
912 | if (flags & F_ZERO) Branch (912:21): [True: 0, False: 16]
|
913 | fill = '0'; |
914 | break; |
915 | |
916 | case 'c': Branch (916:13): [True: 18, False: 923]
|
917 | pbuf = &onechar; |
918 | len = byte_converter(v, &onechar); |
919 | if (!len) Branch (919:21): [True: 6, False: 12]
|
920 | goto error; |
921 | if (width == -1) { Branch (921:21): [True: 8, False: 4]
|
922 | /* Fast path */ |
923 | *res++ = onechar; |
924 | continue; |
925 | } |
926 | break; |
927 | |
928 | default: Branch (928:13): [True: 1, False: 940]
|
929 | PyErr_Format(PyExc_ValueError, |
930 | "unsupported format character '%c' (0x%x) " |
931 | "at index %zd", |
932 | c, c, |
933 | (Py_ssize_t)(fmt - 1 - format)); |
934 | goto error; |
935 | } |
936 | |
937 | if (sign) { Branch (937:17): [True: 202, False: 66]
|
938 | if (*pbuf == '-' || *pbuf == '+'142 ) { Branch (938:21): [True: 60, False: 142]
Branch (938:37): [True: 0, False: 142]
|
939 | sign = *pbuf++; |
940 | len--; |
941 | } |
942 | else if (flags & F_SIGN) Branch (942:26): [True: 30, False: 112]
|
943 | sign = '+'; |
944 | else if (flags & F_BLANK) Branch (944:26): [True: 12, False: 100]
|
945 | sign = ' '; |
946 | else |
947 | sign = 0; |
948 | } |
949 | if (width < len) Branch (949:17): [True: 146, False: 122]
|
950 | width = len; |
951 | |
952 | alloc = width; |
953 | if (sign != 0 && len == width102 ) Branch (953:17): [True: 102, False: 166]
Branch (953:30): [True: 34, False: 68]
|
954 | alloc++; |
955 | /* 2: size preallocated for %s */ |
956 | if (alloc > 2) { Branch (956:17): [True: 262, False: 6]
|
957 | res = _PyBytesWriter_Prepare(&writer, res, alloc - 2); |
958 | if (res == NULL) Branch (958:21): [True: 0, False: 262]
|
959 | goto error; |
960 | } |
961 | #ifndef NDEBUG |
962 | char *before = res; |
963 | #endif |
964 | |
965 | /* Write the sign if needed */ |
966 | if (sign) { Branch (966:17): [True: 102, False: 166]
|
967 | if (fill != ' ') Branch (967:21): [True: 32, False: 70]
|
968 | *res++ = sign; |
969 | if (width > len) Branch (969:21): [True: 68, False: 34]
|
970 | width--; |
971 | } |
972 | |
973 | /* Write the numeric prefix for "x", "X" and "o" formats |
974 | if the alternate form is used. |
975 | For example, write "0x" for the "%#x" format. */ |
976 | if ((flags & F_ALT) && (92 c == 'o'92 || c == 'x'56 || c == 'X'28 )) { Branch (976:17): [True: 92, False: 176]
Branch (976:37): [True: 36, False: 56]
Branch (976:49): [True: 28, False: 28]
Branch (976:61): [True: 14, False: 14]
|
977 | assert(pbuf[0] == '0'); |
978 | assert(pbuf[1] == c); |
979 | if (fill != ' ') { Branch (979:21): [True: 18, False: 60]
|
980 | *res++ = *pbuf++; |
981 | *res++ = *pbuf++; |
982 | } |
983 | width -= 2; |
984 | if (width < 0) Branch (984:21): [True: 0, False: 78]
|
985 | width = 0; |
986 | len -= 2; |
987 | } |
988 | |
989 | /* Pad left with the fill character if needed */ |
990 | if (width > len && !(flags & 114 F_LJUST114 )) { Branch (990:17): [True: 114, False: 154]
Branch (990:32): [True: 80, False: 34]
|
991 | memset(res, fill, width - len); |
992 | res += (width - len); |
993 | width = len; |
994 | } |
995 | |
996 | /* If padding with spaces: write sign if needed and/or numeric |
997 | prefix if the alternate form is used */ |
998 | if (fill == ' ') { Branch (998:17): [True: 218, False: 50]
|
999 | if (sign) Branch (999:21): [True: 70, False: 148]
|
1000 | *res++ = sign; |
1001 | if ((flags & F_ALT) && (74 c == 'o'74 || c == 'x'48 || c == 'X'24 )) { Branch (1001:21): [True: 74, False: 144]
Branch (1001:41): [True: 26, False: 48]
Branch (1001:53): [True: 24, False: 24]
Branch (1001:65): [True: 10, False: 14]
|
1002 | assert(pbuf[0] == '0'); |
1003 | assert(pbuf[1] == c); |
1004 | *res++ = *pbuf++; |
1005 | *res++ = *pbuf++; |
1006 | } |
1007 | } |
1008 | |
1009 | /* Copy bytes */ |
1010 | memcpy(res, pbuf, len); |
1011 | res += len; |
1012 | |
1013 | /* Pad right with the fill character if needed */ |
1014 | if (width > len) { Branch (1014:17): [True: 34, False: 234]
|
1015 | memset(res, ' ', width - len); |
1016 | res += (width - len); |
1017 | } |
1018 | |
1019 | if (dict && (argidx < arglen)0 ) { Branch (1019:17): [True: 0, False: 268]
Branch (1019:25): [True: 0, False: 0]
|
1020 | PyErr_SetString(PyExc_TypeError, |
1021 | "not all arguments converted during bytes formatting"); |
1022 | Py_XDECREF(temp); |
1023 | goto error; |
1024 | } |
1025 | Py_XDECREF(temp); |
1026 | |
1027 | #ifndef NDEBUG |
1028 | /* check that we computed the exact size for this write */ |
1029 | assert((res - before) == alloc); |
1030 | #endif |
1031 | } /* '%' */ |
1032 | |
1033 | /* If overallocation was disabled, ensure that it was the last |
1034 | write. Otherwise, we missed an optimization */ |
1035 | assert(writer.overallocate || fmtcnt == 0 || use_bytearray); |
1036 | } /* until end */ |
1037 | |
1038 | if (argidx < arglen && !dict3 ) { Branch (1038:9): [True: 3, False: 920]
Branch (1038:28): [True: 3, False: 0]
|
1039 | PyErr_SetString(PyExc_TypeError, |
1040 | "not all arguments converted during bytes formatting"); |
1041 | goto error; |
1042 | } |
1043 | |
1044 | if (args_owned) { Branch (1044:9): [True: 0, False: 920]
|
1045 | Py_DECREF(args); |
1046 | } |
1047 | return _PyBytesWriter_Finish(&writer, res); |
1048 | |
1049 | error: |
1050 | _PyBytesWriter_Dealloc(&writer); |
1051 | if (args_owned) { Branch (1051:9): [True: 0, False: 21]
|
1052 | Py_DECREF(args); |
1053 | } |
1054 | return NULL; |
1055 | } |
1056 | |
1057 | /* Unescape a backslash-escaped string. */ |
1058 | PyObject *_PyBytes_DecodeEscape(const char *s, |
1059 | Py_ssize_t len, |
1060 | const char *errors, |
1061 | const char **first_invalid_escape) |
1062 | { |
1063 | int c; |
1064 | char *p; |
1065 | const char *end; |
1066 | _PyBytesWriter writer; |
1067 | |
1068 | _PyBytesWriter_Init(&writer); |
1069 | |
1070 | p = _PyBytesWriter_Alloc(&writer, len); |
1071 | if (p == NULL) Branch (1071:9): [True: 0, False: 11.0k]
|
1072 | return NULL; |
1073 | writer.overallocate = 1; |
1074 | |
1075 | *first_invalid_escape = NULL; |
1076 | |
1077 | end = s + len; |
1078 | while (s < end) { Branch (1078:12): [True: 84.2k, False: 11.0k]
|
1079 | if (*s != '\\') { Branch (1079:13): [True: 72.8k, False: 11.3k]
|
1080 | *p++ = *s++; |
1081 | continue; |
1082 | } |
1083 | |
1084 | s++; |
1085 | if (s == end) { Branch (1085:13): [True: 3, False: 11.3k]
|
1086 | PyErr_SetString(PyExc_ValueError, |
1087 | "Trailing \\ in string"); |
1088 | goto failed; |
1089 | } |
1090 | |
1091 | switch (*s++) { |
1092 | /* XXX This assumes ASCII! */ |
1093 | case '\n': break; Branch (1093:9): [True: 2, False: 11.3k]
|
1094 | case '\\': *p++ = '\\'; break; Branch (1094:9): [True: 238, False: 11.1k]
|
1095 | case '\'': *p++ = '\''; break; Branch (1095:9): [True: 4, False: 11.3k]
|
1096 | case '\"': *p++ = '\"'; break; Branch (1096:9): [True: 1, False: 11.3k]
|
1097 | case 'b': *p++ = '\b'; break; Branch (1097:9): [True: 2, False: 11.3k]
|
1098 | case 'f': *p++ = '\014'; break; /* FF */ Branch (1098:9): [True: 10, False: 11.3k]
|
1099 | case 't': *p++ = '\t'; break; Branch (1099:9): [True: 591, False: 10.8k]
|
1100 | case 'n': *p++ = '\n'; break; Branch (1100:9): [True: 842, False: 10.5k]
|
1101 | case 'r': *p++ = '\r'; break; Branch (1101:9): [True: 640, False: 10.7k]
|
1102 | case 'v': *p++ = '\013'; break; /* VT */ Branch (1102:9): [True: 8, False: 11.3k]
|
1103 | case 'a': *p++ = '\007'; break; /* BEL, not classic C */ Branch (1103:9): [True: 2, False: 11.3k]
|
1104 | case '0': 82 case '1': 85 case '2': 90 case '3': Branch (1104:9): [True: 82, False: 11.3k]
Branch (1104:19): [True: 3, False: 11.3k]
Branch (1104:29): [True: 5, False: 11.3k]
Branch (1104:39): [True: 4, False: 11.3k]
|
1105 | case '4': 226 case '5': 354 case '6': 482 case '7': Branch (1105:9): [True: 132, False: 11.2k]
Branch (1105:19): [True: 128, False: 11.2k]
Branch (1105:29): [True: 128, False: 11.2k]
Branch (1105:39): [True: 130, False: 11.2k]
|
1106 | c = s[-1] - '0'; |
1107 | if (s < end && '0' <= *s591 && *s <= '7'591 ) { Branch (1107:17): [True: 591, False: 21]
Branch (1107:28): [True: 591, False: 0]
Branch (1107:41): [True: 578, False: 13]
|
1108 | c = (c<<3) + *s++ - '0'; |
1109 | if (s < end && '0' <= *s && *s <= '7') Branch (1109:21): [True: 578, False: 0]
Branch (1109:32): [True: 578, False: 0]
Branch (1109:45): [True: 576, False: 2]
|
1110 | c = (c<<3) + *s++ - '0'; |
1111 | } |
1112 | if (c > 0377) { Branch (1112:17): [True: 514, False: 98]
|
1113 | if (*first_invalid_escape == NULL) { Branch (1113:21): [True: 514, False: 0]
|
1114 | *first_invalid_escape = s-3; /* Back up 3 chars, since we've |
1115 | already incremented s. */ |
1116 | } |
1117 | } |
1118 | *p++ = c; |
1119 | break; |
1120 | case 'x': Branch (1120:9): [True: 8.28k, False: 3.10k]
|
1121 | if (s+1 < end) { Branch (1121:17): [True: 8.27k, False: 9]
|
1122 | int digit1, digit2; |
1123 | digit1 = _PyLong_DigitValue[Py_CHARMASK(s[0])]; |
1124 | digit2 = _PyLong_DigitValue[Py_CHARMASK(s[1])]; |
1125 | if (digit1 < 16 && digit2 < 168.27k ) { Branch (1125:21): [True: 8.27k, False: 2]
Branch (1125:36): [True: 8.27k, False: 3]
|
1126 | *p++ = (unsigned char)((digit1 << 4) + digit2); |
1127 | s += 2; |
1128 | break; |
1129 | } |
1130 | } |
1131 | /* invalid hexadecimal digits */ |
1132 | |
1133 | if (!errors || strcmp(errors, "strict") == 08 ) { Branch (1133:17): [True: 6, False: 8]
Branch (1133:28): [True: 0, False: 8]
|
1134 | PyErr_Format(PyExc_ValueError, |
1135 | "invalid \\x escape at position %zd", |
1136 | s - 2 - (end - len)); |
1137 | goto failed; |
1138 | } |
1139 | if (strcmp(errors, "replace") == 0) { Branch (1139:17): [True: 4, False: 4]
|
1140 | *p++ = '?'; |
1141 | } else if (strcmp(errors, "ignore") == 0) Branch (1141:24): [True: 4, False: 0]
|
1142 | /* do nothing */; |
1143 | else { |
1144 | PyErr_Format(PyExc_ValueError, |
1145 | "decoding error; unknown " |
1146 | "error handling code: %.400s", |
1147 | errors); |
1148 | goto failed; |
1149 | } |
1150 | /* skip \x */ |
1151 | if (s < end && Py_ISXDIGIT6 (s[0])) Branch (1151:17): [True: 6, False: 2]
|
1152 | s++; /* and a hexdigit */ |
1153 | break; |
1154 | |
1155 | default: Branch (1155:9): [True: 155, False: 11.2k]
|
1156 | if (*first_invalid_escape == NULL) { Branch (1156:17): [True: 155, False: 0]
|
1157 | *first_invalid_escape = s-1; /* Back up one char, since we've |
1158 | already incremented s. */ |
1159 | } |
1160 | *p++ = '\\'; |
1161 | s--; |
1162 | } |
1163 | } |
1164 | |
1165 | return _PyBytesWriter_Finish(&writer, p); |
1166 | |
1167 | failed: |
1168 | _PyBytesWriter_Dealloc(&writer); |
1169 | return NULL; |
1170 | } |
1171 | |
1172 | PyObject *PyBytes_DecodeEscape(const char *s, |
1173 | Py_ssize_t len, |
1174 | const char *errors, |
1175 | Py_ssize_t Py_UNUSED(unicode), |
1176 | const char *Py_UNUSED(recode_encoding)) |
1177 | { |
1178 | const char* first_invalid_escape; |
1179 | PyObject *result = _PyBytes_DecodeEscape(s, len, errors, |
1180 | &first_invalid_escape); |
1181 | if (result == NULL) Branch (1181:9): [True: 7, False: 8.39k]
|
1182 | return NULL; |
1183 | if (first_invalid_escape != NULL) { Branch (1183:9): [True: 303, False: 8.09k]
|
1184 | unsigned char c = *first_invalid_escape; |
1185 | if ('4' <= c && c <= '7') { Branch (1185:13): [True: 303, False: 0]
Branch (1185:25): [True: 256, False: 47]
|
1186 | if (PyErr_WarnFormat(PyExc_DeprecationWarning, 1, Branch (1186:17): [True: 0, False: 256]
|
1187 | "invalid octal escape sequence '\\%.3s'", |
1188 | first_invalid_escape) < 0) |
1189 | { |
1190 | Py_DECREF(result); |
1191 | return NULL; |
1192 | } |
1193 | } |
1194 | else { |
1195 | if (PyErr_WarnFormat(PyExc_DeprecationWarning, 1, Branch (1195:17): [True: 0, False: 47]
|
1196 | "invalid escape sequence '\\%c'", |
1197 | c) < 0) |
1198 | { |
1199 | Py_DECREF(result); |
1200 | return NULL; |
1201 | } |
1202 | } |
1203 | } |
1204 | return result; |
1205 | |
1206 | } |
1207 | /* -------------------------------------------------------------------- */ |
1208 | /* object api */ |
1209 | |
1210 | Py_ssize_t |
1211 | PyBytes_Size(PyObject *op) |
1212 | { |
1213 | if (!PyBytes_Check(op)) { Branch (1213:9): [True: 0, False: 311k]
|
1214 | PyErr_Format(PyExc_TypeError, |
1215 | "expected bytes, %.200s found", Py_TYPE(op)->tp_name); |
1216 | return -1; |
1217 | } |
1218 | return Py_SIZE(op); |
1219 | } |
1220 | |
1221 | char * |
1222 | PyBytes_AsString(PyObject *op) |
1223 | { |
1224 | if (!PyBytes_Check(op)) { Branch (1224:9): [True: 0, False: 8.23M]
|
1225 | PyErr_Format(PyExc_TypeError, |
1226 | "expected bytes, %.200s found", Py_TYPE(op)->tp_name); |
1227 | return NULL; |
1228 | } |
1229 | return ((PyBytesObject *)op)->ob_sval; |
1230 | } |
1231 | |
1232 | int |
1233 | PyBytes_AsStringAndSize(PyObject *obj, |
1234 | char **s, |
1235 | Py_ssize_t *len) |
1236 | { |
1237 | if (s == NULL) { Branch (1237:9): [True: 0, False: 184k]
|
1238 | PyErr_BadInternalCall(); |
1239 | return -1; |
1240 | } |
1241 | |
1242 | if (!PyBytes_Check(obj)) { Branch (1242:9): [True: 0, False: 184k]
|
1243 | PyErr_Format(PyExc_TypeError, |
1244 | "expected bytes, %.200s found", Py_TYPE(obj)->tp_name); |
1245 | return -1; |
1246 | } |
1247 | |
1248 | *s = PyBytes_AS_STRING(obj); |
1249 | if (len != NULL) Branch (1249:9): [True: 1.28k, False: 182k]
|
1250 | *len = PyBytes_GET_SIZE(obj); |
1251 | else if (strlen(*s) != (size_t)PyBytes_GET_SIZE(obj)) { Branch (1251:14): [True: 7, False: 182k]
|
1252 | PyErr_SetString(PyExc_ValueError, |
1253 | "embedded null byte"); |
1254 | return -1; |
1255 | } |
1256 | return 0; |
1257 | } |
1258 | |
1259 | /* -------------------------------------------------------------------- */ |
1260 | /* Methods */ |
1261 | |
1262 | #define STRINGLIB_GET_EMPTY() bytes_get_empty() |
1263 | |
1264 | #include "stringlib/stringdefs.h" |
1265 | #define STRINGLIB_MUTABLE 0 |
1266 | |
1267 | #include "stringlib/fastsearch.h" |
1268 | #include "stringlib/count.h" |
1269 | #include "stringlib/find.h" |
1270 | #include "stringlib/join.h" |
1271 | #include "stringlib/partition.h" |
1272 | #include "stringlib/split.h" |
1273 | #include "stringlib/ctype.h" |
1274 | |
1275 | #include "stringlib/transmogrify.h" |
1276 | |
1277 | #undef STRINGLIB_GET_EMPTY |
1278 | |
1279 | Py_ssize_t |
1280 | _PyBytes_Find(const char *haystack, Py_ssize_t len_haystack, |
1281 | const char *needle, Py_ssize_t len_needle, |
1282 | Py_ssize_t offset) |
1283 | { |
1284 | return stringlib_find(haystack, len_haystack, |
1285 | needle, len_needle, offset); |
1286 | } |
1287 | |
1288 | Py_ssize_t |
1289 | _PyBytes_ReverseFind(const char *haystack, Py_ssize_t len_haystack, |
1290 | const char *needle, Py_ssize_t len_needle, |
1291 | Py_ssize_t offset) |
1292 | { |
1293 | return stringlib_rfind(haystack, len_haystack, |
1294 | needle, len_needle, offset); |
1295 | } |
1296 | |
1297 | PyObject * |
1298 | PyBytes_Repr(PyObject *obj, int smartquotes) |
1299 | { |
1300 | PyBytesObject* op = (PyBytesObject*) obj; |
1301 | Py_ssize_t i, length = Py_SIZE(op); |
1302 | Py_ssize_t newsize, squotes, dquotes; |
1303 | PyObject *v; |
1304 | unsigned char quote; |
1305 | const unsigned char *s; |
1306 | Py_UCS1 *p; |
1307 | |
1308 | /* Compute size of output string */ |
1309 | squotes = dquotes = 0; |
1310 | newsize = 3; /* b'' */ |
1311 | s = (const unsigned char*)op->ob_sval; |
1312 | for (i = 0; i < length; i++336k ) { Branch (1312:17): [True: 336k, False: 50.5k]
|
1313 | Py_ssize_t incr = 1; |
1314 | switch(s[i]) { |
1315 | case '\'': squotes++; break; Branch (1315:9): [True: 362, False: 336k]
|
1316 | case '"': dquotes++; break; Branch (1316:9): [True: 820, False: 336k]
|
1317 | case '\\': 313 case '\t': 3.41k case '\n': 11.7k case '\r': Branch (1317:9): [True: 313, False: 336k]
Branch (1317:20): [True: 3.10k, False: 333k]
Branch (1317:31): [True: 8.29k, False: 328k]
Branch (1317:42): [True: 4.67k, False: 332k]
|
1318 | incr = 2; break; /* \C */ |
1319 | default: Branch (1319:9): [True: 319k, False: 17.5k]
|
1320 | if (s[i] < ' ' || s[i] >= 0x7f261k ) Branch (1320:17): [True: 57.4k, False: 261k]
Branch (1320:31): [True: 30.0k, False: 231k]
|
1321 | incr = 4; /* \xHH */ |
1322 | } |
1323 | if (newsize > PY_SSIZE_T_MAX - incr) Branch (1323:13): [True: 0, False: 336k]
|
1324 | goto overflow; |
1325 | newsize += incr; |
1326 | } |
1327 | quote = '\''; |
1328 | if (smartquotes && squotes && !dquotes213 ) Branch (1328:9): [True: 50.5k, False: 0]
Branch (1328:24): [True: 213, False: 50.3k]
Branch (1328:35): [True: 205, False: 8]
|
1329 | quote = '"'; |
1330 | if (squotes && quote == '\''213 ) { Branch (1330:9): [True: 213, False: 50.3k]
Branch (1330:20): [True: 8, False: 205]
|
1331 | if (newsize > PY_SSIZE_T_MAX - squotes) Branch (1331:13): [True: 0, False: 8]
|
1332 | goto overflow; |
1333 | newsize += squotes; |
1334 | } |
1335 | |
1336 | v = PyUnicode_New(newsize, 127); |
1337 | if (v == NULL) { Branch (1337:9): [True: 0, False: 50.5k]
|
1338 | return NULL; |
1339 | } |
1340 | p = PyUnicode_1BYTE_DATA(v); |
1341 | |
1342 | *p++ = 'b', *p++ = quote; |
1343 | for (i = 0; i < length; i++336k ) { Branch (1343:17): [True: 336k, False: 50.5k]
|
1344 | unsigned char c = op->ob_sval[i]; |
1345 | if (c == quote || c == '\\'336k ) Branch (1345:13): [True: 12, False: 336k]
Branch (1345:27): [True: 313, False: 336k]
|
1346 | *p++ = '\\', *p++ = c; |
1347 | else if (c == '\t') Branch (1347:18): [True: 3.10k, False: 333k]
|
1348 | *p++ = '\\', *p++ = 't'; |
1349 | else if (c == '\n') Branch (1349:18): [True: 8.29k, False: 325k]
|
1350 | *p++ = '\\', *p++ = 'n'; |
1351 | else if (c == '\r') Branch (1351:18): [True: 4.67k, False: 320k]
|
1352 | *p++ = '\\', *p++ = 'r'; |
1353 | else if (c < ' ' || c >= 0x7f263k ) { Branch (1353:18): [True: 57.4k, False: 263k]
Branch (1353:29): [True: 30.0k, False: 232k]
|
1354 | *p++ = '\\'; |
1355 | *p++ = 'x'; |
1356 | *p++ = Py_hexdigits[(c & 0xf0) >> 4]; |
1357 | *p++ = Py_hexdigits[c & 0xf]; |
1358 | } |
1359 | else |
1360 | *p++ = c; |
1361 | } |
1362 | *p++ = quote; |
1363 | assert(_PyUnicode_CheckConsistency(v, 1)); |
1364 | return v; |
1365 | |
1366 | overflow: |
1367 | PyErr_SetString(PyExc_OverflowError, |
1368 | "bytes object is too large to make repr"); |
1369 | return NULL; |
1370 | } |
1371 | |
1372 | static PyObject * |
1373 | bytes_repr(PyObject *op) |
1374 | { |
1375 | return PyBytes_Repr(op, 1); |
1376 | } |
1377 | |
1378 | static PyObject * |
1379 | bytes_str(PyObject *op) |
1380 | { |
1381 | if (_Py_GetConfig()->bytes_warning) { Branch (1381:9): [True: 0, False: 29]
|
1382 | if (PyErr_WarnEx(PyExc_BytesWarning, Branch (1382:13): [True: 0, False: 0]
|
1383 | "str() on a bytes instance", 1)) { |
1384 | return NULL; |
1385 | } |
1386 | } |
1387 | return bytes_repr(op); |
1388 | } |
1389 | |
1390 | static Py_ssize_t |
1391 | bytes_length(PyBytesObject *a) |
1392 | { |
1393 | return Py_SIZE(a); |
1394 | } |
1395 | |
1396 | /* This is also used by PyBytes_Concat() */ |
1397 | static PyObject * |
1398 | bytes_concat(PyObject *a, PyObject *b) |
1399 | { |
1400 | Py_buffer va, vb; |
1401 | PyObject *result = NULL; |
1402 | |
1403 | va.len = -1; |
1404 | vb.len = -1; |
1405 | if (PyObject_GetBuffer(a, &va, PyBUF_SIMPLE) != 0 || Branch (1405:9): [True: 0, False: 2.16M]
|
1406 | PyObject_GetBuffer(b, &vb, PyBUF_SIMPLE) != 0) { Branch (1406:9): [True: 6, False: 2.16M]
|
1407 | PyErr_Format(PyExc_TypeError, "can't concat %.100s to %.100s", |
1408 | Py_TYPE(b)->tp_name, Py_TYPE(a)->tp_name); |
1409 | goto done; |
1410 | } |
1411 | |
1412 | /* Optimize end cases */ |
1413 | if (va.len == 0 && PyBytes_CheckExact298k (b)) { Branch (1413:9): [True: 298k, False: 1.86M]
|
1414 | result = b; |
1415 | Py_INCREF(result); |
1416 | goto done; |
1417 | } |
1418 | if (vb.len == 0 && PyBytes_CheckExact32.3k (a)) { Branch (1418:9): [True: 32.3k, False: 1.83M]
|
1419 | result = a; |
1420 | Py_INCREF(result); |
1421 | goto done; |
1422 | } |
1423 | |
1424 | if (va.len > PY_SSIZE_T_MAX - vb.len) { Branch (1424:9): [True: 0, False: 1.83M]
|
1425 | PyErr_NoMemory(); |
1426 | goto done; |
1427 | } |
1428 | |
1429 | result = PyBytes_FromStringAndSize(NULL, va.len + vb.len); |
1430 | if (result != NULL) { Branch (1430:9): [True: 1.83M, False: 0]
|
1431 | memcpy(PyBytes_AS_STRING(result), va.buf, va.len); |
1432 | memcpy(PyBytes_AS_STRING(result) + va.len, vb.buf, vb.len); |
1433 | } |
1434 | |
1435 | done: |
1436 | if (va.len != -1) Branch (1436:9): [True: 2.16M, False: 0]
|
1437 | PyBuffer_Release(&va); |
1438 | if (vb.len != -1) Branch (1438:9): [True: 2.16M, False: 6]
|
1439 | PyBuffer_Release(&vb); |
1440 | return result; |
1441 | } |
1442 | |
1443 | static PyObject * |
1444 | bytes_repeat(PyBytesObject *a, Py_ssize_t n) |
1445 | { |
1446 | Py_ssize_t size; |
1447 | PyBytesObject *op; |
1448 | size_t nbytes; |
1449 | if (n < 0) Branch (1449:9): [True: 42, False: 40.2k]
|
1450 | n = 0; |
1451 | /* watch out for overflows: the size can overflow int, |
1452 | * and the # of bytes needed can overflow size_t |
1453 | */ |
1454 | if (n > 0 && Py_SIZE27.3k (a) > 27.3k PY_SSIZE_T_MAX27.3k / n) { Branch (1454:9): [True: 27.3k, False: 12.9k]
Branch (1454:18): [True: 6, False: 27.3k]
|
1455 | PyErr_SetString(PyExc_OverflowError, |
1456 | "repeated bytes are too long"); |
1457 | return NULL; |
1458 | } |
1459 | size = Py_SIZE(a) * n; |
1460 | if (size == Py_SIZE(a) && PyBytes_CheckExact1.89k (a)) { Branch (1460:9): [True: 1.89k, False: 38.4k]
|
1461 | Py_INCREF(a); |
1462 | return (PyObject *)a; |
1463 | } |
1464 | nbytes = (size_t)size; |
1465 | if (nbytes + PyBytesObject_SIZE <= nbytes) { Branch (1465:9): [True: 0, False: 38.4k]
|
1466 | PyErr_SetString(PyExc_OverflowError, |
1467 | "repeated bytes are too long"); |
1468 | return NULL; |
1469 | } |
1470 | op = (PyBytesObject *)PyObject_Malloc(PyBytesObject_SIZE + nbytes); |
1471 | if (op == NULL) { Branch (1471:9): [True: 0, False: 38.4k]
|
1472 | return PyErr_NoMemory(); |
1473 | } |
1474 | _PyObject_InitVar((PyVarObject*)op, &PyBytes_Type, size); |
1475 | _Py_COMP_DIAG_PUSH |
1476 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
1477 | op->ob_shash = -1; |
1478 | _Py_COMP_DIAG_POP |
1479 | op->ob_sval[size] = '\0'; |
1480 | |
1481 | _PyBytes_Repeat(op->ob_sval, size, a->ob_sval, Py_SIZE(a)); |
1482 | |
1483 | return (PyObject *) op; |
1484 | } |
1485 | |
1486 | static int |
1487 | bytes_contains(PyObject *self, PyObject *arg) |
1488 | { |
1489 | return _Py_bytes_contains(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), arg); |
1490 | } |
1491 | |
1492 | static PyObject * |
1493 | bytes_item(PyBytesObject *a, Py_ssize_t i) |
1494 | { |
1495 | if (i < 0 || i >= 115 Py_SIZE115 (a)) { Branch (1495:9): [True: 1, False: 115]
Branch (1495:18): [True: 1, False: 114]
|
1496 | PyErr_SetString(PyExc_IndexError, "index out of range"); |
1497 | return NULL; |
1498 | } |
1499 | return _PyLong_FromUnsignedChar((unsigned char)a->ob_sval[i]); |
1500 | } |
1501 | |
1502 | static int |
1503 | bytes_compare_eq(PyBytesObject *a, PyBytesObject *b) |
1504 | { |
1505 | int cmp; |
1506 | Py_ssize_t len; |
1507 | |
1508 | len = Py_SIZE(a); |
1509 | if (Py_SIZE(b) != len) Branch (1509:9): [True: 272k, False: 4.09M]
|
1510 | return 0; |
1511 | |
1512 | if (a->ob_sval[0] != b->ob_sval[0]) Branch (1512:9): [True: 3.38M, False: 711k]
|
1513 | return 0; |
1514 | |
1515 | cmp = memcmp(a->ob_sval, b->ob_sval, len); |
1516 | return (cmp == 0); |
1517 | } |
1518 | |
1519 | static PyObject* |
1520 | bytes_richcompare(PyBytesObject *a, PyBytesObject *b, int op) |
1521 | { |
1522 | int c; |
1523 | Py_ssize_t len_a, len_b; |
1524 | Py_ssize_t min_len; |
1525 | |
1526 | /* Make sure both arguments are strings. */ |
1527 | if (!(PyBytes_Check(a) && PyBytes_Check(b))) { |
1528 | if (_Py_GetConfig()->bytes_warning && (0 op == 0 Py_EQ0 || op == 0 Py_NE0 )) { Branch (1528:13): [True: 0, False: 51.7k]
Branch (1528:48): [True: 0, False: 0]
Branch (1528:63): [True: 0, False: 0]
|
1529 | if (PyUnicode_Check(a) || PyUnicode_Check(b)) { |
1530 | if (PyErr_WarnEx(PyExc_BytesWarning, Branch (1530:21): [True: 0, False: 0]
|
1531 | "Comparison between bytes and string", 1)) |
1532 | return NULL; |
1533 | } |
1534 | if (PyLong_Check(a) || PyLong_Check(b)) { |
1535 | if (PyErr_WarnEx(PyExc_BytesWarning, Branch (1535:21): [True: 0, False: 0]
|
1536 | "Comparison between bytes and int", 1)) |
1537 | return NULL; |
1538 | } |
1539 | } |
1540 | Py_RETURN_NOTIMPLEMENTED; |
1541 | } |
1542 | else if (a == b) { Branch (1542:14): [True: 152k, False: 4.38M]
|
1543 | switch (op) { |
1544 | case Py_EQ: Branch (1544:9): [True: 18.5k, False: 134k]
|
1545 | case Py_LE: Branch (1545:9): [True: 1, False: 152k]
|
1546 | case Py_GE: Branch (1546:9): [True: 1, False: 152k]
|
1547 | /* a byte string is equal to itself */ |
1548 | Py_RETURN_TRUE; |
1549 | case Py_NE: Branch (1549:9): [True: 134k, False: 18.5k]
|
1550 | case Py_LT: Branch (1550:9): [True: 30, False: 152k]
|
1551 | case Py_GT: Branch (1551:9): [True: 24, False: 152k]
|
1552 | Py_RETURN_FALSE; |
1553 | default: Branch (1553:9): [True: 0, False: 152k]
|
1554 | PyErr_BadArgument(); |
1555 | return NULL; |
1556 | } |
1557 | } |
1558 | else if (op == Py_EQ || op == 60.6k Py_NE60.6k ) { Branch (1558:14): [True: 4.32M, False: 60.6k]
Branch (1558:29): [True: 45.5k, False: 15.1k]
|
1559 | int eq = bytes_compare_eq(a, b); |
1560 | eq ^= (op == Py_NE); |
1561 | return PyBool_FromLong(eq); |
1562 | } |
1563 | else { |
1564 | len_a = Py_SIZE(a); |
1565 | len_b = Py_SIZE(b); |
1566 | min_len = Py_MIN(len_a, len_b); |
1567 | if (min_len > 0) { Branch (1567:13): [True: 15.0k, False: 34]
|
1568 | c = Py_CHARMASK(*a->ob_sval) - Py_CHARMASK(*b->ob_sval); |
1569 | if (c == 0) Branch (1569:17): [True: 4.20k, False: 10.8k]
|
1570 | c = memcmp(a->ob_sval, b->ob_sval, min_len); |
1571 | } |
1572 | else |
1573 | c = 0; |
1574 | if (c != 0) Branch (1574:13): [True: 11.1k, False: 3.96k]
|
1575 | Py_RETURN_RICHCOMPARE(c, 0, op); |
1576 | Py_RETURN_RICHCOMPARE(len_a, len_b, op); |
1577 | } |
1578 | } |
1579 | |
1580 | static Py_hash_t |
1581 | bytes_hash(PyBytesObject *a) |
1582 | { |
1583 | _Py_COMP_DIAG_PUSH |
1584 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
1585 | if (a->ob_shash == -1) { Branch (1585:9): [True: 524k, False: 174k]
|
1586 | /* Can't fail */ |
1587 | a->ob_shash = _Py_HashBytes(a->ob_sval, Py_SIZE(a)); |
1588 | } |
1589 | return a->ob_shash; |
1590 | _Py_COMP_DIAG_POP |
1591 | } |
1592 | |
1593 | static PyObject* |
1594 | bytes_subscript(PyBytesObject* self, PyObject* item) |
1595 | { |
1596 | if (_PyIndex_Check(item)) { Branch (1596:9): [True: 2.90M, False: 2.25M]
|
1597 | Py_ssize_t i = PyNumber_AsSsize_t(item, PyExc_IndexError); |
1598 | if (i == -1 && PyErr_Occurred()4.91k ) Branch (1598:13): [True: 4.91k, False: 2.89M]
Branch (1598:24): [True: 6, False: 4.91k]
|
1599 | return NULL; |
1600 | if (i < 0) Branch (1600:13): [True: 5.01k, False: 2.89M]
|
1601 | i += PyBytes_GET_SIZE(self); |
1602 | if (i < 0 || i >= 2.90M PyBytes_GET_SIZE2.90M (self)) { Branch (1602:13): [True: 4, False: 2.90M]
Branch (1602:22): [True: 16, False: 2.90M]
|
1603 | PyErr_SetString(PyExc_IndexError, |
1604 | "index out of range"); |
1605 | return NULL; |
1606 | } |
1607 | return _PyLong_FromUnsignedChar((unsigned char)self->ob_sval[i]); |
1608 | } |
1609 | else if (PySlice_Check(item)) { |
1610 | Py_ssize_t start, stop, step, slicelength, i; |
1611 | size_t cur; |
1612 | const char* source_buf; |
1613 | char* result_buf; |
1614 | PyObject* result; |
1615 | |
1616 | if (PySlice_Unpack(item, &start, &stop, &step) < 0) { Branch (1616:13): [True: 2, False: 2.25M]
|
1617 | return NULL; |
1618 | } |
1619 | slicelength = PySlice_AdjustIndices(PyBytes_GET_SIZE(self), &start, |
1620 | &stop, step); |
1621 | |
1622 | if (slicelength <= 0) { Branch (1622:13): [True: 308k, False: 1.94M]
|
1623 | return PyBytes_FromStringAndSize("", 0); |
1624 | } |
1625 | else if (start == 0 && step == 1455k && Branch (1625:18): [True: 455k, False: 1.48M]
Branch (1625:32): [True: 454k, False: 806]
|
1626 | slicelength == 454k PyBytes_GET_SIZE454k (self) && Branch (1626:18): [True: 100k, False: 354k]
|
1627 | PyBytes_CheckExact100k (self)) { |
1628 | Py_INCREF(self); |
1629 | return (PyObject *)self; |
1630 | } |
1631 | else if (step == 1) { Branch (1631:18): [True: 1.82M, False: 14.5k]
|
1632 | return PyBytes_FromStringAndSize( |
1633 | PyBytes_AS_STRING(self) + start, |
1634 | slicelength); |
1635 | } |
1636 | else { |
1637 | source_buf = PyBytes_AS_STRING(self); |
1638 | result = PyBytes_FromStringAndSize(NULL, slicelength); |
1639 | if (result == NULL) Branch (1639:17): [True: 0, False: 14.5k]
|
1640 | return NULL; |
1641 | |
1642 | result_buf = PyBytes_AS_STRING(result); |
1643 | for (cur = start, i = 0; i < slicelength; Branch (1643:38): [True: 117k, False: 14.5k]
|
1644 | cur += step, i++) { |
1645 | result_buf[i] = source_buf[cur]; |
1646 | } |
1647 | |
1648 | return result; |
1649 | } |
1650 | } |
1651 | else { |
1652 | PyErr_Format(PyExc_TypeError, |
1653 | "byte indices must be integers or slices, not %.200s", |
1654 | Py_TYPE(item)->tp_name); |
1655 | return NULL; |
1656 | } |
1657 | } |
1658 | |
1659 | static int |
1660 | bytes_buffer_getbuffer(PyBytesObject *self, Py_buffer *view, int flags) |
1661 | { |
1662 | return PyBuffer_FillInfo(view, (PyObject*)self, (void *)self->ob_sval, Py_SIZE(self), |
1663 | 1, flags); |
1664 | } |
1665 | |
1666 | static PySequenceMethods bytes_as_sequence = { |
1667 | (lenfunc)bytes_length, /*sq_length*/ |
1668 | (binaryfunc)bytes_concat, /*sq_concat*/ |
1669 | (ssizeargfunc)bytes_repeat, /*sq_repeat*/ |
1670 | (ssizeargfunc)bytes_item, /*sq_item*/ |
1671 | 0, /*sq_slice*/ |
1672 | 0, /*sq_ass_item*/ |
1673 | 0, /*sq_ass_slice*/ |
1674 | (objobjproc)bytes_contains /*sq_contains*/ |
1675 | }; |
1676 | |
1677 | static PyMappingMethods bytes_as_mapping = { |
1678 | (lenfunc)bytes_length, |
1679 | (binaryfunc)bytes_subscript, |
1680 | 0, |
1681 | }; |
1682 | |
1683 | static PyBufferProcs bytes_as_buffer = { |
1684 | (getbufferproc)bytes_buffer_getbuffer, |
1685 | NULL, |
1686 | }; |
1687 | |
1688 | |
1689 | /*[clinic input] |
1690 | bytes.__bytes__ |
1691 | Convert this value to exact type bytes. |
1692 | [clinic start generated code]*/ |
1693 | |
1694 | static PyObject * |
1695 | bytes___bytes___impl(PyBytesObject *self) |
1696 | /*[clinic end generated code: output=63a306a9bc0caac5 input=34ec5ddba98bd6bb]*/ |
1697 | { |
1698 | if (PyBytes_CheckExact(self)) { |
1699 | Py_INCREF(self); |
1700 | return (PyObject *)self; |
1701 | } |
1702 | else { |
1703 | return PyBytes_FromStringAndSize(self->ob_sval, Py_SIZE(self)); |
1704 | } |
1705 | } |
1706 | |
1707 | |
1708 | #define LEFTSTRIP 0 |
1709 | #define RIGHTSTRIP 1 |
1710 | #define BOTHSTRIP 2 |
1711 | |
1712 | /*[clinic input] |
1713 | bytes.split |
1714 | |
1715 | sep: object = None |
1716 | The delimiter according which to split the bytes. |
1717 | None (the default value) means split on ASCII whitespace characters |
1718 | (space, tab, return, newline, formfeed, vertical tab). |
1719 | maxsplit: Py_ssize_t = -1 |
1720 | Maximum number of splits to do. |
1721 | -1 (the default value) means no limit. |
1722 | |
1723 | Return a list of the sections in the bytes, using sep as the delimiter. |
1724 | [clinic start generated code]*/ |
1725 | |
1726 | static PyObject * |
1727 | bytes_split_impl(PyBytesObject *self, PyObject *sep, Py_ssize_t maxsplit) |
1728 | /*[clinic end generated code: output=52126b5844c1d8ef input=8b809b39074abbfa]*/ |
1729 | { |
1730 | Py_ssize_t len = PyBytes_GET_SIZE(self), n; |
1731 | const char *s = PyBytes_AS_STRING(self), *sub; |
1732 | Py_buffer vsub; |
1733 | PyObject *list; |
1734 | |
1735 | if (maxsplit < 0) Branch (1735:9): [True: 3.21k, False: 3.93k]
|
1736 | maxsplit = PY_SSIZE_T_MAX; |
1737 | if (sep == Py_None) Branch (1737:9): [True: 146, False: 7.00k]
|
1738 | return stringlib_split_whitespace((PyObject*) self, s, len, maxsplit); |
1739 | if (PyObject_GetBuffer(sep, &vsub, PyBUF_SIMPLE) != 0) Branch (1739:9): [True: 7, False: 6.99k]
|
1740 | return NULL; |
1741 | sub = vsub.buf; |
1742 | n = vsub.len; |
1743 | |
1744 | list = stringlib_split((PyObject*) self, s, len, sub, n, maxsplit); |
1745 | PyBuffer_Release(&vsub); |
1746 | return list; |
1747 | } |
1748 | |
1749 | /*[clinic input] |
1750 | bytes.partition |
1751 | |
1752 | sep: Py_buffer |
1753 | / |
1754 | |
1755 | Partition the bytes into three parts using the given separator. |
1756 | |
1757 | This will search for the separator sep in the bytes. If the separator is found, |
1758 | returns a 3-tuple containing the part before the separator, the separator |
1759 | itself, and the part after it. |
1760 | |
1761 | If the separator is not found, returns a 3-tuple containing the original bytes |
1762 | object and two empty bytes objects. |
1763 | [clinic start generated code]*/ |
1764 | |
1765 | static PyObject * |
1766 | bytes_partition_impl(PyBytesObject *self, Py_buffer *sep) |
1767 | /*[clinic end generated code: output=f532b392a17ff695 input=61cca95519406099]*/ |
1768 | { |
1769 | return stringlib_partition( |
1770 | (PyObject*) self, |
1771 | PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), |
1772 | sep->obj, (const char *)sep->buf, sep->len |
1773 | ); |
1774 | } |
1775 | |
1776 | /*[clinic input] |
1777 | bytes.rpartition |
1778 | |
1779 | sep: Py_buffer |
1780 | / |
1781 | |
1782 | Partition the bytes into three parts using the given separator. |
1783 | |
1784 | This will search for the separator sep in the bytes, starting at the end. If |
1785 | the separator is found, returns a 3-tuple containing the part before the |
1786 | separator, the separator itself, and the part after it. |
1787 | |
1788 | If the separator is not found, returns a 3-tuple containing two empty bytes |
1789 | objects and the original bytes object. |
1790 | [clinic start generated code]*/ |
1791 | |
1792 | static PyObject * |
1793 | bytes_rpartition_impl(PyBytesObject *self, Py_buffer *sep) |
1794 | /*[clinic end generated code: output=191b114cbb028e50 input=d78db010c8cfdbe1]*/ |
1795 | { |
1796 | return stringlib_rpartition( |
1797 | (PyObject*) self, |
1798 | PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), |
1799 | sep->obj, (const char *)sep->buf, sep->len |
1800 | ); |
1801 | } |
1802 | |
1803 | /*[clinic input] |
1804 | bytes.rsplit = bytes.split |
1805 | |
1806 | Return a list of the sections in the bytes, using sep as the delimiter. |
1807 | |
1808 | Splitting is done starting at the end of the bytes and working to the front. |
1809 | [clinic start generated code]*/ |
1810 | |
1811 | static PyObject * |
1812 | bytes_rsplit_impl(PyBytesObject *self, PyObject *sep, Py_ssize_t maxsplit) |
1813 | /*[clinic end generated code: output=ba698d9ea01e1c8f input=0f86c9f28f7d7b7b]*/ |
1814 | { |
1815 | Py_ssize_t len = PyBytes_GET_SIZE(self), n; |
1816 | const char *s = PyBytes_AS_STRING(self), *sub; |
1817 | Py_buffer vsub; |
1818 | PyObject *list; |
1819 | |
1820 | if (maxsplit < 0) Branch (1820:9): [True: 42, False: 45]
|
1821 | maxsplit = PY_SSIZE_T_MAX; |
1822 | if (sep == Py_None) Branch (1822:9): [True: 39, False: 48]
|
1823 | return stringlib_rsplit_whitespace((PyObject*) self, s, len, maxsplit); |
1824 | if (PyObject_GetBuffer(sep, &vsub, PyBUF_SIMPLE) != 0) Branch (1824:9): [True: 2, False: 46]
|
1825 | return NULL; |
1826 | sub = vsub.buf; |
1827 | n = vsub.len; |
1828 | |
1829 | list = stringlib_rsplit((PyObject*) self, s, len, sub, n, maxsplit); |
1830 | PyBuffer_Release(&vsub); |
1831 | return list; |
1832 | } |
1833 | |
1834 | |
1835 | /*[clinic input] |
1836 | bytes.join |
1837 | |
1838 | iterable_of_bytes: object |
1839 | / |
1840 | |
1841 | Concatenate any number of bytes objects. |
1842 | |
1843 | The bytes whose method is called is inserted in between each pair. |
1844 | |
1845 | The result is returned as a new bytes object. |
1846 | |
1847 | Example: b'.'.join([b'ab', b'pq', b'rs']) -> b'ab.pq.rs'. |
1848 | [clinic start generated code]*/ |
1849 | |
1850 | static PyObject * |
1851 | bytes_join(PyBytesObject *self, PyObject *iterable_of_bytes) |
1852 | /*[clinic end generated code: output=a046f379f626f6f8 input=7fe377b95bd549d2]*/ |
1853 | { |
1854 | return stringlib_bytes_join((PyObject*)self, iterable_of_bytes); |
1855 | } |
1856 | |
1857 | PyObject * |
1858 | _PyBytes_Join(PyObject *sep, PyObject *x) |
1859 | { |
1860 | assert(sep != NULL && PyBytes_Check(sep)); |
1861 | assert(x != NULL); |
1862 | return bytes_join((PyBytesObject*)sep, x); |
1863 | } |
1864 | |
1865 | static PyObject * |
1866 | bytes_find(PyBytesObject *self, PyObject *args) |
1867 | { |
1868 | return _Py_bytes_find(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
1869 | } |
1870 | |
1871 | static PyObject * |
1872 | bytes_index(PyBytesObject *self, PyObject *args) |
1873 | { |
1874 | return _Py_bytes_index(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
1875 | } |
1876 | |
1877 | |
1878 | static PyObject * |
1879 | bytes_rfind(PyBytesObject *self, PyObject *args) |
1880 | { |
1881 | return _Py_bytes_rfind(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
1882 | } |
1883 | |
1884 | |
1885 | static PyObject * |
1886 | bytes_rindex(PyBytesObject *self, PyObject *args) |
1887 | { |
1888 | return _Py_bytes_rindex(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
1889 | } |
1890 | |
1891 | |
1892 | Py_LOCAL_INLINE(PyObject *) |
1893 | do_xstrip(PyBytesObject *self, int striptype, PyObject *sepobj) |
1894 | { |
1895 | Py_buffer vsep; |
1896 | const char *s = PyBytes_AS_STRING(self); |
1897 | Py_ssize_t len = PyBytes_GET_SIZE(self); |
1898 | char *sep; |
1899 | Py_ssize_t seplen; |
1900 | Py_ssize_t i, j; |
1901 | |
1902 | if (PyObject_GetBuffer(sepobj, &vsep, PyBUF_SIMPLE) != 0) Branch (1902:9): [True: 6, False: 5.60k]
|
1903 | return NULL; |
1904 | sep = vsep.buf; |
1905 | seplen = vsep.len; |
1906 | |
1907 | i = 0; |
1908 | if (striptype != RIGHTSTRIP) { Branch (1908:9): [True: 685, False: 4.91k]
|
1909 | while (i < len && memchr(sep, 31.7k Py_CHARMASK31.7k (s[i]), seplen)) { Branch (1909:16): [True: 31.7k, False: 69]
Branch (1909:27): [True: 31.1k, False: 616]
|
1910 | i++; |
1911 | } |
1912 | } |
1913 | |
1914 | j = len; |
1915 | if (striptype != LEFTSTRIP) { Branch (1915:9): [True: 5.47k, False: 129]
|
1916 | do { |
1917 | j--; |
1918 | } while (j >= i && memchr(sep, 20.2k Py_CHARMASK20.2k (s[j]), seplen)); Branch (1918:18): [True: 20.2k, False: 958]
Branch (1918:28): [True: 15.7k, False: 4.51k]
|
1919 | j++; |
1920 | } |
1921 | |
1922 | PyBuffer_Release(&vsep); |
1923 | |
1924 | if (i == 0 && j == len5.39k && PyBytes_CheckExact598 (self)) { Branch (1924:9): [True: 5.39k, False: 211]
Branch (1924:19): [True: 598, False: 4.79k]
|
1925 | Py_INCREF(self); |
1926 | return (PyObject*)self; |
1927 | } |
1928 | else |
1929 | return PyBytes_FromStringAndSize(s+i, j-i); |
1930 | } |
1931 | |
1932 | |
1933 | Py_LOCAL_INLINE(PyObject *) |
1934 | do_strip(PyBytesObject *self, int striptype) |
1935 | { |
1936 | const char *s = PyBytes_AS_STRING(self); |
1937 | Py_ssize_t len = PyBytes_GET_SIZE(self), i, j; |
1938 | |
1939 | i = 0; |
1940 | if (striptype != RIGHTSTRIP) { Branch (1940:9): [True: 1.51k, False: 236]
|
1941 | while (i < len && Py_ISSPACE115k (s[i])) { Branch (1941:16): [True: 115k, False: 141]
|
1942 | i++; |
1943 | } |
1944 | } |
1945 | |
1946 | j = len; |
1947 | if (striptype != LEFTSTRIP) { Branch (1947:9): [True: 1.74k, False: 8]
|
1948 | do { |
1949 | j--; |
1950 | } while (j >= i && Py_ISSPACE25.2k (s[j])); Branch (1950:18): [True: 25.2k, False: 146]
|
1951 | j++; |
1952 | } |
1953 | |
1954 | if (i == 0 && j == len1.57k && PyBytes_CheckExact1.05k (self)) { Branch (1954:9): [True: 1.57k, False: 180]
Branch (1954:19): [True: 1.05k, False: 524]
|
1955 | Py_INCREF(self); |
1956 | return (PyObject*)self; |
1957 | } |
1958 | else |
1959 | return PyBytes_FromStringAndSize(s+i, j-i); |
1960 | } |
1961 | |
1962 | |
1963 | Py_LOCAL_INLINE(PyObject *) |
1964 | do_argstrip(PyBytesObject *self, int striptype, PyObject *bytes) |
1965 | { |
1966 | if (bytes != Py_None) { Branch (1966:9): [True: 5.61k, False: 1.75k]
|
1967 | return do_xstrip(self, striptype, bytes); |
1968 | } |
1969 | return do_strip(self, striptype); |
1970 | } |
1971 | |
1972 | /*[clinic input] |
1973 | bytes.strip |
1974 | |
1975 | bytes: object = None |
1976 | / |
1977 | |
1978 | Strip leading and trailing bytes contained in the argument. |
1979 | |
1980 | If the argument is omitted or None, strip leading and trailing ASCII whitespace. |
1981 | [clinic start generated code]*/ |
1982 | |
1983 | static PyObject * |
1984 | bytes_strip_impl(PyBytesObject *self, PyObject *bytes) |
1985 | /*[clinic end generated code: output=c7c228d3bd104a1b input=8a354640e4e0b3ef]*/ |
1986 | { |
1987 | return do_argstrip(self, BOTHSTRIP, bytes); |
1988 | } |
1989 | |
1990 | /*[clinic input] |
1991 | bytes.lstrip |
1992 | |
1993 | bytes: object = None |
1994 | / |
1995 | |
1996 | Strip leading bytes contained in the argument. |
1997 | |
1998 | If the argument is omitted or None, strip leading ASCII whitespace. |
1999 | [clinic start generated code]*/ |
2000 | |
2001 | static PyObject * |
2002 | bytes_lstrip_impl(PyBytesObject *self, PyObject *bytes) |
2003 | /*[clinic end generated code: output=28602e586f524e82 input=9baff4398c3f6857]*/ |
2004 | { |
2005 | return do_argstrip(self, LEFTSTRIP, bytes); |
2006 | } |
2007 | |
2008 | /*[clinic input] |
2009 | bytes.rstrip |
2010 | |
2011 | bytes: object = None |
2012 | / |
2013 | |
2014 | Strip trailing bytes contained in the argument. |
2015 | |
2016 | If the argument is omitted or None, strip trailing ASCII whitespace. |
2017 | [clinic start generated code]*/ |
2018 | |
2019 | static PyObject * |
2020 | bytes_rstrip_impl(PyBytesObject *self, PyObject *bytes) |
2021 | /*[clinic end generated code: output=547e3815c95447da input=b78af445c727e32b]*/ |
2022 | { |
2023 | return do_argstrip(self, RIGHTSTRIP, bytes); |
2024 | } |
2025 | |
2026 | |
2027 | static PyObject * |
2028 | bytes_count(PyBytesObject *self, PyObject *args) |
2029 | { |
2030 | return _Py_bytes_count(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
2031 | } |
2032 | |
2033 | |
2034 | /*[clinic input] |
2035 | bytes.translate |
2036 | |
2037 | table: object |
2038 | Translation table, which must be a bytes object of length 256. |
2039 | / |
2040 | delete as deletechars: object(c_default="NULL") = b'' |
2041 | |
2042 | Return a copy with each character mapped by the given translation table. |
2043 | |
2044 | All characters occurring in the optional argument delete are removed. |
2045 | The remaining characters are mapped through the given translation table. |
2046 | [clinic start generated code]*/ |
2047 | |
2048 | static PyObject * |
2049 | bytes_translate_impl(PyBytesObject *self, PyObject *table, |
2050 | PyObject *deletechars) |
2051 | /*[clinic end generated code: output=43be3437f1956211 input=0ecdf159f654233c]*/ |
2052 | { |
2053 | const char *input; |
2054 | char *output; |
2055 | Py_buffer table_view = {NULL, NULL}; |
2056 | Py_buffer del_table_view = {NULL, NULL}; |
2057 | const char *table_chars; |
2058 | Py_ssize_t i, c, changed = 0; |
2059 | PyObject *input_obj = (PyObject*)self; |
2060 | const char *output_start, *del_table_chars=NULL; |
2061 | Py_ssize_t inlen, tablen, dellen = 0; |
2062 | PyObject *result; |
2063 | int trans_table[256]; |
2064 | |
2065 | if (PyBytes_Check(table)) { |
2066 | table_chars = PyBytes_AS_STRING(table); |
2067 | tablen = PyBytes_GET_SIZE(table); |
2068 | } |
2069 | else if (table == Py_None) { Branch (2069:14): [True: 3, False: 6]
|
2070 | table_chars = NULL; |
2071 | tablen = 256; |
2072 | } |
2073 | else { |
2074 | if (PyObject_GetBuffer(table, &table_view, PyBUF_SIMPLE) != 0) Branch (2074:13): [True: 0, False: 6]
|
2075 | return NULL; |
2076 | table_chars = table_view.buf; |
2077 | tablen = table_view.len; |
2078 | } |
2079 | |
2080 | if (tablen != 256) { Branch (2080:9): [True: 1, False: 150]
|
2081 | PyErr_SetString(PyExc_ValueError, |
2082 | "translation table must be 256 characters long"); |
2083 | PyBuffer_Release(&table_view); |
2084 | return NULL; |
2085 | } |
2086 | |
2087 | if (deletechars != NULL) { Branch (2087:9): [True: 8, False: 142]
|
2088 | if (PyBytes_Check(deletechars)) { |
2089 | del_table_chars = PyBytes_AS_STRING(deletechars); |
2090 | dellen = PyBytes_GET_SIZE(deletechars); |
2091 | } |
2092 | else { |
2093 | if (PyObject_GetBuffer(deletechars, &del_table_view, PyBUF_SIMPLE) != 0) { Branch (2093:17): [True: 1, False: 0]
|
2094 | PyBuffer_Release(&table_view); |
2095 | return NULL; |
2096 | } |
2097 | del_table_chars = del_table_view.buf; |
2098 | dellen = del_table_view.len; |
2099 | } |
2100 | } |
2101 | else { |
2102 | del_table_chars = NULL; |
2103 | dellen = 0; |
2104 | } |
2105 | |
2106 | inlen = PyBytes_GET_SIZE(input_obj); |
2107 | result = PyBytes_FromStringAndSize((char *)NULL, inlen); |
2108 | if (result == NULL) { Branch (2108:9): [True: 0, False: 149]
|
2109 | PyBuffer_Release(&del_table_view); |
2110 | PyBuffer_Release(&table_view); |
2111 | return NULL; |
2112 | } |
2113 | output_start = output = PyBytes_AS_STRING(result); |
2114 | input = PyBytes_AS_STRING(input_obj); |
2115 | |
2116 | if (dellen == 0 && table_chars != NULL144 ) { Branch (2116:9): [True: 144, False: 5]
Branch (2116:24): [True: 144, False: 0]
|
2117 | /* If no deletions are required, use faster code */ |
2118 | for (i = inlen; --i >= 0; ) { Branch (2118:25): [True: 12.6k, False: 144]
|
2119 | c = Py_CHARMASK(*input++); |
2120 | if (Py_CHARMASK((*output++ = table_chars[c])) != c) Branch (2120:17): [True: 10.0k, False: 2.57k]
|
2121 | changed = 1; |
2122 | } |
2123 | if (!changed && PyBytes_CheckExact25 (input_obj)) { Branch (2123:13): [True: 25, False: 119]
|
2124 | Py_INCREF(input_obj); |
2125 | Py_DECREF(result); |
2126 | result = input_obj; |
2127 | } |
2128 | PyBuffer_Release(&del_table_view); |
2129 | PyBuffer_Release(&table_view); |
2130 | return result; |
2131 | } |
2132 | |
2133 | if (table_chars == NULL) { Branch (2133:9): [True: 2, False: 3]
|
2134 | for (i = 0; i < 256; i++512 ) Branch (2134:21): [True: 512, False: 2]
|
2135 | trans_table[i] = Py_CHARMASK(i); |
2136 | } else { |
2137 | for (i = 0; i < 256; i++768 ) Branch (2137:21): [True: 768, False: 3]
|
2138 | trans_table[i] = Py_CHARMASK(table_chars[i]); |
2139 | } |
2140 | PyBuffer_Release(&table_view); |
2141 | |
2142 | for (i = 0; i < dellen; i++9 ) Branch (2142:17): [True: 9, False: 5]
|
2143 | trans_table[(int) Py_CHARMASK(del_table_chars[i])] = -1; |
2144 | PyBuffer_Release(&del_table_view); |
2145 | |
2146 | for (i = inlen; --i >= 0; ) { Branch (2146:21): [True: 25, False: 5]
|
2147 | c = Py_CHARMASK(*input++); |
2148 | if (trans_table[c] != -1) Branch (2148:13): [True: 14, False: 11]
|
2149 | if (Py_CHARMASK(*output++ = (char)trans_table[c]) == c) Branch (2149:17): [True: 12, False: 2]
|
2150 | continue; |
2151 | changed = 1; |
2152 | } |
2153 | if (!changed && PyBytes_CheckExact0 (input_obj)) { Branch (2153:9): [True: 0, False: 5]
|
2154 | Py_DECREF(result); |
2155 | Py_INCREF(input_obj); |
2156 | return input_obj; |
2157 | } |
2158 | /* Fix the size of the resulting byte string */ |
2159 | if (inlen > 0) Branch (2159:9): [True: 5, False: 0]
|
2160 | _PyBytes_Resize(&result, output - output_start); |
2161 | return result; |
2162 | } |
2163 | |
2164 | |
2165 | /*[clinic input] |
2166 | |
2167 | @staticmethod |
2168 | bytes.maketrans |
2169 | |
2170 | frm: Py_buffer |
2171 | to: Py_buffer |
2172 | / |
2173 | |
2174 | Return a translation table useable for the bytes or bytearray translate method. |
2175 | |
2176 | The returned table will be one where each byte in frm is mapped to the byte at |
2177 | the same position in to. |
2178 | |
2179 | The bytes objects frm and to must be of the same length. |
2180 | [clinic start generated code]*/ |
2181 | |
2182 | static PyObject * |
2183 | bytes_maketrans_impl(Py_buffer *frm, Py_buffer *to) |
2184 | /*[clinic end generated code: output=a36f6399d4b77f6f input=de7a8fc5632bb8f1]*/ |
2185 | { |
2186 | return _Py_bytes_maketrans(frm, to); |
2187 | } |
2188 | |
2189 | |
2190 | /*[clinic input] |
2191 | bytes.replace |
2192 | |
2193 | old: Py_buffer |
2194 | new: Py_buffer |
2195 | count: Py_ssize_t = -1 |
2196 | Maximum number of occurrences to replace. |
2197 | -1 (the default value) means replace all occurrences. |
2198 | / |
2199 | |
2200 | Return a copy with all occurrences of substring old replaced by new. |
2201 | |
2202 | If the optional argument count is given, only the first count occurrences are |
2203 | replaced. |
2204 | [clinic start generated code]*/ |
2205 | |
2206 | static PyObject * |
2207 | bytes_replace_impl(PyBytesObject *self, Py_buffer *old, Py_buffer *new, |
2208 | Py_ssize_t count) |
2209 | /*[clinic end generated code: output=994fa588b6b9c104 input=b2fbbf0bf04de8e5]*/ |
2210 | { |
2211 | return stringlib_replace((PyObject *)self, |
2212 | (const char *)old->buf, old->len, |
2213 | (const char *)new->buf, new->len, count); |
2214 | } |
2215 | |
2216 | /** End DALKE **/ |
2217 | |
2218 | /*[clinic input] |
2219 | bytes.removeprefix as bytes_removeprefix |
2220 | |
2221 | prefix: Py_buffer |
2222 | / |
2223 | |
2224 | Return a bytes object with the given prefix string removed if present. |
2225 | |
2226 | If the bytes starts with the prefix string, return bytes[len(prefix):]. |
2227 | Otherwise, return a copy of the original bytes. |
2228 | [clinic start generated code]*/ |
2229 | |
2230 | static PyObject * |
2231 | bytes_removeprefix_impl(PyBytesObject *self, Py_buffer *prefix) |
2232 | /*[clinic end generated code: output=f006865331a06ab6 input=0c93bac817a8502c]*/ |
2233 | { |
2234 | const char *self_start = PyBytes_AS_STRING(self); |
2235 | Py_ssize_t self_len = PyBytes_GET_SIZE(self); |
2236 | const char *prefix_start = prefix->buf; |
2237 | Py_ssize_t prefix_len = prefix->len; |
2238 | |
2239 | if (self_len >= prefix_len Branch (2239:9): [True: 19, False: 8]
|
2240 | && prefix_len > 019 Branch (2240:12): [True: 15, False: 4]
|
2241 | && memcmp(self_start, prefix_start, prefix_len) == 015 ) Branch (2241:12): [True: 15, False: 0]
|
2242 | { |
2243 | return PyBytes_FromStringAndSize(self_start + prefix_len, |
2244 | self_len - prefix_len); |
2245 | } |
2246 | |
2247 | if (PyBytes_CheckExact(self)) { |
2248 | Py_INCREF(self); |
2249 | return (PyObject *)self; |
2250 | } |
2251 | |
2252 | return PyBytes_FromStringAndSize(self_start, self_len); |
2253 | } |
2254 | |
2255 | /*[clinic input] |
2256 | bytes.removesuffix as bytes_removesuffix |
2257 | |
2258 | suffix: Py_buffer |
2259 | / |
2260 | |
2261 | Return a bytes object with the given suffix string removed if present. |
2262 | |
2263 | If the bytes ends with the suffix string and that suffix is not empty, |
2264 | return bytes[:-len(prefix)]. Otherwise, return a copy of the original |
2265 | bytes. |
2266 | [clinic start generated code]*/ |
2267 | |
2268 | static PyObject * |
2269 | bytes_removesuffix_impl(PyBytesObject *self, Py_buffer *suffix) |
2270 | /*[clinic end generated code: output=d887d308e3242eeb input=9f4e1da8c637bbf1]*/ |
2271 | { |
2272 | const char *self_start = PyBytes_AS_STRING(self); |
2273 | Py_ssize_t self_len = PyBytes_GET_SIZE(self); |
2274 | const char *suffix_start = suffix->buf; |
2275 | Py_ssize_t suffix_len = suffix->len; |
2276 | |
2277 | if (self_len >= suffix_len Branch (2277:9): [True: 9, False: 6]
|
2278 | && suffix_len > 09 Branch (2278:12): [True: 5, False: 4]
|
2279 | && memcmp(self_start + self_len - suffix_len, Branch (2279:12): [True: 3, False: 2]
|
2280 | suffix_start, suffix_len) == 0) |
2281 | { |
2282 | return PyBytes_FromStringAndSize(self_start, |
2283 | self_len - suffix_len); |
2284 | } |
2285 | |
2286 | if (PyBytes_CheckExact(self)) { |
2287 | Py_INCREF(self); |
2288 | return (PyObject *)self; |
2289 | } |
2290 | |
2291 | return PyBytes_FromStringAndSize(self_start, self_len); |
2292 | } |
2293 | |
2294 | static PyObject * |
2295 | bytes_startswith(PyBytesObject *self, PyObject *args) |
2296 | { |
2297 | return _Py_bytes_startswith(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
2298 | } |
2299 | |
2300 | static PyObject * |
2301 | bytes_endswith(PyBytesObject *self, PyObject *args) |
2302 | { |
2303 | return _Py_bytes_endswith(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), args); |
2304 | } |
2305 | |
2306 | |
2307 | /*[clinic input] |
2308 | bytes.decode |
2309 | |
2310 | encoding: str(c_default="NULL") = 'utf-8' |
2311 | The encoding with which to decode the bytes. |
2312 | errors: str(c_default="NULL") = 'strict' |
2313 | The error handling scheme to use for the handling of decoding errors. |
2314 | The default is 'strict' meaning that decoding errors raise a |
2315 | UnicodeDecodeError. Other possible values are 'ignore' and 'replace' |
2316 | as well as any other name registered with codecs.register_error that |
2317 | can handle UnicodeDecodeErrors. |
2318 | |
2319 | Decode the bytes using the codec registered for encoding. |
2320 | [clinic start generated code]*/ |
2321 | |
2322 | static PyObject * |
2323 | bytes_decode_impl(PyBytesObject *self, const char *encoding, |
2324 | const char *errors) |
2325 | /*[clinic end generated code: output=5649a53dde27b314 input=958174769d2a40ca]*/ |
2326 | { |
2327 | return PyUnicode_FromEncodedObject((PyObject*)self, encoding, errors); |
2328 | } |
2329 | |
2330 | |
2331 | /*[clinic input] |
2332 | bytes.splitlines |
2333 | |
2334 | keepends: bool(accept={int}) = False |
2335 | |
2336 | Return a list of the lines in the bytes, breaking at line boundaries. |
2337 | |
2338 | Line breaks are not included in the resulting list unless keepends is given and |
2339 | true. |
2340 | [clinic start generated code]*/ |
2341 | |
2342 | static PyObject * |
2343 | bytes_splitlines_impl(PyBytesObject *self, int keepends) |
2344 | /*[clinic end generated code: output=3484149a5d880ffb input=a8b32eb01ff5a5ed]*/ |
2345 | { |
2346 | return stringlib_splitlines( |
2347 | (PyObject*) self, PyBytes_AS_STRING(self), |
2348 | PyBytes_GET_SIZE(self), keepends |
2349 | ); |
2350 | } |
2351 | |
2352 | /*[clinic input] |
2353 | @classmethod |
2354 | bytes.fromhex |
2355 | |
2356 | string: unicode |
2357 | / |
2358 | |
2359 | Create a bytes object from a string of hexadecimal numbers. |
2360 | |
2361 | Spaces between two numbers are accepted. |
2362 | Example: bytes.fromhex('B9 01EF') -> b'\\xb9\\x01\\xef'. |
2363 | [clinic start generated code]*/ |
2364 | |
2365 | static PyObject * |
2366 | bytes_fromhex_impl(PyTypeObject *type, PyObject *string) |
2367 | /*[clinic end generated code: output=0973acc63661bb2e input=bf4d1c361670acd3]*/ |
2368 | { |
2369 | PyObject *result = _PyBytes_FromHex(string, 0); |
2370 | if (type != &PyBytes_Type && result != NULL3 ) { Branch (2370:9): [True: 3, False: 1.49k]
Branch (2370:34): [True: 3, False: 0]
|
2371 | Py_SETREF(result, PyObject_CallOneArg((PyObject *)type, result)); |
2372 | } |
2373 | return result; |
2374 | } |
2375 | |
2376 | PyObject* |
2377 | _PyBytes_FromHex(PyObject *string, int use_bytearray) |
2378 | { |
2379 | char *buf; |
2380 | Py_ssize_t hexlen, invalid_char; |
2381 | unsigned int top, bot; |
2382 | const Py_UCS1 *str, *end; |
2383 | _PyBytesWriter writer; |
2384 | |
2385 | _PyBytesWriter_Init(&writer); |
2386 | writer.use_bytearray = use_bytearray; |
2387 | |
2388 | assert(PyUnicode_Check(string)); |
2389 | if (PyUnicode_READY(string)) |
2390 | return NULL; |
2391 | hexlen = PyUnicode_GET_LENGTH(string); |
2392 | |
2393 | if (!PyUnicode_IS_ASCII(string)) { Branch (2393:9): [True: 12, False: 1.52k]
|
2394 | const void *data = PyUnicode_DATA(string); |
2395 | int kind = PyUnicode_KIND(string); |
2396 | Py_ssize_t i; |
2397 | |
2398 | /* search for the first non-ASCII character */ |
2399 | for (i = 0; i < hexlen; i++8 ) { Branch (2399:21): [True: 20, False: 0]
|
2400 | if (PyUnicode_READ(kind, data, i) >= 128) Branch (2400:17): [True: 12, False: 8]
|
2401 | break; |
2402 | } |
2403 | invalid_char = i; |
2404 | goto error; |
2405 | } |
2406 | |
2407 | assert(PyUnicode_KIND(string) == PyUnicode_1BYTE_KIND); |
2408 | str = PyUnicode_1BYTE_DATA(string); |
2409 | |
2410 | /* This overestimates if there are spaces */ |
2411 | buf = _PyBytesWriter_Alloc(&writer, hexlen / 2); |
2412 | if (buf == NULL) Branch (2412:9): [True: 0, False: 1.52k]
|
2413 | return NULL; |
2414 | |
2415 | end = str + hexlen; |
2416 | while (str < end) { Branch (2416:12): [True: 10.6k, False: 1.46k]
|
2417 | /* skip over spaces in the input */ |
2418 | if (Py_ISSPACE(*str)) { |
2419 | do { |
2420 | str++; |
2421 | } while (Py_ISSPACE(*str)); |
2422 | if (str >= end) Branch (2422:17): [True: 35, False: 1.39k]
|
2423 | break; |
2424 | } |
2425 | |
2426 | top = _PyLong_DigitValue[*str]; |
2427 | if (top >= 16) { Branch (2427:13): [True: 18, False: 10.6k]
|
2428 | invalid_char = str - PyUnicode_1BYTE_DATA(string); |
2429 | goto error; |
2430 | } |
2431 | str++; |
2432 | |
2433 | bot = _PyLong_DigitValue[*str]; |
2434 | if (bot >= 16) { Branch (2434:13): [True: 6, False: 10.6k]
|
2435 | invalid_char = str - PyUnicode_1BYTE_DATA(string); |
2436 | goto error; |
2437 | } |
2438 | str++; |
2439 | |
2440 | *buf++ = (unsigned char)((top << 4) + bot); |
2441 | } |
2442 | |
2443 | return _PyBytesWriter_Finish(&writer, buf); |
2444 | |
2445 | error: |
2446 | PyErr_Format(PyExc_ValueError, |
2447 | "non-hexadecimal number found in " |
2448 | "fromhex() arg at position %zd", invalid_char); |
2449 | _PyBytesWriter_Dealloc(&writer); |
2450 | return NULL; |
2451 | } |
2452 | |
2453 | /*[clinic input] |
2454 | bytes.hex |
2455 | |
2456 | sep: object = NULL |
2457 | An optional single character or byte to separate hex bytes. |
2458 | bytes_per_sep: int = 1 |
2459 | How many bytes between separators. Positive values count from the |
2460 | right, negative values count from the left. |
2461 | |
2462 | Create a string of hexadecimal numbers from a bytes object. |
2463 | |
2464 | Example: |
2465 | >>> value = b'\xb9\x01\xef' |
2466 | >>> value.hex() |
2467 | 'b901ef' |
2468 | >>> value.hex(':') |
2469 | 'b9:01:ef' |
2470 | >>> value.hex(':', 2) |
2471 | 'b9:01ef' |
2472 | >>> value.hex(':', -2) |
2473 | 'b901:ef' |
2474 | [clinic start generated code]*/ |
2475 | |
2476 | static PyObject * |
2477 | bytes_hex_impl(PyBytesObject *self, PyObject *sep, int bytes_per_sep) |
2478 | /*[clinic end generated code: output=1f134da504064139 input=1a21282b1f1ae595]*/ |
2479 | { |
2480 | const char *argbuf = PyBytes_AS_STRING(self); |
2481 | Py_ssize_t arglen = PyBytes_GET_SIZE(self); |
2482 | return _Py_strhex_with_sep(argbuf, arglen, sep, bytes_per_sep); |
2483 | } |
2484 | |
2485 | static PyObject * |
2486 | bytes_getnewargs(PyBytesObject *v, PyObject *Py_UNUSED(ignored)) |
2487 | { |
2488 | return Py_BuildValue("(y#)", v->ob_sval, Py_SIZE(v)); |
2489 | } |
2490 | |
2491 | |
2492 | static PyMethodDef |
2493 | bytes_methods[] = { |
2494 | {"__getnewargs__", (PyCFunction)bytes_getnewargs, METH_NOARGS}, |
2495 | BYTES___BYTES___METHODDEF |
2496 | {"capitalize", stringlib_capitalize, METH_NOARGS, |
2497 | _Py_capitalize__doc__}, |
2498 | STRINGLIB_CENTER_METHODDEF |
2499 | {"count", (PyCFunction)bytes_count, METH_VARARGS, |
2500 | _Py_count__doc__}, |
2501 | BYTES_DECODE_METHODDEF |
2502 | {"endswith", (PyCFunction)bytes_endswith, METH_VARARGS, |
2503 | _Py_endswith__doc__}, |
2504 | STRINGLIB_EXPANDTABS_METHODDEF |
2505 | {"find", (PyCFunction)bytes_find, METH_VARARGS, |
2506 | _Py_find__doc__}, |
2507 | BYTES_FROMHEX_METHODDEF |
2508 | BYTES_HEX_METHODDEF |
2509 | {"index", (PyCFunction)bytes_index, METH_VARARGS, _Py_index__doc__}, |
2510 | {"isalnum", stringlib_isalnum, METH_NOARGS, |
2511 | _Py_isalnum__doc__}, |
2512 | {"isalpha", stringlib_isalpha, METH_NOARGS, |
2513 | _Py_isalpha__doc__}, |
2514 | {"isascii", stringlib_isascii, METH_NOARGS, |
2515 | _Py_isascii__doc__}, |
2516 | {"isdigit", stringlib_isdigit, METH_NOARGS, |
2517 | _Py_isdigit__doc__}, |
2518 | {"islower", stringlib_islower, METH_NOARGS, |
2519 | _Py_islower__doc__}, |
2520 | {"isspace", stringlib_isspace, METH_NOARGS, |
2521 | _Py_isspace__doc__}, |
2522 | {"istitle", stringlib_istitle, METH_NOARGS, |
2523 | _Py_istitle__doc__}, |
2524 | {"isupper", stringlib_isupper, METH_NOARGS, |
2525 | _Py_isupper__doc__}, |
2526 | BYTES_JOIN_METHODDEF |
2527 | STRINGLIB_LJUST_METHODDEF |
2528 | {"lower", stringlib_lower, METH_NOARGS, _Py_lower__doc__}, |
2529 | BYTES_LSTRIP_METHODDEF |
2530 | BYTES_MAKETRANS_METHODDEF |
2531 | BYTES_PARTITION_METHODDEF |
2532 | BYTES_REPLACE_METHODDEF |
2533 | BYTES_REMOVEPREFIX_METHODDEF |
2534 | BYTES_REMOVESUFFIX_METHODDEF |
2535 | {"rfind", (PyCFunction)bytes_rfind, METH_VARARGS, _Py_rfind__doc__}, |
2536 | {"rindex", (PyCFunction)bytes_rindex, METH_VARARGS, _Py_rindex__doc__}, |
2537 | STRINGLIB_RJUST_METHODDEF |
2538 | BYTES_RPARTITION_METHODDEF |
2539 | BYTES_RSPLIT_METHODDEF |
2540 | BYTES_RSTRIP_METHODDEF |
2541 | BYTES_SPLIT_METHODDEF |
2542 | BYTES_SPLITLINES_METHODDEF |
2543 | {"startswith", (PyCFunction)bytes_startswith, METH_VARARGS, |
2544 | _Py_startswith__doc__}, |
2545 | BYTES_STRIP_METHODDEF |
2546 | {"swapcase", stringlib_swapcase, METH_NOARGS, |
2547 | _Py_swapcase__doc__}, |
2548 | {"title", stringlib_title, METH_NOARGS, _Py_title__doc__}, |
2549 | BYTES_TRANSLATE_METHODDEF |
2550 | {"upper", stringlib_upper, METH_NOARGS, _Py_upper__doc__}, |
2551 | STRINGLIB_ZFILL_METHODDEF |
2552 | {NULL, NULL} /* sentinel */ |
2553 | }; |
2554 | |
2555 | static PyObject * |
2556 | bytes_mod(PyObject *self, PyObject *arg) |
2557 | { |
2558 | if (!PyBytes_Check(self)) { Branch (2558:9): [True: 2, False: 776]
|
2559 | Py_RETURN_NOTIMPLEMENTED; |
2560 | } |
2561 | return _PyBytes_FormatEx(PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self), |
2562 | arg, 0); |
2563 | } |
2564 | |
2565 | static PyNumberMethods bytes_as_number = { |
2566 | 0, /*nb_add*/ |
2567 | 0, /*nb_subtract*/ |
2568 | 0, /*nb_multiply*/ |
2569 | bytes_mod, /*nb_remainder*/ |
2570 | }; |
2571 | |
2572 | static PyObject * |
2573 | bytes_subtype_new(PyTypeObject *, PyObject *); |
2574 | |
2575 | /*[clinic input] |
2576 | @classmethod |
2577 | bytes.__new__ as bytes_new |
2578 | |
2579 | source as x: object = NULL |
2580 | encoding: str = NULL |
2581 | errors: str = NULL |
2582 | |
2583 | [clinic start generated code]*/ |
2584 | |
2585 | static PyObject * |
2586 | bytes_new_impl(PyTypeObject *type, PyObject *x, const char *encoding, |
2587 | const char *errors) |
2588 | /*[clinic end generated code: output=1e0c471be311a425 input=f0a966d19b7262b4]*/ |
2589 | { |
2590 | PyObject *bytes; |
2591 | PyObject *func; |
2592 | Py_ssize_t size; |
2593 | |
2594 | if (x == NULL) { Branch (2594:9): [True: 121, False: 478k]
|
2595 | if (encoding != NULL || errors != NULL120 ) { Branch (2595:13): [True: 1, False: 120]
Branch (2595:33): [True: 1, False: 119]
|
2596 | PyErr_SetString(PyExc_TypeError, |
2597 | encoding != NULL ? Branch (2597:29): [True: 1, False: 1]
|
2598 | "encoding without a string argument" : |
2599 | "errors without a string argument"1 ); |
2600 | return NULL; |
2601 | } |
2602 | bytes = PyBytes_FromStringAndSize(NULL, 0); |
2603 | } |
2604 | else if (encoding != NULL) { Branch (2604:14): [True: 36.6k, False: 442k]
|
2605 | /* Encode via the codec registry */ |
2606 | if (!PyUnicode_Check(x)) { Branch (2606:13): [True: 2, False: 36.6k]
|
2607 | PyErr_SetString(PyExc_TypeError, |
2608 | "encoding without a string argument"); |
2609 | return NULL; |
2610 | } |
2611 | bytes = PyUnicode_AsEncodedString(x, encoding, errors); |
2612 | } |
2613 | else if (errors != NULL) { Branch (2613:14): [True: 3, False: 442k]
|
2614 | PyErr_SetString(PyExc_TypeError, |
2615 | PyUnicode_Check(x) ? |
2616 | "string argument without an encoding" : |
2617 | "errors without a string argument"2 ); |
2618 | return NULL; |
2619 | } |
2620 | /* We'd like to call PyObject_Bytes here, but we need to check for an |
2621 | integer argument before deferring to PyBytes_FromObject, something |
2622 | PyObject_Bytes doesn't do. */ |
2623 | else if ((func = _PyObject_LookupSpecial(x, &_Py_ID(__bytes__))) != NULL) { Branch (2623:14): [True: 105k, False: 336k]
|
2624 | bytes = _PyObject_CallNoArgs(func); |
2625 | Py_DECREF(func); |
2626 | if (bytes == NULL) Branch (2626:13): [True: 4, False: 105k]
|
2627 | return NULL; |
2628 | if (!PyBytes_Check(bytes)) { Branch (2628:13): [True: 2, False: 105k]
|
2629 | PyErr_Format(PyExc_TypeError, |
2630 | "__bytes__ returned non-bytes (type %.200s)", |
2631 | Py_TYPE(bytes)->tp_name); |
2632 | Py_DECREF(bytes); |
2633 | return NULL; |
2634 | } |
2635 | } |
2636 | else if (PyErr_Occurred()) Branch (2636:14): [True: 1, False: 336k]
|
2637 | return NULL; |
2638 | else if (PyUnicode_Check(x)) { |
2639 | PyErr_SetString(PyExc_TypeError, |
2640 | "string argument without an encoding"); |
2641 | return NULL; |
2642 | } |
2643 | /* Is it an integer? */ |
2644 | else if (_PyIndex_Check(x)) { Branch (2644:14): [True: 133, False: 336k]
|
2645 | size = PyNumber_AsSsize_t(x, PyExc_OverflowError); |
2646 | if (size == -1 && PyErr_Occurred()3 ) { Branch (2646:13): [True: 3, False: 130]
Branch (2646:27): [True: 2, False: 1]
|
2647 | if (!PyErr_ExceptionMatches(PyExc_TypeError)) Branch (2647:17): [True: 2, False: 0]
|
2648 | return NULL; |
2649 | PyErr_Clear(); /* fall through */ |
2650 | bytes = PyBytes_FromObject(x); |
2651 | } |
2652 | else { |
2653 | if (size < 0) { Branch (2653:17): [True: 1, False: 130]
|
2654 | PyErr_SetString(PyExc_ValueError, "negative count"); |
2655 | return NULL; |
2656 | } |
2657 | bytes = _PyBytes_FromSize(size, 1); |
2658 | } |
2659 | } |
2660 | else { |
2661 | bytes = PyBytes_FromObject(x); |
2662 | } |
2663 | |
2664 | if (bytes != NULL && type != &PyBytes_Type478k ) { Branch (2664:9): [True: 478k, False: 57]
Branch (2664:26): [True: 858, False: 477k]
|
2665 | Py_SETREF(bytes, bytes_subtype_new(type, bytes)); |
2666 | } |
2667 | |
2668 | return bytes; |
2669 | } |
2670 | |
2671 | static PyObject* |
2672 | _PyBytes_FromBuffer(PyObject *x) |
2673 | { |
2674 | PyObject *new; |
2675 | Py_buffer view; |
2676 | |
2677 | if (PyObject_GetBuffer(x, &view, PyBUF_FULL_RO) < 0) Branch (2677:9): [True: 27, False: 157k]
|
2678 | return NULL; |
2679 | |
2680 | new = PyBytes_FromStringAndSize(NULL, view.len); |
2681 | if (!new) Branch (2681:9): [True: 0, False: 157k]
|
2682 | goto fail; |
2683 | if (PyBuffer_ToContiguous(((PyBytesObject *)new)->ob_sval, Branch (2683:9): [True: 0, False: 157k]
|
2684 | &view, view.len, 'C') < 0) |
2685 | goto fail; |
2686 | PyBuffer_Release(&view); |
2687 | return new; |
2688 | |
2689 | fail: |
2690 | Py_XDECREF(new); |
2691 | PyBuffer_Release(&view); |
2692 | return NULL; |
2693 | } |
2694 | |
2695 | static PyObject* |
2696 | _PyBytes_FromList(PyObject *x) |
2697 | { |
2698 | Py_ssize_t i, size = PyList_GET_SIZE(x); |
2699 | Py_ssize_t value; |
2700 | char *str; |
2701 | PyObject *item; |
2702 | _PyBytesWriter writer; |
2703 | |
2704 | _PyBytesWriter_Init(&writer); |
2705 | str = _PyBytesWriter_Alloc(&writer, size); |
2706 | if (str == NULL) Branch (2706:9): [True: 0, False: 173k]
|
2707 | return NULL; |
2708 | writer.overallocate = 1; |
2709 | size = writer.allocated; |
2710 | |
2711 | for (i = 0; i < PyList_GET_SIZE(x); i++542k ) { Branch (2711:17): [True: 542k, False: 173k]
|
2712 | item = PyList_GET_ITEM(x, i); |
2713 | Py_INCREF(item); |
2714 | value = PyNumber_AsSsize_t(item, NULL); |
2715 | Py_DECREF(item); |
2716 | if (value == -1 && PyErr_Occurred()7 ) Branch (2716:13): [True: 7, False: 542k]
Branch (2716:28): [True: 5, False: 2]
|
2717 | goto error; |
2718 | |
2719 | if (value < 0 || value >= 256542k ) { Branch (2719:13): [True: 6, False: 542k]
Branch (2719:26): [True: 7, False: 542k]
|
2720 | PyErr_SetString(PyExc_ValueError, |
2721 | "bytes must be in range(0, 256)"); |
2722 | goto error; |
2723 | } |
2724 | |
2725 | if (i >= size) { Branch (2725:13): [True: 6, False: 542k]
|
2726 | str = _PyBytesWriter_Resize(&writer, str, size+1); |
2727 | if (str == NULL) Branch (2727:17): [True: 0, False: 6]
|
2728 | return NULL; |
2729 | size = writer.allocated; |
2730 | } |
2731 | *str++ = (char) value; |
2732 | } |
2733 | return _PyBytesWriter_Finish(&writer, str); |
2734 | |
2735 | error: |
2736 | _PyBytesWriter_Dealloc(&writer); |
2737 | return NULL; |
2738 | } |
2739 | |
2740 | static PyObject* |
2741 | _PyBytes_FromTuple(PyObject *x) |
2742 | { |
2743 | PyObject *bytes; |
2744 | Py_ssize_t i, size = PyTuple_GET_SIZE(x); |
2745 | Py_ssize_t value; |
2746 | char *str; |
2747 | PyObject *item; |
2748 | |
2749 | bytes = PyBytes_FromStringAndSize(NULL, size); |
2750 | if (bytes == NULL) Branch (2750:9): [True: 0, False: 4.43k]
|
2751 | return NULL; |
2752 | str = ((PyBytesObject *)bytes)->ob_sval; |
2753 | |
2754 | for (i = 0; i < size; i++5.02k ) { Branch (2754:17): [True: 5.02k, False: 4.43k]
|
2755 | item = PyTuple_GET_ITEM(x, i); |
2756 | value = PyNumber_AsSsize_t(item, NULL); |
2757 | if (value == -1 && PyErr_Occurred()0 ) Branch (2757:13): [True: 0, False: 5.02k]
Branch (2757:28): [True: 0, False: 0]
|
2758 | goto error; |
2759 | |
2760 | if (value < 0 || value >= 256) { Branch (2760:13): [True: 0, False: 5.02k]
Branch (2760:26): [True: 0, False: 5.02k]
|
2761 | PyErr_SetString(PyExc_ValueError, |
2762 | "bytes must be in range(0, 256)"); |
2763 | goto error; |
2764 | } |
2765 | *str++ = (char) value; |
2766 | } |
2767 | return bytes; |
2768 | |
2769 | error: |
2770 | Py_DECREF(bytes); |
2771 | return NULL; |
2772 | } |
2773 | |
2774 | static PyObject * |
2775 | _PyBytes_FromIterator(PyObject *it, PyObject *x) |
2776 | { |
2777 | char *str; |
2778 | Py_ssize_t i, size; |
2779 | _PyBytesWriter writer; |
2780 | |
2781 | /* For iterator version, create a bytes object and resize as needed */ |
2782 | size = PyObject_LengthHint(x, 64); |
2783 | if (size == -1 && PyErr_Occurred()0 ) Branch (2783:9): [True: 0, False: 2.85k]
Branch (2783:23): [True: 0, False: 0]
|
2784 | return NULL; |
2785 | |
2786 | _PyBytesWriter_Init(&writer); |
2787 | str = _PyBytesWriter_Alloc(&writer, size); |
2788 | if (str == NULL) Branch (2788:9): [True: 0, False: 2.85k]
|
2789 | return NULL; |
2790 | writer.overallocate = 1; |
2791 | size = writer.allocated; |
2792 | |
2793 | /* Run the iterator to exhaustion */ |
2794 | for (i = 0; ; i++6.05M ) { |
2795 | PyObject *item; |
2796 | Py_ssize_t value; |
2797 | |
2798 | /* Get the next item */ |
2799 | item = PyIter_Next(it); |
2800 | if (item == NULL) { Branch (2800:13): [True: 2.85k, False: 6.05M]
|
2801 | if (PyErr_Occurred()) Branch (2801:17): [True: 64, False: 2.79k]
|
2802 | goto error; |
2803 | break; |
2804 | } |
2805 | |
2806 | /* Interpret it as an int (__index__) */ |
2807 | value = PyNumber_AsSsize_t(item, NULL); |
2808 | Py_DECREF(item); |
2809 | if (value == -1 && PyErr_Occurred()2 ) Branch (2809:13): [True: 2, False: 6.05M]
Branch (2809:28): [True: 2, False: 0]
|
2810 | goto error; |
2811 | |
2812 | /* Range check */ |
2813 | if (value < 0 || value >= 256) { Branch (2813:13): [True: 0, False: 6.05M]
Branch (2813:26): [True: 0, False: 6.05M]
|
2814 | PyErr_SetString(PyExc_ValueError, |
2815 | "bytes must be in range(0, 256)"); |
2816 | goto error; |
2817 | } |
2818 | |
2819 | /* Append the byte */ |
2820 | if (i >= size) { Branch (2820:13): [True: 1.53k, False: 6.05M]
|
2821 | str = _PyBytesWriter_Resize(&writer, str, size+1); |
2822 | if (str == NULL) Branch (2822:17): [True: 0, False: 1.53k]
|
2823 | return NULL; |
2824 | size = writer.allocated; |
2825 | } |
2826 | *str++ = (char) value; |
2827 | } |
2828 | |
2829 | return _PyBytesWriter_Finish(&writer, str); |
2830 | |
2831 | error: |
2832 | _PyBytesWriter_Dealloc(&writer); |
2833 | return NULL; |
2834 | } |
2835 | |
2836 | PyObject * |
2837 | PyBytes_FromObject(PyObject *x) |
2838 | { |
2839 | PyObject *it, *result; |
2840 | |
2841 | if (x == NULL) { Branch (2841:9): [True: 0, False: 338k]
|
2842 | PyErr_BadInternalCall(); |
2843 | return NULL; |
2844 | } |
2845 | |
2846 | if (PyBytes_CheckExact(x)) { |
2847 | Py_INCREF(x); |
2848 | return x; |
2849 | } |
2850 | |
2851 | /* Use the modern buffer interface */ |
2852 | if (PyObject_CheckBuffer(x)) Branch (2852:9): [True: 157k, False: 181k]
|
2853 | return _PyBytes_FromBuffer(x); |
2854 | |
2855 | if (PyList_CheckExact(x)) |
2856 | return _PyBytes_FromList(x); |
2857 | |
2858 | if (PyTuple_CheckExact(x)) |
2859 | return _PyBytes_FromTuple(x); |
2860 | |
2861 | if (!PyUnicode_Check(x)) { Branch (2861:9): [True: 2.86k, False: 4]
|
2862 | it = PyObject_GetIter(x); |
2863 | if (it != NULL) { Branch (2863:13): [True: 2.85k, False: 9]
|
2864 | result = _PyBytes_FromIterator(it, x); |
2865 | Py_DECREF(it); |
2866 | return result; |
2867 | } |
2868 | if (!PyErr_ExceptionMatches(PyExc_TypeError)) { Branch (2868:13): [True: 1, False: 8]
|
2869 | return NULL; |
2870 | } |
2871 | } |
2872 | |
2873 | PyErr_Format(PyExc_TypeError, |
2874 | "cannot convert '%.200s' object to bytes", |
2875 | Py_TYPE(x)->tp_name); |
2876 | return NULL; |
2877 | } |
2878 | |
2879 | /* This allocator is needed for subclasses don't want to use __new__. |
2880 | * See https://github.com/python/cpython/issues/91020#issuecomment-1096793239 |
2881 | * |
2882 | * This allocator will be removed when ob_shash is removed. |
2883 | */ |
2884 | static PyObject * |
2885 | bytes_alloc(PyTypeObject *self, Py_ssize_t nitems) |
2886 | { |
2887 | PyBytesObject *obj = (PyBytesObject*)PyType_GenericAlloc(self, nitems); |
2888 | if (obj == NULL) { Branch (2888:9): [True: 0, False: 0]
|
2889 | return NULL; |
2890 | } |
2891 | _Py_COMP_DIAG_PUSH |
2892 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
2893 | obj->ob_shash = -1; |
2894 | _Py_COMP_DIAG_POP |
2895 | return (PyObject*)obj; |
2896 | } |
2897 | |
2898 | static PyObject * |
2899 | bytes_subtype_new(PyTypeObject *type, PyObject *tmp) |
2900 | { |
2901 | PyObject *pnew; |
2902 | Py_ssize_t n; |
2903 | |
2904 | assert(PyType_IsSubtype(type, &PyBytes_Type)); |
2905 | assert(PyBytes_Check(tmp)); |
2906 | n = PyBytes_GET_SIZE(tmp); |
2907 | pnew = type->tp_alloc(type, n); |
2908 | if (pnew != NULL) { Branch (2908:9): [True: 858, False: 0]
|
2909 | memcpy(PyBytes_AS_STRING(pnew), |
2910 | PyBytes_AS_STRING(tmp), n+1); |
2911 | _Py_COMP_DIAG_PUSH |
2912 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
2913 | ((PyBytesObject *)pnew)->ob_shash = |
2914 | ((PyBytesObject *)tmp)->ob_shash; |
2915 | _Py_COMP_DIAG_POP |
2916 | } |
2917 | return pnew; |
2918 | } |
2919 | |
2920 | PyDoc_STRVAR(bytes_doc, |
2921 | "bytes(iterable_of_ints) -> bytes\n\ |
2922 | bytes(string, encoding[, errors]) -> bytes\n\ |
2923 | bytes(bytes_or_buffer) -> immutable copy of bytes_or_buffer\n\ |
2924 | bytes(int) -> bytes object of size given by the parameter initialized with null bytes\n\ |
2925 | bytes() -> empty bytes object\n\ |
2926 | \n\ |
2927 | Construct an immutable array of bytes from:\n\ |
2928 | - an iterable yielding integers in range(256)\n\ |
2929 | - a text string encoded using the specified encoding\n\ |
2930 | - any object implementing the buffer API.\n\ |
2931 | - an integer"); |
2932 | |
2933 | static PyObject *bytes_iter(PyObject *seq); |
2934 | |
2935 | PyTypeObject PyBytes_Type = { |
2936 | PyVarObject_HEAD_INIT(&PyType_Type, 0) |
2937 | "bytes", |
2938 | PyBytesObject_SIZE, |
2939 | sizeof(char), |
2940 | 0, /* tp_dealloc */ |
2941 | 0, /* tp_vectorcall_offset */ |
2942 | 0, /* tp_getattr */ |
2943 | 0, /* tp_setattr */ |
2944 | 0, /* tp_as_async */ |
2945 | (reprfunc)bytes_repr, /* tp_repr */ |
2946 | &bytes_as_number, /* tp_as_number */ |
2947 | &bytes_as_sequence, /* tp_as_sequence */ |
2948 | &bytes_as_mapping, /* tp_as_mapping */ |
2949 | (hashfunc)bytes_hash, /* tp_hash */ |
2950 | 0, /* tp_call */ |
2951 | bytes_str, /* tp_str */ |
2952 | PyObject_GenericGetAttr, /* tp_getattro */ |
2953 | 0, /* tp_setattro */ |
2954 | &bytes_as_buffer, /* tp_as_buffer */ |
2955 | Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | |
2956 | Py_TPFLAGS_BYTES_SUBCLASS | |
2957 | _Py_TPFLAGS_MATCH_SELF, /* tp_flags */ |
2958 | bytes_doc, /* tp_doc */ |
2959 | 0, /* tp_traverse */ |
2960 | 0, /* tp_clear */ |
2961 | (richcmpfunc)bytes_richcompare, /* tp_richcompare */ |
2962 | 0, /* tp_weaklistoffset */ |
2963 | bytes_iter, /* tp_iter */ |
2964 | 0, /* tp_iternext */ |
2965 | bytes_methods, /* tp_methods */ |
2966 | 0, /* tp_members */ |
2967 | 0, /* tp_getset */ |
2968 | 0, /* tp_base */ |
2969 | 0, /* tp_dict */ |
2970 | 0, /* tp_descr_get */ |
2971 | 0, /* tp_descr_set */ |
2972 | 0, /* tp_dictoffset */ |
2973 | 0, /* tp_init */ |
2974 | bytes_alloc, /* tp_alloc */ |
2975 | bytes_new, /* tp_new */ |
2976 | PyObject_Del, /* tp_free */ |
2977 | }; |
2978 | |
2979 | void |
2980 | PyBytes_Concat(PyObject **pv, PyObject *w) |
2981 | { |
2982 | assert(pv != NULL); |
2983 | if (*pv == NULL) Branch (2983:9): [True: 0, False: 17.3k]
|
2984 | return; |
2985 | if (w == NULL) { Branch (2985:9): [True: 0, False: 17.3k]
|
2986 | Py_CLEAR(*pv); |
2987 | return; |
2988 | } |
2989 | |
2990 | if (Py_REFCNT(*pv) == 1 && PyBytes_CheckExact4.25k (*pv)) { Branch (2990:9): [True: 4.25k, False: 13.0k]
|
2991 | /* Only one reference, so we can resize in place */ |
2992 | Py_ssize_t oldsize; |
2993 | Py_buffer wb; |
2994 | |
2995 | if (PyObject_GetBuffer(w, &wb, PyBUF_SIMPLE) != 0) { Branch (2995:13): [True: 0, False: 4.25k]
|
2996 | PyErr_Format(PyExc_TypeError, "can't concat %.100s to %.100s", |
2997 | Py_TYPE(w)->tp_name, Py_TYPE(*pv)->tp_name); |
2998 | Py_CLEAR(*pv); |
2999 | return; |
3000 | } |
3001 | |
3002 | oldsize = PyBytes_GET_SIZE(*pv); |
3003 | if (oldsize > PY_SSIZE_T_MAX - wb.len) { Branch (3003:13): [True: 0, False: 4.25k]
|
3004 | PyErr_NoMemory(); |
3005 | goto error; |
3006 | } |
3007 | if (_PyBytes_Resize(pv, oldsize + wb.len) < 0) Branch (3007:13): [True: 0, False: 4.25k]
|
3008 | goto error; |
3009 | |
3010 | memcpy(PyBytes_AS_STRING(*pv) + oldsize, wb.buf, wb.len); |
3011 | PyBuffer_Release(&wb); |
3012 | return; |
3013 | |
3014 | error: |
3015 | PyBuffer_Release(&wb); |
3016 | Py_CLEAR(*pv); |
3017 | return; |
3018 | } |
3019 | |
3020 | else { |
3021 | /* Multiple references, need to create new object */ |
3022 | PyObject *v; |
3023 | v = bytes_concat(*pv, w); |
3024 | Py_SETREF(*pv, v); |
3025 | } |
3026 | } |
3027 | |
3028 | void |
3029 | PyBytes_ConcatAndDel(PyObject **pv, PyObject *w) |
3030 | { |
3031 | PyBytes_Concat(pv, w); |
3032 | Py_XDECREF(w); |
3033 | } |
3034 | |
3035 | |
3036 | /* The following function breaks the notion that bytes are immutable: |
3037 | it changes the size of a bytes object. We get away with this only if there |
3038 | is only one module referencing the object. You can also think of it |
3039 | as creating a new bytes object and destroying the old one, only |
3040 | more efficiently. In any case, don't use this if the bytes object may |
3041 | already be known to some other part of the code... |
3042 | Note that if there's not enough memory to resize the bytes object, the |
3043 | original bytes object at *pv is deallocated, *pv is set to NULL, an "out of |
3044 | memory" exception is set, and -1 is returned. Else (on success) 0 is |
3045 | returned, and the value in *pv may or may not be the same as on input. |
3046 | As always, an extra byte is allocated for a trailing \0 byte (newsize |
3047 | does *not* include that), and a trailing \0 byte is stored. |
3048 | */ |
3049 | |
3050 | int |
3051 | _PyBytes_Resize(PyObject **pv, Py_ssize_t newsize) |
3052 | { |
3053 | PyObject *v; |
3054 | PyBytesObject *sv; |
3055 | v = *pv; |
3056 | if (!PyBytes_Check(v) || newsize < 0) { Branch (3056:9): [True: 0, False: 2.30M]
Branch (3056:30): [True: 0, False: 2.30M]
|
3057 | goto error; |
3058 | } |
3059 | if (Py_SIZE(v) == newsize) { Branch (3059:9): [True: 5.12k, False: 2.29M]
|
3060 | /* return early if newsize equals to v->ob_size */ |
3061 | return 0; |
3062 | } |
3063 | if (Py_SIZE(v) == 0) { Branch (3063:9): [True: 0, False: 2.29M]
|
3064 | if (newsize == 0) { Branch (3064:13): [True: 0, False: 0]
|
3065 | return 0; |
3066 | } |
3067 | *pv = _PyBytes_FromSize(newsize, 0); |
3068 | Py_DECREF(v); |
3069 | return (*pv == NULL) ? -1 : 0; Branch (3069:16): [True: 0, False: 0]
|
3070 | } |
3071 | if (Py_REFCNT(v) != 1) { Branch (3071:9): [True: 0, False: 2.29M]
|
3072 | goto error; |
3073 | } |
3074 | if (newsize == 0) { Branch (3074:9): [True: 1.14M, False: 1.15M]
|
3075 | *pv = bytes_new_empty(); |
3076 | Py_DECREF(v); |
3077 | return 0; |
3078 | } |
3079 | /* XXX UNREF/NEWREF interface should be more symmetrical */ |
3080 | #ifdef Py_REF_DEBUG |
3081 | _Py_RefTotal--; |
3082 | #endif |
3083 | #ifdef Py_TRACE_REFS |
3084 | _Py_ForgetReference(v); |
3085 | #endif |
3086 | *pv = (PyObject *) |
3087 | PyObject_Realloc(v, PyBytesObject_SIZE + newsize); |
3088 | if (*pv == NULL) { Branch (3088:9): [True: 0, False: 1.15M]
|
3089 | PyObject_Free(v); |
3090 | PyErr_NoMemory(); |
3091 | return -1; |
3092 | } |
3093 | _Py_NewReference(*pv); |
3094 | sv = (PyBytesObject *) *pv; |
3095 | Py_SET_SIZE(sv, newsize); |
3096 | sv->ob_sval[newsize] = '\0'; |
3097 | _Py_COMP_DIAG_PUSH |
3098 | _Py_COMP_DIAG_IGNORE_DEPR_DECLS |
3099 | sv->ob_shash = -1; /* invalidate cached hash value */ |
3100 | _Py_COMP_DIAG_POP |
3101 | return 0; |
3102 | error: |
3103 | *pv = 0; |
3104 | Py_DECREF(v); |
3105 | PyErr_BadInternalCall(); |
3106 | return -1; |
3107 | } |
3108 | |
3109 | |
3110 | PyStatus |
3111 | _PyBytes_InitTypes(PyInterpreterState *interp) |
3112 | { |
3113 | if (!_Py_IsMainInterpreter(interp)) { Branch (3113:9): [True: 171, False: 107]
|
3114 | return _PyStatus_OK(); |
3115 | } |
3116 | |
3117 | if (PyType_Ready(&PyBytes_Type) < 0) { Branch (3117:9): [True: 0, False: 107]
|
3118 | return _PyStatus_ERR("Can't initialize bytes type"); |
3119 | } |
3120 | |
3121 | if (PyType_Ready(&PyBytesIter_Type) < 0) { Branch (3121:9): [True: 0, False: 107]
|
3122 | return _PyStatus_ERR("Can't initialize bytes iterator type"); |
3123 | } |
3124 | |
3125 | return _PyStatus_OK(); |
3126 | } |
3127 | |
3128 | |
3129 | /*********************** Bytes Iterator ****************************/ |
3130 | |
3131 | typedef struct { |
3132 | PyObject_HEAD |
3133 | Py_ssize_t it_index; |
3134 | PyBytesObject *it_seq; /* Set to NULL when iterator is exhausted */ |
3135 | } striterobject; |
3136 | |
3137 | static void |
3138 | striter_dealloc(striterobject *it) |
3139 | { |
3140 | _PyObject_GC_UNTRACK(it); |
3141 | Py_XDECREF(it->it_seq); |
3142 | PyObject_GC_Del(it); |
3143 | } |
3144 | |
3145 | static int |
3146 | striter_traverse(striterobject *it, visitproc visit, void *arg) |
3147 | { |
3148 | Py_VISIT(it->it_seq); |
3149 | return 0; |
3150 | } |
3151 | |
3152 | static PyObject * |
3153 | striter_next(striterobject *it) |
3154 | { |
3155 | PyBytesObject *seq; |
3156 | |
3157 | assert(it != NULL); |
3158 | seq = it->it_seq; |
3159 | if (seq == NULL) Branch (3159:9): [True: 1, False: 1.45M]
|
3160 | return NULL; |
3161 | assert(PyBytes_Check(seq)); |
3162 | |
3163 | if (it->it_index < PyBytes_GET_SIZE(seq)) { Branch (3163:9): [True: 1.38M, False: 63.1k]
|
3164 | return _PyLong_FromUnsignedChar( |
3165 | (unsigned char)seq->ob_sval[it->it_index++]); |
3166 | } |
3167 | |
3168 | it->it_seq = NULL; |
3169 | Py_DECREF(seq); |
3170 | return NULL; |
3171 | } |
3172 | |
3173 | static PyObject * |
3174 | striter_len(striterobject *it, PyObject *Py_UNUSED(ignored)) |
3175 | { |
3176 | Py_ssize_t len = 0; |
3177 | if (it->it_seq) Branch (3177:9): [True: 1.27k, False: 0]
|
3178 | len = PyBytes_GET_SIZE(it->it_seq) - it->it_index; |
3179 | return PyLong_FromSsize_t(len); |
3180 | } |
3181 | |
3182 | PyDoc_STRVAR(length_hint_doc, |
3183 | "Private method returning an estimate of len(list(it))."); |
3184 | |
3185 | static PyObject * |
3186 | striter_reduce(striterobject *it, PyObject *Py_UNUSED(ignored)) |
3187 | { |
3188 | if (it->it_seq != NULL) { Branch (3188:9): [True: 54, False: 0]
|
3189 | return Py_BuildValue("N(O)n", _PyEval_GetBuiltin(&_Py_ID(iter)), |
3190 | it->it_seq, it->it_index); |
3191 | } else { |
3192 | return Py_BuildValue("N(())", _PyEval_GetBuiltin(&_Py_ID(iter))); |
3193 | } |
3194 | } |
3195 | |
3196 | PyDoc_STRVAR(reduce_doc, "Return state information for pickling."); |
3197 | |
3198 | static PyObject * |
3199 | striter_setstate(striterobject *it, PyObject *state) |
3200 | { |
3201 | Py_ssize_t index = PyLong_AsSsize_t(state); |
3202 | if (index == -1 && PyErr_Occurred()0 ) Branch (3202:9): [True: 0, False: 84]
Branch (3202:24): [True: 0, False: 0]
|
3203 | return NULL; |
3204 | if (it->it_seq != NULL) { Branch (3204:9): [True: 84, False: 0]
|
3205 | if (index < 0) Branch (3205:13): [True: 0, False: 84]
|
3206 | index = 0; |
3207 | else if (index > PyBytes_GET_SIZE(it->it_seq)) Branch (3207:18): [True: 0, False: 84]
|
3208 | index = PyBytes_GET_SIZE(it->it_seq); /* iterator exhausted */ |
3209 | it->it_index = index; |
3210 | } |
3211 | Py_RETURN_NONE; |
3212 | } |
3213 | |
3214 | PyDoc_STRVAR(setstate_doc, "Set state information for unpickling."); |
3215 | |
3216 | static PyMethodDef striter_methods[] = { |
3217 | {"__length_hint__", (PyCFunction)striter_len, METH_NOARGS, |
3218 | length_hint_doc}, |
3219 | {"__reduce__", (PyCFunction)striter_reduce, METH_NOARGS, |
3220 | reduce_doc}, |
3221 | {"__setstate__", (PyCFunction)striter_setstate, METH_O, |
3222 | setstate_doc}, |
3223 | {NULL, NULL} /* sentinel */ |
3224 | }; |
3225 | |
3226 | PyTypeObject PyBytesIter_Type = { |
3227 | PyVarObject_HEAD_INIT(&PyType_Type, 0) |
3228 | "bytes_iterator", /* tp_name */ |
3229 | sizeof(striterobject), /* tp_basicsize */ |
3230 | 0, /* tp_itemsize */ |
3231 | /* methods */ |
3232 | (destructor)striter_dealloc, /* tp_dealloc */ |
3233 | 0, /* tp_vectorcall_offset */ |
3234 | 0, /* tp_getattr */ |
3235 | 0, /* tp_setattr */ |
3236 | 0, /* tp_as_async */ |
3237 | 0, /* tp_repr */ |
3238 | 0, /* tp_as_number */ |
3239 | 0, /* tp_as_sequence */ |
3240 | 0, /* tp_as_mapping */ |
3241 | 0, /* tp_hash */ |
3242 | 0, /* tp_call */ |
3243 | 0, /* tp_str */ |
3244 | PyObject_GenericGetAttr, /* tp_getattro */ |
3245 | 0, /* tp_setattro */ |
3246 | 0, /* tp_as_buffer */ |
3247 | Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC,/* tp_flags */ |
3248 | 0, /* tp_doc */ |
3249 | (traverseproc)striter_traverse, /* tp_traverse */ |
3250 | 0, /* tp_clear */ |
3251 | 0, /* tp_richcompare */ |
3252 | 0, /* tp_weaklistoffset */ |
3253 | PyObject_SelfIter, /* tp_iter */ |
3254 | (iternextfunc)striter_next, /* tp_iternext */ |
3255 | striter_methods, /* tp_methods */ |
3256 | 0, |
3257 | }; |
3258 | |
3259 | static PyObject * |
3260 | bytes_iter(PyObject *seq) |
3261 | { |
3262 | striterobject *it; |
3263 | |
3264 | if (!PyBytes_Check(seq)) { Branch (3264:9): [True: 0, False: 64.4k]
|
3265 | PyErr_BadInternalCall(); |
3266 | return NULL; |
3267 | } |
3268 | it = PyObject_GC_New(striterobject, &PyBytesIter_Type); |
3269 | if (it == NULL) Branch (3269:9): [True: 0, False: 64.4k]
|
3270 | return NULL; |
3271 | it->it_index = 0; |
3272 | Py_INCREF(seq); |
3273 | it->it_seq = (PyBytesObject *)seq; |
3274 | _PyObject_GC_TRACK(it); |
3275 | return (PyObject *)it; |
3276 | } |
3277 | |
3278 | |
3279 | /* _PyBytesWriter API */ |
3280 | |
3281 | #ifdef MS_WINDOWS |
3282 | /* On Windows, overallocate by 50% is the best factor */ |
3283 | # define OVERALLOCATE_FACTOR 2 |
3284 | #else |
3285 | /* On Linux, overallocate by 25% is the best factor */ |
3286 | # define OVERALLOCATE_FACTOR 4 |
3287 | #endif |
3288 | |
3289 | void |
3290 | _PyBytesWriter_Init(_PyBytesWriter *writer) |
3291 | { |
3292 | /* Set all attributes before small_buffer to 0 */ |
3293 | memset(writer, 0, offsetof(_PyBytesWriter, small_buffer)); |
3294 | #ifndef NDEBUG |
3295 | memset(writer->small_buffer, PYMEM_CLEANBYTE, |
3296 | sizeof(writer->small_buffer)); |
3297 | #endif |
3298 | } |
3299 | |
3300 | void |
3301 | _PyBytesWriter_Dealloc(_PyBytesWriter *writer) |
3302 | { |
3303 | Py_CLEAR(writer->buffer); |
3304 | } |
3305 | |
3306 | Py_LOCAL_INLINE(char*) |
3307 | _PyBytesWriter_AsString(_PyBytesWriter *writer) |
3308 | { |
3309 | if (writer->use_small_buffer) { Branch (3309:9): [True: 1.59M, False: 5.87k]
|
3310 | assert(writer->buffer == NULL); |
3311 | return writer->small_buffer; |
3312 | } |
3313 | else if (writer->use_bytearray) { Branch (3313:14): [True: 2, False: 5.87k]
|
3314 | assert(writer->buffer != NULL); |
3315 | return PyByteArray_AS_STRING(writer->buffer); |
3316 | } |
3317 | else { |
3318 | assert(writer->buffer != NULL); |
3319 | return PyBytes_AS_STRING(writer->buffer); |
3320 | } |
3321 | } |
3322 | |
3323 | Py_LOCAL_INLINE(Py_ssize_t) |
3324 | _PyBytesWriter_GetSize(_PyBytesWriter *writer, char *str) |
3325 | { |
3326 | const char *start = _PyBytesWriter_AsString(writer); |
3327 | assert(str != NULL); |
3328 | assert(str >= start); |
3329 | assert(str - start <= writer->allocated); |
3330 | return str - start; |
3331 | } |
3332 | |
3333 | #ifndef NDEBUG |
3334 | Py_LOCAL_INLINE(int) |
3335 | _PyBytesWriter_CheckConsistency(_PyBytesWriter *writer, char *str) |
3336 | { |
3337 | const char *start, *end; |
3338 | |
3339 | if (writer->use_small_buffer) { |
3340 | assert(writer->buffer == NULL); |
3341 | } |
3342 | else { |
3343 | assert(writer->buffer != NULL); |
3344 | if (writer->use_bytearray) |
3345 | assert(PyByteArray_CheckExact(writer->buffer)); |
3346 | else |
3347 | assert(PyBytes_CheckExact(writer->buffer)); |
3348 | assert(Py_REFCNT(writer->buffer) == 1); |
3349 | } |
3350 | |
3351 | if (writer->use_bytearray) { |
3352 | /* bytearray has its own overallocation algorithm, |
3353 | writer overallocation must be disabled */ |
3354 | assert(!writer->overallocate); |
3355 | } |
3356 | |
3357 | assert(0 <= writer->allocated); |
3358 | assert(0 <= writer->min_size && writer->min_size <= writer->allocated); |
3359 | /* the last byte must always be null */ |
3360 | start = _PyBytesWriter_AsString(writer); |
3361 | assert(start[writer->allocated] == 0); |
3362 | |
3363 | end = start + writer->allocated; |
3364 | assert(str != NULL); |
3365 | assert(start <= str && str <= end); |
3366 | return 1; |
3367 | } |
3368 | #endif |
3369 | |
3370 | void* |
3371 | _PyBytesWriter_Resize(_PyBytesWriter *writer, void *str, Py_ssize_t size) |
3372 | { |
3373 | Py_ssize_t allocated, pos; |
3374 | |
3375 | assert(_PyBytesWriter_CheckConsistency(writer, str)); |
3376 | assert(writer->allocated < size); |
3377 | |
3378 | allocated = size; |
3379 | if (writer->overallocate Branch (3379:9): [True: 1.82k, False: 1.28k]
|
3380 | && allocated <= (1.82k PY_SSIZE_T_MAX1.82k - allocated / OVERALLOCATE_FACTOR1.82k )) { Branch (3380:12): [True: 1.82k, False: 0]
|
3381 | /* overallocate to limit the number of realloc() */ |
3382 | allocated += allocated / OVERALLOCATE_FACTOR; |
3383 | } |
3384 | |
3385 | pos = _PyBytesWriter_GetSize(writer, str); |
3386 | if (!writer->use_small_buffer) { Branch (3386:9): [True: 1.58k, False: 1.52k]
|
3387 | if (writer->use_bytearray) { Branch (3387:13): [True: 0, False: 1.58k]
|
3388 | if (PyByteArray_Resize(writer->buffer, allocated)) Branch (3388:17): [True: 0, False: 0]
|
3389 | goto error; |
3390 | /* writer->allocated can be smaller than writer->buffer->ob_alloc, |
3391 | but we cannot use ob_alloc because bytes may need to be moved |
3392 | to use the whole buffer. bytearray uses an internal optimization |
3393 | to avoid moving or copying bytes when bytes are removed at the |
3394 | beginning (ex: del bytearray[:1]). */ |
3395 | } |
3396 | else { |
3397 | if (_PyBytes_Resize(&writer->buffer, allocated)) Branch (3397:17): [True: 0, False: 1.58k]
|
3398 | goto error; |
3399 | } |
3400 | } |
3401 | else { |
3402 | /* convert from stack buffer to bytes object buffer */ |
3403 | assert(writer->buffer == NULL); |
3404 | |
3405 | if (writer->use_bytearray) Branch (3405:13): [True: 1, False: 1.52k]
|
3406 | writer->buffer = PyByteArray_FromStringAndSize(NULL, allocated); |
3407 | else |
3408 | writer->buffer = PyBytes_FromStringAndSize(NULL, allocated); |
3409 | if (writer->buffer == NULL) Branch (3409:13): [True: 0, False: 1.52k]
|
3410 | goto error; |
3411 | |
3412 | if (pos != 0) { Branch (3412:13): [True: 261, False: 1.26k]
|
3413 | char *dest; |
3414 | if (writer->use_bytearray) Branch (3414:17): [True: 0, False: 261]
|
3415 | dest = PyByteArray_AS_STRING(writer->buffer); |
3416 | else |
3417 | dest = PyBytes_AS_STRING(writer->buffer); |
3418 | memcpy(dest, |
3419 | writer->small_buffer, |
3420 | pos); |
3421 | } |
3422 | |
3423 | writer->use_small_buffer = 0; |
3424 | #ifndef NDEBUG |
3425 | memset(writer->small_buffer, PYMEM_CLEANBYTE, |
3426 | sizeof(writer->small_buffer)); |
3427 | #endif |
3428 | } |
3429 | writer->allocated = allocated; |
3430 | |
3431 | str = _PyBytesWriter_AsString(writer) + pos; |
3432 | assert(_PyBytesWriter_CheckConsistency(writer, str)); |
3433 | return str; |
3434 | |
3435 | error: |
3436 | _PyBytesWriter_Dealloc(writer); |
3437 | return NULL; |
3438 | } |
3439 | |
3440 | void* |
3441 | _PyBytesWriter_Prepare(_PyBytesWriter *writer, void *str, Py_ssize_t size) |
3442 | { |
3443 | Py_ssize_t new_min_size; |
3444 | |
3445 | assert(_PyBytesWriter_CheckConsistency(writer, str)); |
3446 | assert(size >= 0); |
3447 | |
3448 | if (size == 0) { Branch (3448:9): [True: 2.38k, False: 1.63M]
|
3449 | /* nothing to do */ |
3450 | return str; |
3451 | } |
3452 | |
3453 | if (writer->min_size > PY_SSIZE_T_MAX - size) { Branch (3453:9): [True: 0, False: 1.63M]
|
3454 | PyErr_NoMemory(); |
3455 | _PyBytesWriter_Dealloc(writer); |
3456 | return NULL; |
3457 | } |
3458 | new_min_size = writer->min_size + size; |
3459 | |
3460 | if (new_min_size > writer->allocated) Branch (3460:9): [True: 1.56k, False: 1.63M]
|
3461 | str = _PyBytesWriter_Resize(writer, str, new_min_size); |
3462 | |
3463 | writer->min_size = new_min_size; |
3464 | return str; |
3465 | } |
3466 | |
3467 | /* Allocate the buffer to write size bytes. |
3468 | Return the pointer to the beginning of buffer data. |
3469 | Raise an exception and return NULL on error. */ |
3470 | void* |
3471 | _PyBytesWriter_Alloc(_PyBytesWriter *writer, Py_ssize_t size) |
3472 | { |
3473 | /* ensure that _PyBytesWriter_Alloc() is only called once */ |
3474 | assert(writer->min_size == 0 && writer->buffer == NULL); |
3475 | assert(size >= 0); |
3476 | |
3477 | writer->use_small_buffer = 1; |
3478 | #ifndef NDEBUG |
3479 | writer->allocated = sizeof(writer->small_buffer) - 1; |
3480 | /* In debug mode, don't use the full small buffer because it is less |
3481 | efficient than bytes and bytearray objects to detect buffer underflow |
3482 | and buffer overflow. Use 10 bytes of the small buffer to test also |
3483 | code using the smaller buffer in debug mode. |
3484 | |
3485 | Don't modify the _PyBytesWriter structure (use a shorter small buffer) |
3486 | in debug mode to also be able to detect stack overflow when running |
3487 | tests in debug mode. The _PyBytesWriter is large (more than 512 bytes), |
3488 | if _Py_EnterRecursiveCall() is not used in deep C callback, we may hit a |
3489 | stack overflow. */ |
3490 | writer->allocated = Py_MIN(writer->allocated, 10); |
3491 | /* _PyBytesWriter_CheckConsistency() requires the last byte to be 0, |
3492 | to detect buffer overflow */ |
3493 | writer->small_buffer[writer->allocated] = 0; |
3494 | #else |
3495 | writer->allocated = sizeof(writer->small_buffer); |
3496 | #endif |
3497 | return _PyBytesWriter_Prepare(writer, writer->small_buffer, size); |
3498 | } |
3499 | |
3500 | PyObject * |
3501 | _PyBytesWriter_Finish(_PyBytesWriter *writer, void *str) |
3502 | { |
3503 | Py_ssize_t size; |
3504 | PyObject *result; |
3505 | |
3506 | assert(_PyBytesWriter_CheckConsistency(writer, str)); |
3507 | |
3508 | size = _PyBytesWriter_GetSize(writer, str); |
3509 | if (size == 0 && !writer->use_bytearray1.43k ) { Branch (3509:9): [True: 1.43k, False: 1.59M]
Branch (3509:22): [True: 1.42k, False: 7]
|
3510 | Py_CLEAR(writer->buffer); |
3511 | /* Get the empty byte string singleton */ |
3512 | result = PyBytes_FromStringAndSize(NULL, 0); |
3513 | } |
3514 | else if (writer->use_small_buffer) { Branch (3514:14): [True: 1.58M, False: 1.18k]
|
3515 | if (writer->use_bytearray) { Branch (3515:13): [True: 174, False: 1.58M]
|
3516 | result = PyByteArray_FromStringAndSize(writer->small_buffer, size); |
3517 | } |
3518 | else { |
3519 | result = PyBytes_FromStringAndSize(writer->small_buffer, size); |
3520 | } |
3521 | } |
3522 | else { |
3523 | result = writer->buffer; |
3524 | writer->buffer = NULL; |
3525 | |
3526 | if (size != writer->allocated) { Branch (3526:13): [True: 1.10k, False: 73]
|
3527 | if (writer->use_bytearray) { Branch (3527:17): [True: 1, False: 1.10k]
|
3528 | if (PyByteArray_Resize(result, size)) { Branch (3528:21): [True: 0, False: 1]
|
3529 | Py_DECREF(result); |
3530 | return NULL; |
3531 | } |
3532 | } |
3533 | else { |
3534 | if (_PyBytes_Resize(&result, size)) { Branch (3534:21): [True: 0, False: 1.10k]
|
3535 | assert(result == NULL); |
3536 | return NULL; |
3537 | } |
3538 | } |
3539 | } |
3540 | } |
3541 | return result; |
3542 | } |
3543 | |
3544 | void* |
3545 | _PyBytesWriter_WriteBytes(_PyBytesWriter *writer, void *ptr, |
3546 | const void *bytes, Py_ssize_t size) |
3547 | { |
3548 | char *str = (char *)ptr; |
3549 | |
3550 | str = _PyBytesWriter_Prepare(writer, str, size); |
3551 | if (str == NULL) Branch (3551:9): [True: 0, False: 17.8k]
|
3552 | return NULL; |
3553 | |
3554 | memcpy(str, bytes, size); |
3555 | str += size; |
3556 | |
3557 | return str; |
3558 | } |
3559 | |
3560 | |
3561 | void |
3562 | _PyBytes_Repeat(char* dest, Py_ssize_t len_dest, |
3563 | const char* src, Py_ssize_t len_src) |
3564 | { |
3565 | if (len_dest == 0) { Branch (3565:9): [True: 13.5k, False: 77.6k]
|
3566 | return; |
3567 | } |
3568 | if (len_src == 1) { Branch (3568:9): [True: 24.6k, False: 52.9k]
|
3569 | memset(dest, src[0], len_dest); |
3570 | } |
3571 | else { |
3572 | if (src != dest) { Branch (3572:13): [True: 52.9k, False: 14]
|
3573 | memcpy(dest, src, len_src); |
3574 | } |
3575 | Py_ssize_t copied = len_src; |
3576 | while (copied < len_dest) { Branch (3576:16): [True: 175k, False: 52.9k]
|
3577 | Py_ssize_t bytes_to_copy = Py_MIN(copied, len_dest - copied); |
3578 | memcpy(dest + copied, dest, bytes_to_copy); |
3579 | copied += bytes_to_copy; |
3580 | } |
3581 | } |
3582 | } |
3583 | |