]> git.proxmox.com Git - mirror_edk2.git/blob - AppPkg/Applications/Python/Python-2.7.10/Modules/stropmodule.c
cdc6cb8ef8bb19f60d6441f4d944d0db61722ebc
[mirror_edk2.git] / AppPkg / Applications / Python / Python-2.7.10 / Modules / stropmodule.c
1 /* strop module */
2
3 #define PY_SSIZE_T_CLEAN
4 #include "Python.h"
5 #include <ctype.h>
6
7 PyDoc_STRVAR(strop_module__doc__,
8 "Common string manipulations, optimized for speed.\n"
9 "\n"
10 "Always use \"import string\" rather than referencing\n"
11 "this module directly.");
12
13 /* XXX This file assumes that the <ctype.h> is*() functions
14 XXX are defined for all 8-bit characters! */
15
16 #define WARN if (PyErr_Warn(PyExc_DeprecationWarning, \
17 "strop functions are obsolete; use string methods")) \
18 return NULL
19
20 /* The lstrip(), rstrip() and strip() functions are implemented
21 in do_strip(), which uses an additional parameter to indicate what
22 type of strip should occur. */
23
24 #define LEFTSTRIP 0
25 #define RIGHTSTRIP 1
26 #define BOTHSTRIP 2
27
28
29 static PyObject *
30 split_whitespace(char *s, Py_ssize_t len, Py_ssize_t maxsplit)
31 {
32 Py_ssize_t i = 0, j;
33 int err;
34 Py_ssize_t countsplit = 0;
35 PyObject* item;
36 PyObject *list = PyList_New(0);
37
38 if (list == NULL)
39 return NULL;
40
41 while (i < len) {
42 while (i < len && isspace(Py_CHARMASK(s[i]))) {
43 i = i+1;
44 }
45 j = i;
46 while (i < len && !isspace(Py_CHARMASK(s[i]))) {
47 i = i+1;
48 }
49 if (j < i) {
50 item = PyString_FromStringAndSize(s+j, i-j);
51 if (item == NULL)
52 goto finally;
53
54 err = PyList_Append(list, item);
55 Py_DECREF(item);
56 if (err < 0)
57 goto finally;
58
59 countsplit++;
60 while (i < len && isspace(Py_CHARMASK(s[i]))) {
61 i = i+1;
62 }
63 if (maxsplit && (countsplit >= maxsplit) && i < len) {
64 item = PyString_FromStringAndSize(
65 s+i, len - i);
66 if (item == NULL)
67 goto finally;
68
69 err = PyList_Append(list, item);
70 Py_DECREF(item);
71 if (err < 0)
72 goto finally;
73
74 i = len;
75 }
76 }
77 }
78 return list;
79 finally:
80 Py_DECREF(list);
81 return NULL;
82 }
83
84
85 PyDoc_STRVAR(splitfields__doc__,
86 "split(s [,sep [,maxsplit]]) -> list of strings\n"
87 "splitfields(s [,sep [,maxsplit]]) -> list of strings\n"
88 "\n"
89 "Return a list of the words in the string s, using sep as the\n"
90 "delimiter string. If maxsplit is nonzero, splits into at most\n"
91 "maxsplit words. If sep is not specified, any whitespace string\n"
92 "is a separator. Maxsplit defaults to 0.\n"
93 "\n"
94 "(split and splitfields are synonymous)");
95
96 static PyObject *
97 strop_splitfields(PyObject *self, PyObject *args)
98 {
99 Py_ssize_t len, n, i, j, err;
100 Py_ssize_t splitcount, maxsplit;
101 char *s, *sub;
102 PyObject *list, *item;
103
104 WARN;
105 sub = NULL;
106 n = 0;
107 splitcount = 0;
108 maxsplit = 0;
109 if (!PyArg_ParseTuple(args, "t#|z#n:split", &s, &len, &sub, &n, &maxsplit))
110 return NULL;
111 if (sub == NULL)
112 return split_whitespace(s, len, maxsplit);
113 if (n == 0) {
114 PyErr_SetString(PyExc_ValueError, "empty separator");
115 return NULL;
116 }
117
118 list = PyList_New(0);
119 if (list == NULL)
120 return NULL;
121
122 i = j = 0;
123 while (i+n <= len) {
124 if (s[i] == sub[0] && (n == 1 || memcmp(s+i, sub, n) == 0)) {
125 item = PyString_FromStringAndSize(s+j, i-j);
126 if (item == NULL)
127 goto fail;
128 err = PyList_Append(list, item);
129 Py_DECREF(item);
130 if (err < 0)
131 goto fail;
132 i = j = i + n;
133 splitcount++;
134 if (maxsplit && (splitcount >= maxsplit))
135 break;
136 }
137 else
138 i++;
139 }
140 item = PyString_FromStringAndSize(s+j, len-j);
141 if (item == NULL)
142 goto fail;
143 err = PyList_Append(list, item);
144 Py_DECREF(item);
145 if (err < 0)
146 goto fail;
147
148 return list;
149
150 fail:
151 Py_DECREF(list);
152 return NULL;
153 }
154
155
156 PyDoc_STRVAR(joinfields__doc__,
157 "join(list [,sep]) -> string\n"
158 "joinfields(list [,sep]) -> string\n"
159 "\n"
160 "Return a string composed of the words in list, with\n"
161 "intervening occurrences of sep. Sep defaults to a single\n"
162 "space.\n"
163 "\n"
164 "(join and joinfields are synonymous)");
165
166 static PyObject *
167 strop_joinfields(PyObject *self, PyObject *args)
168 {
169 PyObject *seq;
170 char *sep = NULL;
171 Py_ssize_t seqlen, seplen = 0;
172 Py_ssize_t i, reslen = 0, slen = 0, sz = 100;
173 PyObject *res = NULL;
174 char* p = NULL;
175 ssizeargfunc getitemfunc;
176
177 WARN;
178 if (!PyArg_ParseTuple(args, "O|t#:join", &seq, &sep, &seplen))
179 return NULL;
180 if (sep == NULL) {
181 sep = " ";
182 seplen = 1;
183 }
184
185 seqlen = PySequence_Size(seq);
186 if (seqlen < 0 && PyErr_Occurred())
187 return NULL;
188
189 if (seqlen == 1) {
190 /* Optimization if there's only one item */
191 PyObject *item = PySequence_GetItem(seq, 0);
192 if (item && !PyString_Check(item)) {
193 PyErr_SetString(PyExc_TypeError,
194 "first argument must be sequence of strings");
195 Py_DECREF(item);
196 return NULL;
197 }
198 return item;
199 }
200
201 if (!(res = PyString_FromStringAndSize((char*)NULL, sz)))
202 return NULL;
203 p = PyString_AsString(res);
204
205 /* optimize for lists, since it's the most common case. all others
206 * (tuples and arbitrary sequences) just use the sequence abstract
207 * interface.
208 */
209 if (PyList_Check(seq)) {
210 for (i = 0; i < seqlen; i++) {
211 PyObject *item = PyList_GET_ITEM(seq, i);
212 if (!PyString_Check(item)) {
213 PyErr_SetString(PyExc_TypeError,
214 "first argument must be sequence of strings");
215 Py_DECREF(res);
216 return NULL;
217 }
218 slen = PyString_GET_SIZE(item);
219 if (slen > PY_SSIZE_T_MAX - reslen ||
220 seplen > PY_SSIZE_T_MAX - reslen - seplen) {
221 PyErr_SetString(PyExc_OverflowError,
222 "input too long");
223 Py_DECREF(res);
224 return NULL;
225 }
226 while (reslen + slen + seplen >= sz) {
227 if (_PyString_Resize(&res, sz * 2) < 0)
228 return NULL;
229 sz *= 2;
230 p = PyString_AsString(res) + reslen;
231 }
232 if (i > 0) {
233 memcpy(p, sep, seplen);
234 p += seplen;
235 reslen += seplen;
236 }
237 memcpy(p, PyString_AS_STRING(item), slen);
238 p += slen;
239 reslen += slen;
240 }
241 _PyString_Resize(&res, reslen);
242 return res;
243 }
244
245 if (seq->ob_type->tp_as_sequence == NULL ||
246 (getitemfunc = seq->ob_type->tp_as_sequence->sq_item) == NULL)
247 {
248 PyErr_SetString(PyExc_TypeError,
249 "first argument must be a sequence");
250 return NULL;
251 }
252 /* This is now type safe */
253 for (i = 0; i < seqlen; i++) {
254 PyObject *item = getitemfunc(seq, i);
255 if (!item || !PyString_Check(item)) {
256 PyErr_SetString(PyExc_TypeError,
257 "first argument must be sequence of strings");
258 Py_DECREF(res);
259 Py_XDECREF(item);
260 return NULL;
261 }
262 slen = PyString_GET_SIZE(item);
263 if (slen > PY_SSIZE_T_MAX - reslen ||
264 seplen > PY_SSIZE_T_MAX - reslen - seplen) {
265 PyErr_SetString(PyExc_OverflowError,
266 "input too long");
267 Py_DECREF(res);
268 Py_XDECREF(item);
269 return NULL;
270 }
271 while (reslen + slen + seplen >= sz) {
272 if (_PyString_Resize(&res, sz * 2) < 0) {
273 Py_DECREF(item);
274 return NULL;
275 }
276 sz *= 2;
277 p = PyString_AsString(res) + reslen;
278 }
279 if (i > 0) {
280 memcpy(p, sep, seplen);
281 p += seplen;
282 reslen += seplen;
283 }
284 memcpy(p, PyString_AS_STRING(item), slen);
285 p += slen;
286 reslen += slen;
287 Py_DECREF(item);
288 }
289 _PyString_Resize(&res, reslen);
290 return res;
291 }
292
293
294 PyDoc_STRVAR(find__doc__,
295 "find(s, sub [,start [,end]]) -> in\n"
296 "\n"
297 "Return the lowest index in s where substring sub is found,\n"
298 "such that sub is contained within s[start,end]. Optional\n"
299 "arguments start and end are interpreted as in slice notation.\n"
300 "\n"
301 "Return -1 on failure.");
302
303 static PyObject *
304 strop_find(PyObject *self, PyObject *args)
305 {
306 char *s, *sub;
307 Py_ssize_t len, n, i = 0, last = PY_SSIZE_T_MAX;
308
309 WARN;
310 if (!PyArg_ParseTuple(args, "t#t#|nn:find", &s, &len, &sub, &n, &i, &last))
311 return NULL;
312
313 if (last > len)
314 last = len;
315 if (last < 0)
316 last += len;
317 if (last < 0)
318 last = 0;
319 if (i < 0)
320 i += len;
321 if (i < 0)
322 i = 0;
323
324 if (n == 0 && i <= last)
325 return PyInt_FromLong((long)i);
326
327 last -= n;
328 for (; i <= last; ++i)
329 if (s[i] == sub[0] &&
330 (n == 1 || memcmp(&s[i+1], &sub[1], n-1) == 0))
331 return PyInt_FromLong((long)i);
332
333 return PyInt_FromLong(-1L);
334 }
335
336
337 PyDoc_STRVAR(rfind__doc__,
338 "rfind(s, sub [,start [,end]]) -> int\n"
339 "\n"
340 "Return the highest index in s where substring sub is found,\n"
341 "such that sub is contained within s[start,end]. Optional\n"
342 "arguments start and end are interpreted as in slice notation.\n"
343 "\n"
344 "Return -1 on failure.");
345
346 static PyObject *
347 strop_rfind(PyObject *self, PyObject *args)
348 {
349 char *s, *sub;
350 Py_ssize_t len, n, j;
351 Py_ssize_t i = 0, last = PY_SSIZE_T_MAX;
352
353 WARN;
354 if (!PyArg_ParseTuple(args, "t#t#|nn:rfind", &s, &len, &sub, &n, &i, &last))
355 return NULL;
356
357 if (last > len)
358 last = len;
359 if (last < 0)
360 last += len;
361 if (last < 0)
362 last = 0;
363 if (i < 0)
364 i += len;
365 if (i < 0)
366 i = 0;
367
368 if (n == 0 && i <= last)
369 return PyInt_FromLong((long)last);
370
371 for (j = last-n; j >= i; --j)
372 if (s[j] == sub[0] &&
373 (n == 1 || memcmp(&s[j+1], &sub[1], n-1) == 0))
374 return PyInt_FromLong((long)j);
375
376 return PyInt_FromLong(-1L);
377 }
378
379
380 static PyObject *
381 do_strip(PyObject *args, int striptype)
382 {
383 char *s;
384 Py_ssize_t len, i, j;
385
386
387 if (PyString_AsStringAndSize(args, &s, &len))
388 return NULL;
389
390 i = 0;
391 if (striptype != RIGHTSTRIP) {
392 while (i < len && isspace(Py_CHARMASK(s[i]))) {
393 i++;
394 }
395 }
396
397 j = len;
398 if (striptype != LEFTSTRIP) {
399 do {
400 j--;
401 } while (j >= i && isspace(Py_CHARMASK(s[j])));
402 j++;
403 }
404
405 if (i == 0 && j == len) {
406 Py_INCREF(args);
407 return args;
408 }
409 else
410 return PyString_FromStringAndSize(s+i, j-i);
411 }
412
413
414 PyDoc_STRVAR(strip__doc__,
415 "strip(s) -> string\n"
416 "\n"
417 "Return a copy of the string s with leading and trailing\n"
418 "whitespace removed.");
419
420 static PyObject *
421 strop_strip(PyObject *self, PyObject *args)
422 {
423 WARN;
424 return do_strip(args, BOTHSTRIP);
425 }
426
427
428 PyDoc_STRVAR(lstrip__doc__,
429 "lstrip(s) -> string\n"
430 "\n"
431 "Return a copy of the string s with leading whitespace removed.");
432
433 static PyObject *
434 strop_lstrip(PyObject *self, PyObject *args)
435 {
436 WARN;
437 return do_strip(args, LEFTSTRIP);
438 }
439
440
441 PyDoc_STRVAR(rstrip__doc__,
442 "rstrip(s) -> string\n"
443 "\n"
444 "Return a copy of the string s with trailing whitespace removed.");
445
446 static PyObject *
447 strop_rstrip(PyObject *self, PyObject *args)
448 {
449 WARN;
450 return do_strip(args, RIGHTSTRIP);
451 }
452
453
454 PyDoc_STRVAR(lower__doc__,
455 "lower(s) -> string\n"
456 "\n"
457 "Return a copy of the string s converted to lowercase.");
458
459 static PyObject *
460 strop_lower(PyObject *self, PyObject *args)
461 {
462 char *s, *s_new;
463 Py_ssize_t i, n;
464 PyObject *newstr;
465 int changed;
466
467 WARN;
468 if (PyString_AsStringAndSize(args, &s, &n))
469 return NULL;
470 newstr = PyString_FromStringAndSize(NULL, n);
471 if (newstr == NULL)
472 return NULL;
473 s_new = PyString_AsString(newstr);
474 changed = 0;
475 for (i = 0; i < n; i++) {
476 int c = Py_CHARMASK(*s++);
477 if (isupper(c)) {
478 changed = 1;
479 *s_new = tolower(c);
480 } else
481 *s_new = c;
482 s_new++;
483 }
484 if (!changed) {
485 Py_DECREF(newstr);
486 Py_INCREF(args);
487 return args;
488 }
489 return newstr;
490 }
491
492
493 PyDoc_STRVAR(upper__doc__,
494 "upper(s) -> string\n"
495 "\n"
496 "Return a copy of the string s converted to uppercase.");
497
498 static PyObject *
499 strop_upper(PyObject *self, PyObject *args)
500 {
501 char *s, *s_new;
502 Py_ssize_t i, n;
503 PyObject *newstr;
504 int changed;
505
506 WARN;
507 if (PyString_AsStringAndSize(args, &s, &n))
508 return NULL;
509 newstr = PyString_FromStringAndSize(NULL, n);
510 if (newstr == NULL)
511 return NULL;
512 s_new = PyString_AsString(newstr);
513 changed = 0;
514 for (i = 0; i < n; i++) {
515 int c = Py_CHARMASK(*s++);
516 if (islower(c)) {
517 changed = 1;
518 *s_new = toupper(c);
519 } else
520 *s_new = c;
521 s_new++;
522 }
523 if (!changed) {
524 Py_DECREF(newstr);
525 Py_INCREF(args);
526 return args;
527 }
528 return newstr;
529 }
530
531
532 PyDoc_STRVAR(capitalize__doc__,
533 "capitalize(s) -> string\n"
534 "\n"
535 "Return a copy of the string s with only its first character\n"
536 "capitalized.");
537
538 static PyObject *
539 strop_capitalize(PyObject *self, PyObject *args)
540 {
541 char *s, *s_new;
542 Py_ssize_t i, n;
543 PyObject *newstr;
544 int changed;
545
546 WARN;
547 if (PyString_AsStringAndSize(args, &s, &n))
548 return NULL;
549 newstr = PyString_FromStringAndSize(NULL, n);
550 if (newstr == NULL)
551 return NULL;
552 s_new = PyString_AsString(newstr);
553 changed = 0;
554 if (0 < n) {
555 int c = Py_CHARMASK(*s++);
556 if (islower(c)) {
557 changed = 1;
558 *s_new = toupper(c);
559 } else
560 *s_new = c;
561 s_new++;
562 }
563 for (i = 1; i < n; i++) {
564 int c = Py_CHARMASK(*s++);
565 if (isupper(c)) {
566 changed = 1;
567 *s_new = tolower(c);
568 } else
569 *s_new = c;
570 s_new++;
571 }
572 if (!changed) {
573 Py_DECREF(newstr);
574 Py_INCREF(args);
575 return args;
576 }
577 return newstr;
578 }
579
580
581 PyDoc_STRVAR(expandtabs__doc__,
582 "expandtabs(string, [tabsize]) -> string\n"
583 "\n"
584 "Expand tabs in a string, i.e. replace them by one or more spaces,\n"
585 "depending on the current column and the given tab size (default 8).\n"
586 "The column number is reset to zero after each newline occurring in the\n"
587 "string. This doesn't understand other non-printing characters.");
588
589 static PyObject *
590 strop_expandtabs(PyObject *self, PyObject *args)
591 {
592 /* Original by Fredrik Lundh */
593 char* e;
594 char* p;
595 char* q;
596 Py_ssize_t i, j;
597 PyObject* out;
598 char* string;
599 Py_ssize_t stringlen;
600 int tabsize = 8;
601
602 WARN;
603 /* Get arguments */
604 if (!PyArg_ParseTuple(args, "s#|i:expandtabs", &string, &stringlen, &tabsize))
605 return NULL;
606 if (tabsize < 1) {
607 PyErr_SetString(PyExc_ValueError,
608 "tabsize must be at least 1");
609 return NULL;
610 }
611
612 /* First pass: determine size of output string */
613 i = j = 0; /* j: current column; i: total of previous lines */
614 e = string + stringlen;
615 for (p = string; p < e; p++) {
616 if (*p == '\t') {
617 Py_ssize_t incr = tabsize - (j%tabsize);
618 if (j > PY_SSIZE_T_MAX - incr)
619 goto overflow;
620 j += incr;
621 } else {
622 if (j > PY_SSIZE_T_MAX - 1)
623 goto overflow;
624 j++;
625 if (*p == '\n') {
626 if (i > PY_SSIZE_T_MAX - j)
627 goto overflow;
628 i += j;
629 j = 0;
630 }
631 }
632 }
633
634 if (i > PY_SSIZE_T_MAX - j)
635 goto overflow;
636
637 /* Second pass: create output string and fill it */
638 out = PyString_FromStringAndSize(NULL, i+j);
639 if (out == NULL)
640 return NULL;
641
642 i = 0;
643 q = PyString_AS_STRING(out);
644
645 for (p = string; p < e; p++) {
646 if (*p == '\t') {
647 j = tabsize - (i%tabsize);
648 i += j;
649 while (j-- > 0)
650 *q++ = ' ';
651 } else {
652 *q++ = *p;
653 i++;
654 if (*p == '\n')
655 i = 0;
656 }
657 }
658
659 return out;
660 overflow:
661 PyErr_SetString(PyExc_OverflowError, "result is too long");
662 return NULL;
663 }
664
665
666 PyDoc_STRVAR(count__doc__,
667 "count(s, sub[, start[, end]]) -> int\n"
668 "\n"
669 "Return the number of occurrences of substring sub in string\n"
670 "s[start:end]. Optional arguments start and end are\n"
671 "interpreted as in slice notation.");
672
673 static PyObject *
674 strop_count(PyObject *self, PyObject *args)
675 {
676 char *s, *sub;
677 Py_ssize_t len, n;
678 Py_ssize_t i = 0, last = PY_SSIZE_T_MAX;
679 Py_ssize_t m, r;
680
681 WARN;
682 if (!PyArg_ParseTuple(args, "t#t#|nn:count", &s, &len, &sub, &n, &i, &last))
683 return NULL;
684 if (last > len)
685 last = len;
686 if (last < 0)
687 last += len;
688 if (last < 0)
689 last = 0;
690 if (i < 0)
691 i += len;
692 if (i < 0)
693 i = 0;
694 m = last + 1 - n;
695 if (n == 0)
696 return PyInt_FromLong((long) (m-i));
697
698 r = 0;
699 while (i < m) {
700 if (!memcmp(s+i, sub, n)) {
701 r++;
702 i += n;
703 } else {
704 i++;
705 }
706 }
707 return PyInt_FromLong((long) r);
708 }
709
710
711 PyDoc_STRVAR(swapcase__doc__,
712 "swapcase(s) -> string\n"
713 "\n"
714 "Return a copy of the string s with upper case characters\n"
715 "converted to lowercase and vice versa.");
716
717 static PyObject *
718 strop_swapcase(PyObject *self, PyObject *args)
719 {
720 char *s, *s_new;
721 Py_ssize_t i, n;
722 PyObject *newstr;
723 int changed;
724
725 WARN;
726 if (PyString_AsStringAndSize(args, &s, &n))
727 return NULL;
728 newstr = PyString_FromStringAndSize(NULL, n);
729 if (newstr == NULL)
730 return NULL;
731 s_new = PyString_AsString(newstr);
732 changed = 0;
733 for (i = 0; i < n; i++) {
734 int c = Py_CHARMASK(*s++);
735 if (islower(c)) {
736 changed = 1;
737 *s_new = toupper(c);
738 }
739 else if (isupper(c)) {
740 changed = 1;
741 *s_new = tolower(c);
742 }
743 else
744 *s_new = c;
745 s_new++;
746 }
747 if (!changed) {
748 Py_DECREF(newstr);
749 Py_INCREF(args);
750 return args;
751 }
752 return newstr;
753 }
754
755
756 PyDoc_STRVAR(atoi__doc__,
757 "atoi(s [,base]) -> int\n"
758 "\n"
759 "Return the integer represented by the string s in the given\n"
760 "base, which defaults to 10. The string s must consist of one\n"
761 "or more digits, possibly preceded by a sign. If base is 0, it\n"
762 "is chosen from the leading characters of s, 0 for octal, 0x or\n"
763 "0X for hexadecimal. If base is 16, a preceding 0x or 0X is\n"
764 "accepted.");
765
766 static PyObject *
767 strop_atoi(PyObject *self, PyObject *args)
768 {
769 char *s, *end;
770 int base = 10;
771 long x;
772 char buffer[256]; /* For errors */
773
774 WARN;
775 if (!PyArg_ParseTuple(args, "s|i:atoi", &s, &base))
776 return NULL;
777
778 if ((base != 0 && base < 2) || base > 36) {
779 PyErr_SetString(PyExc_ValueError, "invalid base for atoi()");
780 return NULL;
781 }
782
783 while (*s && isspace(Py_CHARMASK(*s)))
784 s++;
785 errno = 0;
786 if (base == 0 && s[0] == '0')
787 x = (long) PyOS_strtoul(s, &end, base);
788 else
789 x = PyOS_strtol(s, &end, base);
790 if (end == s || !isalnum(Py_CHARMASK(end[-1])))
791 goto bad;
792 while (*end && isspace(Py_CHARMASK(*end)))
793 end++;
794 if (*end != '\0') {
795 bad:
796 PyOS_snprintf(buffer, sizeof(buffer),
797 "invalid literal for atoi(): %.200s", s);
798 PyErr_SetString(PyExc_ValueError, buffer);
799 return NULL;
800 }
801 else if (errno != 0) {
802 PyOS_snprintf(buffer, sizeof(buffer),
803 "atoi() literal too large: %.200s", s);
804 PyErr_SetString(PyExc_ValueError, buffer);
805 return NULL;
806 }
807 return PyInt_FromLong(x);
808 }
809
810
811 PyDoc_STRVAR(atol__doc__,
812 "atol(s [,base]) -> long\n"
813 "\n"
814 "Return the long integer represented by the string s in the\n"
815 "given base, which defaults to 10. The string s must consist\n"
816 "of one or more digits, possibly preceded by a sign. If base\n"
817 "is 0, it is chosen from the leading characters of s, 0 for\n"
818 "octal, 0x or 0X for hexadecimal. If base is 16, a preceding\n"
819 "0x or 0X is accepted. A trailing L or l is not accepted,\n"
820 "unless base is 0.");
821
822 static PyObject *
823 strop_atol(PyObject *self, PyObject *args)
824 {
825 char *s, *end;
826 int base = 10;
827 PyObject *x;
828 char buffer[256]; /* For errors */
829
830 WARN;
831 if (!PyArg_ParseTuple(args, "s|i:atol", &s, &base))
832 return NULL;
833
834 if ((base != 0 && base < 2) || base > 36) {
835 PyErr_SetString(PyExc_ValueError, "invalid base for atol()");
836 return NULL;
837 }
838
839 while (*s && isspace(Py_CHARMASK(*s)))
840 s++;
841 if (s[0] == '\0') {
842 PyErr_SetString(PyExc_ValueError, "empty string for atol()");
843 return NULL;
844 }
845 x = PyLong_FromString(s, &end, base);
846 if (x == NULL)
847 return NULL;
848 if (base == 0 && (*end == 'l' || *end == 'L'))
849 end++;
850 while (*end && isspace(Py_CHARMASK(*end)))
851 end++;
852 if (*end != '\0') {
853 PyOS_snprintf(buffer, sizeof(buffer),
854 "invalid literal for atol(): %.200s", s);
855 PyErr_SetString(PyExc_ValueError, buffer);
856 Py_DECREF(x);
857 return NULL;
858 }
859 return x;
860 }
861
862
863 PyDoc_STRVAR(atof__doc__,
864 "atof(s) -> float\n"
865 "\n"
866 "Return the floating point number represented by the string s.");
867
868 static PyObject *
869 strop_atof(PyObject *self, PyObject *args)
870 {
871 char *s, *end;
872 double x;
873 char buffer[256]; /* For errors */
874
875 WARN;
876 if (!PyArg_ParseTuple(args, "s:atof", &s))
877 return NULL;
878 while (*s && isspace(Py_CHARMASK(*s)))
879 s++;
880 if (s[0] == '\0') {
881 PyErr_SetString(PyExc_ValueError, "empty string for atof()");
882 return NULL;
883 }
884
885 PyFPE_START_PROTECT("strop_atof", return 0)
886 x = PyOS_string_to_double(s, &end, PyExc_OverflowError);
887 PyFPE_END_PROTECT(x)
888 if (x == -1 && PyErr_Occurred())
889 return NULL;
890 while (*end && isspace(Py_CHARMASK(*end)))
891 end++;
892 if (*end != '\0') {
893 PyOS_snprintf(buffer, sizeof(buffer),
894 "invalid literal for atof(): %.200s", s);
895 PyErr_SetString(PyExc_ValueError, buffer);
896 return NULL;
897 }
898 return PyFloat_FromDouble(x);
899 }
900
901
902 PyDoc_STRVAR(maketrans__doc__,
903 "maketrans(frm, to) -> string\n"
904 "\n"
905 "Return a translation table (a string of 256 bytes long)\n"
906 "suitable for use in string.translate. The strings frm and to\n"
907 "must be of the same length.");
908
909 static PyObject *
910 strop_maketrans(PyObject *self, PyObject *args)
911 {
912 unsigned char *c, *from=NULL, *to=NULL;
913 Py_ssize_t i, fromlen=0, tolen=0;
914 PyObject *result;
915
916 if (!PyArg_ParseTuple(args, "t#t#:maketrans", &from, &fromlen, &to, &tolen))
917 return NULL;
918
919 if (fromlen != tolen) {
920 PyErr_SetString(PyExc_ValueError,
921 "maketrans arguments must have same length");
922 return NULL;
923 }
924
925 result = PyString_FromStringAndSize((char *)NULL, 256);
926 if (result == NULL)
927 return NULL;
928 c = (unsigned char *) PyString_AS_STRING((PyStringObject *)result);
929 for (i = 0; i < 256; i++)
930 c[i]=(unsigned char)i;
931 for (i = 0; i < fromlen; i++)
932 c[from[i]]=to[i];
933
934 return result;
935 }
936
937
938 PyDoc_STRVAR(translate__doc__,
939 "translate(s,table [,deletechars]) -> string\n"
940 "\n"
941 "Return a copy of the string s, where all characters occurring\n"
942 "in the optional argument deletechars are removed, and the\n"
943 "remaining characters have been mapped through the given\n"
944 "translation table, which must be a string of length 256.");
945
946 static PyObject *
947 strop_translate(PyObject *self, PyObject *args)
948 {
949 register char *input, *table, *output;
950 Py_ssize_t i;
951 int c, changed = 0;
952 PyObject *input_obj;
953 char *table1, *output_start, *del_table=NULL;
954 Py_ssize_t inlen, tablen, dellen = 0;
955 PyObject *result;
956 int trans_table[256];
957
958 WARN;
959 if (!PyArg_ParseTuple(args, "St#|t#:translate", &input_obj,
960 &table1, &tablen, &del_table, &dellen))
961 return NULL;
962 if (tablen != 256) {
963 PyErr_SetString(PyExc_ValueError,
964 "translation table must be 256 characters long");
965 return NULL;
966 }
967
968 table = table1;
969 inlen = PyString_GET_SIZE(input_obj);
970 result = PyString_FromStringAndSize((char *)NULL, inlen);
971 if (result == NULL)
972 return NULL;
973 output_start = output = PyString_AsString(result);
974 input = PyString_AsString(input_obj);
975
976 if (dellen == 0) {
977 /* If no deletions are required, use faster code */
978 for (i = inlen; --i >= 0; ) {
979 c = Py_CHARMASK(*input++);
980 if (Py_CHARMASK((*output++ = table[c])) != c)
981 changed = 1;
982 }
983 if (changed)
984 return result;
985 Py_DECREF(result);
986 Py_INCREF(input_obj);
987 return input_obj;
988 }
989
990 for (i = 0; i < 256; i++)
991 trans_table[i] = Py_CHARMASK(table[i]);
992
993 for (i = 0; i < dellen; i++)
994 trans_table[(int) Py_CHARMASK(del_table[i])] = -1;
995
996 for (i = inlen; --i >= 0; ) {
997 c = Py_CHARMASK(*input++);
998 if (trans_table[c] != -1)
999 if (Py_CHARMASK(*output++ = (char)trans_table[c]) == c)
1000 continue;
1001 changed = 1;
1002 }
1003 if (!changed) {
1004 Py_DECREF(result);
1005 Py_INCREF(input_obj);
1006 return input_obj;
1007 }
1008 /* Fix the size of the resulting string */
1009 if (inlen > 0)
1010 _PyString_Resize(&result, output - output_start);
1011 return result;
1012 }
1013
1014
1015 /* What follows is used for implementing replace(). Perry Stoll. */
1016
1017 /*
1018 mymemfind
1019
1020 strstr replacement for arbitrary blocks of memory.
1021
1022 Locates the first occurrence in the memory pointed to by MEM of the
1023 contents of memory pointed to by PAT. Returns the index into MEM if
1024 found, or -1 if not found. If len of PAT is greater than length of
1025 MEM, the function returns -1.
1026 */
1027 static Py_ssize_t
1028 mymemfind(const char *mem, Py_ssize_t len, const char *pat, Py_ssize_t pat_len)
1029 {
1030 register Py_ssize_t ii;
1031
1032 /* pattern can not occur in the last pat_len-1 chars */
1033 len -= pat_len;
1034
1035 for (ii = 0; ii <= len; ii++) {
1036 if (mem[ii] == pat[0] &&
1037 (pat_len == 1 ||
1038 memcmp(&mem[ii+1], &pat[1], pat_len-1) == 0)) {
1039 return ii;
1040 }
1041 }
1042 return -1;
1043 }
1044
1045 /*
1046 mymemcnt
1047
1048 Return the number of distinct times PAT is found in MEM.
1049 meaning mem=1111 and pat==11 returns 2.
1050 mem=11111 and pat==11 also return 2.
1051 */
1052 static Py_ssize_t
1053 mymemcnt(const char *mem, Py_ssize_t len, const char *pat, Py_ssize_t pat_len)
1054 {
1055 register Py_ssize_t offset = 0;
1056 Py_ssize_t nfound = 0;
1057
1058 while (len >= 0) {
1059 offset = mymemfind(mem, len, pat, pat_len);
1060 if (offset == -1)
1061 break;
1062 mem += offset + pat_len;
1063 len -= offset + pat_len;
1064 nfound++;
1065 }
1066 return nfound;
1067 }
1068
1069 /*
1070 mymemreplace
1071
1072 Return a string in which all occurrences of PAT in memory STR are
1073 replaced with SUB.
1074
1075 If length of PAT is less than length of STR or there are no occurrences
1076 of PAT in STR, then the original string is returned. Otherwise, a new
1077 string is allocated here and returned.
1078
1079 on return, out_len is:
1080 the length of output string, or
1081 -1 if the input string is returned, or
1082 unchanged if an error occurs (no memory).
1083
1084 return value is:
1085 the new string allocated locally, or
1086 NULL if an error occurred.
1087 */
1088 static char *
1089 mymemreplace(const char *str, Py_ssize_t len, /* input string */
1090 const char *pat, Py_ssize_t pat_len, /* pattern string to find */
1091 const char *sub, Py_ssize_t sub_len, /* substitution string */
1092 Py_ssize_t count, /* number of replacements */
1093 Py_ssize_t *out_len)
1094 {
1095 char *out_s;
1096 char *new_s;
1097 Py_ssize_t nfound, offset, new_len;
1098
1099 if (len == 0 || pat_len > len)
1100 goto return_same;
1101
1102 /* find length of output string */
1103 nfound = mymemcnt(str, len, pat, pat_len);
1104 if (count < 0)
1105 count = PY_SSIZE_T_MAX;
1106 else if (nfound > count)
1107 nfound = count;
1108 if (nfound == 0)
1109 goto return_same;
1110
1111 new_len = len + nfound*(sub_len - pat_len);
1112 if (new_len == 0) {
1113 /* Have to allocate something for the caller to free(). */
1114 out_s = (char *)PyMem_MALLOC(1);
1115 if (out_s == NULL)
1116 return NULL;
1117 out_s[0] = '\0';
1118 }
1119 else {
1120 assert(new_len > 0);
1121 new_s = (char *)PyMem_MALLOC(new_len);
1122 if (new_s == NULL)
1123 return NULL;
1124 out_s = new_s;
1125
1126 for (; count > 0 && len > 0; --count) {
1127 /* find index of next instance of pattern */
1128 offset = mymemfind(str, len, pat, pat_len);
1129 if (offset == -1)
1130 break;
1131
1132 /* copy non matching part of input string */
1133 memcpy(new_s, str, offset);
1134 str += offset + pat_len;
1135 len -= offset + pat_len;
1136
1137 /* copy substitute into the output string */
1138 new_s += offset;
1139 memcpy(new_s, sub, sub_len);
1140 new_s += sub_len;
1141 }
1142 /* copy any remaining values into output string */
1143 if (len > 0)
1144 memcpy(new_s, str, len);
1145 }
1146 *out_len = new_len;
1147 return out_s;
1148
1149 return_same:
1150 *out_len = -1;
1151 return (char *)str; /* cast away const */
1152 }
1153
1154
1155 PyDoc_STRVAR(replace__doc__,
1156 "replace (str, old, new[, maxsplit]) -> string\n"
1157 "\n"
1158 "Return a copy of string str with all occurrences of substring\n"
1159 "old replaced by new. If the optional argument maxsplit is\n"
1160 "given, only the first maxsplit occurrences are replaced.");
1161
1162 static PyObject *
1163 strop_replace(PyObject *self, PyObject *args)
1164 {
1165 char *str, *pat,*sub,*new_s;
1166 Py_ssize_t len,pat_len,sub_len,out_len;
1167 Py_ssize_t count = -1;
1168 PyObject *newstr;
1169
1170 WARN;
1171 if (!PyArg_ParseTuple(args, "t#t#t#|n:replace",
1172 &str, &len, &pat, &pat_len, &sub, &sub_len,
1173 &count))
1174 return NULL;
1175 if (pat_len <= 0) {
1176 PyErr_SetString(PyExc_ValueError, "empty pattern string");
1177 return NULL;
1178 }
1179 /* CAUTION: strop treats a replace count of 0 as infinity, unlke
1180 * current (2.1) string.py and string methods. Preserve this for
1181 * ... well, hard to say for what <wink>.
1182 */
1183 if (count == 0)
1184 count = -1;
1185 new_s = mymemreplace(str,len,pat,pat_len,sub,sub_len,count,&out_len);
1186 if (new_s == NULL) {
1187 PyErr_NoMemory();
1188 return NULL;
1189 }
1190 if (out_len == -1) {
1191 /* we're returning another reference to the input string */
1192 newstr = PyTuple_GetItem(args, 0);
1193 Py_XINCREF(newstr);
1194 }
1195 else {
1196 newstr = PyString_FromStringAndSize(new_s, out_len);
1197 PyMem_FREE(new_s);
1198 }
1199 return newstr;
1200 }
1201
1202
1203 /* List of functions defined in the module */
1204
1205 static PyMethodDef
1206 strop_methods[] = {
1207 {"atof", strop_atof, METH_VARARGS, atof__doc__},
1208 {"atoi", strop_atoi, METH_VARARGS, atoi__doc__},
1209 {"atol", strop_atol, METH_VARARGS, atol__doc__},
1210 {"capitalize", strop_capitalize, METH_O, capitalize__doc__},
1211 {"count", strop_count, METH_VARARGS, count__doc__},
1212 {"expandtabs", strop_expandtabs, METH_VARARGS, expandtabs__doc__},
1213 {"find", strop_find, METH_VARARGS, find__doc__},
1214 {"join", strop_joinfields, METH_VARARGS, joinfields__doc__},
1215 {"joinfields", strop_joinfields, METH_VARARGS, joinfields__doc__},
1216 {"lstrip", strop_lstrip, METH_O, lstrip__doc__},
1217 {"lower", strop_lower, METH_O, lower__doc__},
1218 {"maketrans", strop_maketrans, METH_VARARGS, maketrans__doc__},
1219 {"replace", strop_replace, METH_VARARGS, replace__doc__},
1220 {"rfind", strop_rfind, METH_VARARGS, rfind__doc__},
1221 {"rstrip", strop_rstrip, METH_O, rstrip__doc__},
1222 {"split", strop_splitfields, METH_VARARGS, splitfields__doc__},
1223 {"splitfields", strop_splitfields, METH_VARARGS, splitfields__doc__},
1224 {"strip", strop_strip, METH_O, strip__doc__},
1225 {"swapcase", strop_swapcase, METH_O, swapcase__doc__},
1226 {"translate", strop_translate, METH_VARARGS, translate__doc__},
1227 {"upper", strop_upper, METH_O, upper__doc__},
1228 {NULL, NULL} /* sentinel */
1229 };
1230
1231
1232 PyMODINIT_FUNC
1233 initstrop(void)
1234 {
1235 PyObject *m, *s;
1236 char buf[256];
1237 int c, n;
1238 m = Py_InitModule4("strop", strop_methods, strop_module__doc__,
1239 (PyObject*)NULL, PYTHON_API_VERSION);
1240 if (m == NULL)
1241 return;
1242
1243 /* Create 'whitespace' object */
1244 n = 0;
1245 for (c = 0; c < 256; c++) {
1246 if (isspace(c))
1247 buf[n++] = c;
1248 }
1249 s = PyString_FromStringAndSize(buf, n);
1250 if (s)
1251 PyModule_AddObject(m, "whitespace", s);
1252
1253 /* Create 'lowercase' object */
1254 n = 0;
1255 for (c = 0; c < 256; c++) {
1256 if (islower(c))
1257 buf[n++] = c;
1258 }
1259 s = PyString_FromStringAndSize(buf, n);
1260 if (s)
1261 PyModule_AddObject(m, "lowercase", s);
1262
1263 /* Create 'uppercase' object */
1264 n = 0;
1265 for (c = 0; c < 256; c++) {
1266 if (isupper(c))
1267 buf[n++] = c;
1268 }
1269 s = PyString_FromStringAndSize(buf, n);
1270 if (s)
1271 PyModule_AddObject(m, "uppercase", s);
1272 }