/[pcre]/code/trunk/pcrecpp.cc
ViewVC logotype

Contents of /code/trunk/pcrecpp.cc

Parent Directory Parent Directory | Revision Log Revision Log


Revision 77 - (show annotations) (download)
Sat Feb 24 21:40:45 2007 UTC (7 years, 2 months ago) by nigel
File size: 26432 byte(s)
Load pcre-6.0 into code/trunk.

1 // Copyright (c) 2005, Google Inc.
2 // All rights reserved.
3 //
4 // Redistribution and use in source and binary forms, with or without
5 // modification, are permitted provided that the following conditions are
6 // met:
7 //
8 // * Redistributions of source code must retain the above copyright
9 // notice, this list of conditions and the following disclaimer.
10 // * Redistributions in binary form must reproduce the above
11 // copyright notice, this list of conditions and the following disclaimer
12 // in the documentation and/or other materials provided with the
13 // distribution.
14 // * Neither the name of Google Inc. nor the names of its
15 // contributors may be used to endorse or promote products derived from
16 // this software without specific prior written permission.
17 //
18 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
19 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
20 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
21 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
22 // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
23 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
24 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
25 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
26 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
27 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
28 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 //
30 // Author: Sanjay Ghemawat
31
32 #include <stdlib.h>
33 #include <stdio.h>
34 #include <ctype.h>
35 #include <limits.h> /* for SHRT_MIN, USHRT_MAX, etc */
36 #include <assert.h>
37 #include <errno.h>
38 #include <string>
39 #include "config.h"
40 // We need this to compile the proper dll on windows/msys. This is copied
41 // from pcre_internal.h. It would probably be better just to include that.
42 #define PCRE_DEFINITION /* Win32 __declspec(export) trigger for .dll */
43 #include "pcre.h"
44 #include "pcre_stringpiece.h"
45 #include "pcrecpp.h"
46
47
48 namespace pcrecpp {
49
50 // Maximum number of args we can set
51 static const int kMaxArgs = 16;
52 static const int kVecSize = (1 + kMaxArgs) * 3; // results + PCRE workspace
53
54 // Special object that stands-in for no argument
55 Arg no_arg((void*)NULL);
56
57 // If a regular expression has no error, its error_ field points here
58 static const string empty_string;
59
60 // If the user doesn't ask for any options, we just use this one
61 static RE_Options default_options;
62
63 void RE::Init(const char* pat, const RE_Options* options) {
64 pattern_ = pat;
65 if (options == NULL) {
66 options_ = default_options;
67 } else {
68 options_ = *options;
69 }
70 error_ = &empty_string;
71 re_full_ = NULL;
72 re_partial_ = NULL;
73
74 re_partial_ = Compile(UNANCHORED);
75 if (re_partial_ != NULL) {
76 // Check for complicated patterns. The following change is
77 // conservative in that it may treat some "simple" patterns
78 // as "complex" (e.g., if the vertical bar is in a character
79 // class or is escaped). But it seems good enough.
80 if (strchr(pat, '|') == NULL) {
81 // Simple pattern: we can use position-based checks to perform
82 // fully anchored matches
83 re_full_ = re_partial_;
84 } else {
85 // We need a special pattern for anchored matches
86 re_full_ = Compile(ANCHOR_BOTH);
87 }
88 }
89 }
90
91 RE::~RE() {
92 if (re_full_ != NULL && re_full_ != re_partial_) (*pcre_free)(re_full_);
93 if (re_partial_ != NULL) (*pcre_free)(re_partial_);
94 if (error_ != &empty_string) delete error_;
95 }
96
97 pcre* RE::Compile(Anchor anchor) {
98 // First, convert RE_Options into pcre options
99 int pcre_options = 0;
100 if (options_.utf8())
101 pcre_options |= PCRE_UTF8;
102
103 // Special treatment for anchoring. This is needed because at
104 // runtime pcre only provides an option for anchoring at the
105 // beginning of a string (unless you use offset).
106 //
107 // There are three types of anchoring we want:
108 // UNANCHORED Compile the original pattern, and use
109 // a pcre unanchored match.
110 // ANCHOR_START Compile the original pattern, and use
111 // a pcre anchored match.
112 // ANCHOR_BOTH Tack a "\z" to the end of the original pattern
113 // and use a pcre anchored match.
114
115 const char* compile_error;
116 int eoffset;
117 pcre* re;
118 if (anchor != ANCHOR_BOTH) {
119 re = pcre_compile(pattern_.c_str(), pcre_options,
120 &compile_error, &eoffset, NULL);
121 } else {
122 // Tack a '\z' at the end of RE. Parenthesize it first so that
123 // the '\z' applies to all top-level alternatives in the regexp.
124 string wrapped = "(?:"; // A non-counting grouping operator
125 wrapped += pattern_;
126 wrapped += ")\\z";
127 re = pcre_compile(wrapped.c_str(), pcre_options,
128 &compile_error, &eoffset, NULL);
129 }
130 if (re == NULL) {
131 if (error_ == &empty_string) error_ = new string(compile_error);
132 }
133 return re;
134 }
135
136 /***** Matching interfaces *****/
137
138 bool RE::FullMatch(const StringPiece& text,
139 const Arg& ptr1,
140 const Arg& ptr2,
141 const Arg& ptr3,
142 const Arg& ptr4,
143 const Arg& ptr5,
144 const Arg& ptr6,
145 const Arg& ptr7,
146 const Arg& ptr8,
147 const Arg& ptr9,
148 const Arg& ptr10,
149 const Arg& ptr11,
150 const Arg& ptr12,
151 const Arg& ptr13,
152 const Arg& ptr14,
153 const Arg& ptr15,
154 const Arg& ptr16) const {
155 const Arg* args[kMaxArgs];
156 int n = 0;
157 if (&ptr1 == &no_arg) goto done; args[n++] = &ptr1;
158 if (&ptr2 == &no_arg) goto done; args[n++] = &ptr2;
159 if (&ptr3 == &no_arg) goto done; args[n++] = &ptr3;
160 if (&ptr4 == &no_arg) goto done; args[n++] = &ptr4;
161 if (&ptr5 == &no_arg) goto done; args[n++] = &ptr5;
162 if (&ptr6 == &no_arg) goto done; args[n++] = &ptr6;
163 if (&ptr7 == &no_arg) goto done; args[n++] = &ptr7;
164 if (&ptr8 == &no_arg) goto done; args[n++] = &ptr8;
165 if (&ptr9 == &no_arg) goto done; args[n++] = &ptr9;
166 if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
167 if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
168 if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
169 if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
170 if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
171 if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
172 if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
173 done:
174
175 int consumed;
176 int vec[kVecSize];
177 return DoMatchImpl(text, ANCHOR_BOTH, &consumed, args, n, vec, kVecSize);
178 }
179
180 bool RE::PartialMatch(const StringPiece& text,
181 const Arg& ptr1,
182 const Arg& ptr2,
183 const Arg& ptr3,
184 const Arg& ptr4,
185 const Arg& ptr5,
186 const Arg& ptr6,
187 const Arg& ptr7,
188 const Arg& ptr8,
189 const Arg& ptr9,
190 const Arg& ptr10,
191 const Arg& ptr11,
192 const Arg& ptr12,
193 const Arg& ptr13,
194 const Arg& ptr14,
195 const Arg& ptr15,
196 const Arg& ptr16) const {
197 const Arg* args[kMaxArgs];
198 int n = 0;
199 if (&ptr1 == &no_arg) goto done; args[n++] = &ptr1;
200 if (&ptr2 == &no_arg) goto done; args[n++] = &ptr2;
201 if (&ptr3 == &no_arg) goto done; args[n++] = &ptr3;
202 if (&ptr4 == &no_arg) goto done; args[n++] = &ptr4;
203 if (&ptr5 == &no_arg) goto done; args[n++] = &ptr5;
204 if (&ptr6 == &no_arg) goto done; args[n++] = &ptr6;
205 if (&ptr7 == &no_arg) goto done; args[n++] = &ptr7;
206 if (&ptr8 == &no_arg) goto done; args[n++] = &ptr8;
207 if (&ptr9 == &no_arg) goto done; args[n++] = &ptr9;
208 if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
209 if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
210 if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
211 if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
212 if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
213 if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
214 if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
215 done:
216
217 int consumed;
218 int vec[kVecSize];
219 return DoMatchImpl(text, UNANCHORED, &consumed, args, n, vec, kVecSize);
220 }
221
222 bool RE::Consume(StringPiece* input,
223 const Arg& ptr1,
224 const Arg& ptr2,
225 const Arg& ptr3,
226 const Arg& ptr4,
227 const Arg& ptr5,
228 const Arg& ptr6,
229 const Arg& ptr7,
230 const Arg& ptr8,
231 const Arg& ptr9,
232 const Arg& ptr10,
233 const Arg& ptr11,
234 const Arg& ptr12,
235 const Arg& ptr13,
236 const Arg& ptr14,
237 const Arg& ptr15,
238 const Arg& ptr16) const {
239 const Arg* args[kMaxArgs];
240 int n = 0;
241 if (&ptr1 == &no_arg) goto done; args[n++] = &ptr1;
242 if (&ptr2 == &no_arg) goto done; args[n++] = &ptr2;
243 if (&ptr3 == &no_arg) goto done; args[n++] = &ptr3;
244 if (&ptr4 == &no_arg) goto done; args[n++] = &ptr4;
245 if (&ptr5 == &no_arg) goto done; args[n++] = &ptr5;
246 if (&ptr6 == &no_arg) goto done; args[n++] = &ptr6;
247 if (&ptr7 == &no_arg) goto done; args[n++] = &ptr7;
248 if (&ptr8 == &no_arg) goto done; args[n++] = &ptr8;
249 if (&ptr9 == &no_arg) goto done; args[n++] = &ptr9;
250 if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
251 if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
252 if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
253 if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
254 if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
255 if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
256 if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
257 done:
258
259 int consumed;
260 int vec[kVecSize];
261 if (DoMatchImpl(*input, ANCHOR_START, &consumed,
262 args, n, vec, kVecSize)) {
263 input->remove_prefix(consumed);
264 return true;
265 } else {
266 return false;
267 }
268 }
269
270 bool RE::FindAndConsume(StringPiece* input,
271 const Arg& ptr1,
272 const Arg& ptr2,
273 const Arg& ptr3,
274 const Arg& ptr4,
275 const Arg& ptr5,
276 const Arg& ptr6,
277 const Arg& ptr7,
278 const Arg& ptr8,
279 const Arg& ptr9,
280 const Arg& ptr10,
281 const Arg& ptr11,
282 const Arg& ptr12,
283 const Arg& ptr13,
284 const Arg& ptr14,
285 const Arg& ptr15,
286 const Arg& ptr16) const {
287 const Arg* args[kMaxArgs];
288 int n = 0;
289 if (&ptr1 == &no_arg) goto done; args[n++] = &ptr1;
290 if (&ptr2 == &no_arg) goto done; args[n++] = &ptr2;
291 if (&ptr3 == &no_arg) goto done; args[n++] = &ptr3;
292 if (&ptr4 == &no_arg) goto done; args[n++] = &ptr4;
293 if (&ptr5 == &no_arg) goto done; args[n++] = &ptr5;
294 if (&ptr6 == &no_arg) goto done; args[n++] = &ptr6;
295 if (&ptr7 == &no_arg) goto done; args[n++] = &ptr7;
296 if (&ptr8 == &no_arg) goto done; args[n++] = &ptr8;
297 if (&ptr9 == &no_arg) goto done; args[n++] = &ptr9;
298 if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
299 if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
300 if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
301 if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
302 if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
303 if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
304 if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
305 done:
306
307 int consumed;
308 int vec[kVecSize];
309 if (DoMatchImpl(*input, UNANCHORED, &consumed,
310 args, n, vec, kVecSize)) {
311 input->remove_prefix(consumed);
312 return true;
313 } else {
314 return false;
315 }
316 }
317
318 bool RE::Replace(const StringPiece& rewrite,
319 string *str) const {
320 int vec[kVecSize];
321 int matches = TryMatch(*str, 0, UNANCHORED, vec, kVecSize);
322 if (matches == 0)
323 return false;
324
325 string s;
326 if (!Rewrite(&s, rewrite, *str, vec, matches))
327 return false;
328
329 assert(vec[0] >= 0);
330 assert(vec[1] >= 0);
331 str->replace(vec[0], vec[1] - vec[0], s);
332 return true;
333 }
334
335 int RE::GlobalReplace(const StringPiece& rewrite,
336 string *str) const {
337 int count = 0;
338 int vec[kVecSize];
339 string out;
340 int start = 0;
341 int lastend = -1;
342
343 for (; start <= static_cast<int>(str->length()); count++) {
344 int matches = TryMatch(*str, start, UNANCHORED, vec, kVecSize);
345 if (matches <= 0)
346 break;
347 int matchstart = vec[0], matchend = vec[1];
348 assert(matchstart >= start);
349 assert(matchend >= matchstart);
350 if (matchstart == matchend && matchstart == lastend) {
351 // advance one character if we matched an empty string at the same
352 // place as the last match occurred
353 if (start < static_cast<int>(str->length()))
354 out.push_back((*str)[start]);
355 start++;
356 } else {
357 out.append(*str, start, matchstart - start);
358 Rewrite(&out, rewrite, *str, vec, matches);
359 start = matchend;
360 lastend = matchend;
361 count++;
362 }
363 }
364
365 if (count == 0)
366 return 0;
367
368 if (start < static_cast<int>(str->length()))
369 out.append(*str, start, str->length() - start);
370 swap(out, *str);
371 return count;
372 }
373
374 bool RE::Extract(const StringPiece& rewrite,
375 const StringPiece& text,
376 string *out) const {
377 int vec[kVecSize];
378 int matches = TryMatch(text, 0, UNANCHORED, vec, kVecSize);
379 if (matches == 0)
380 return false;
381 out->clear();
382 return Rewrite(out, rewrite, text, vec, matches);
383 }
384
385 /***** Actual matching and rewriting code *****/
386
387 int RE::TryMatch(const StringPiece& text,
388 int startpos,
389 Anchor anchor,
390 int *vec,
391 int vecsize) const {
392 pcre* re = (anchor == ANCHOR_BOTH) ? re_full_ : re_partial_;
393 if (re == NULL) {
394 //fprintf(stderr, "Matching against invalid re: %s\n", error_->c_str());
395 return 0;
396 }
397
398 pcre_extra extra = { 0 };
399 if (options_.match_limit() > 0) {
400 extra.flags = PCRE_EXTRA_MATCH_LIMIT;
401 extra.match_limit = options_.match_limit();
402 }
403 int rc = pcre_exec(re, // The regular expression object
404 &extra,
405 text.data(),
406 text.size(),
407 startpos,
408 (anchor == UNANCHORED) ? 0 : PCRE_ANCHORED,
409 vec,
410 vecsize);
411
412 // Handle errors
413 if (rc == PCRE_ERROR_NOMATCH) {
414 return 0;
415 } else if (rc < 0) {
416 //fprintf(stderr, "Unexpected return code: %d when matching '%s'\n",
417 // re, pattern_.c_str());
418 return 0;
419 } else if (rc == 0) {
420 // pcre_exec() returns 0 as a special case when the number of
421 // capturing subpatterns exceeds the size of the vector.
422 // When this happens, there is a match and the output vector
423 // is filled, but we miss out on the positions of the extra subpatterns.
424 rc = vecsize / 2;
425 }
426
427 if ((anchor == ANCHOR_BOTH) && (re_full_ == re_partial_)) {
428 // We need an extra check to make sure that the match extended
429 // to the end of the input string
430 assert(vec[0] == 0); // PCRE_ANCHORED forces starting match
431 if (vec[1] != text.size()) return 0; // Did not get ending match
432 }
433
434 return rc;
435 }
436
437 bool RE::DoMatchImpl(const StringPiece& text,
438 Anchor anchor,
439 int* consumed,
440 const Arg* const* args,
441 int n,
442 int* vec,
443 int vecsize) const {
444 assert((1 + n) * 3 <= vecsize); // results + PCRE workspace
445 int matches = TryMatch(text, 0, anchor, vec, vecsize);
446 assert(matches >= 0); // TryMatch never returns negatives
447 if (matches == 0)
448 return false;
449
450 *consumed = vec[1];
451
452 if (args == NULL) {
453 // We are not interested in results
454 return true;
455 }
456
457 // If we got here, we must have matched the whole pattern.
458 // We do not need (can not do) any more checks on the value of 'matches' here
459 // -- see the comment for TryMatch.
460 for (int i = 0; i < n; i++) {
461 const int start = vec[2*(i+1)];
462 const int limit = vec[2*(i+1)+1];
463 if (!args[i]->Parse(text.data() + start, limit-start)) {
464 // TODO: Should we indicate what the error was?
465 return false;
466 }
467 }
468
469 return true;
470 }
471
472 bool RE::DoMatch(const StringPiece& text,
473 Anchor anchor,
474 int* consumed,
475 const Arg* const args[],
476 int n) const {
477 assert(n >= 0);
478 size_t const vecsize = (1 + n) * 3; // results + PCRE workspace
479 // (as for kVecSize)
480 int space[21]; // use stack allocation for small vecsize (common case)
481 int* vec = vecsize <= 21 ? space : new int[vecsize];
482 bool retval = DoMatchImpl(text, anchor, consumed, args, n, vec, vecsize);
483 if (vec != space) delete [] vec;
484 return retval;
485 }
486
487 bool RE::Rewrite(string *out, const StringPiece &rewrite,
488 const StringPiece &text, int *vec, int veclen) const {
489 for (const char *s = rewrite.data(), *end = s + rewrite.size();
490 s < end; s++) {
491 int c = *s;
492 if (c == '\\') {
493 c = *++s;
494 if (isdigit(c)) {
495 int n = (c - '0');
496 if (n >= veclen) {
497 //fprintf(stderr, requested group %d in regexp %.*s\n",
498 // n, rewrite.size(), rewrite.data());
499 return false;
500 }
501 int start = vec[2 * n];
502 if (start >= 0)
503 out->append(text.data() + start, vec[2 * n + 1] - start);
504 } else if (c == '\\') {
505 out->push_back('\\');
506 } else {
507 //fprintf(stderr, "invalid rewrite pattern: %.*s\n",
508 // rewrite.size(), rewrite.data());
509 return false;
510 }
511 } else {
512 out->push_back(c);
513 }
514 }
515 return true;
516 }
517
518 // Return the number of capturing subpatterns, or -1 if the
519 // regexp wasn't valid on construction.
520 int RE::NumberOfCapturingGroups() {
521 if (re_partial_ == NULL) return -1;
522
523 int result;
524 int pcre_retval = pcre_fullinfo(re_partial_, // The regular expression object
525 NULL, // We did not study the pattern
526 PCRE_INFO_CAPTURECOUNT,
527 &result);
528 assert(pcre_retval == 0);
529 return result;
530 }
531
532 /***** Parsers for various types *****/
533
534 bool Arg::parse_null(const char* str, int n, void* dest) {
535 // We fail if somebody asked us to store into a non-NULL void* pointer
536 return (dest == NULL);
537 }
538
539 bool Arg::parse_string(const char* str, int n, void* dest) {
540 reinterpret_cast<string*>(dest)->assign(str, n);
541 return true;
542 }
543
544 bool Arg::parse_stringpiece(const char* str, int n, void* dest) {
545 reinterpret_cast<StringPiece*>(dest)->set(str, n);
546 return true;
547 }
548
549 bool Arg::parse_char(const char* str, int n, void* dest) {
550 if (n != 1) return false;
551 *(reinterpret_cast<char*>(dest)) = str[0];
552 return true;
553 }
554
555 bool Arg::parse_uchar(const char* str, int n, void* dest) {
556 if (n != 1) return false;
557 *(reinterpret_cast<unsigned char*>(dest)) = str[0];
558 return true;
559 }
560
561 // Largest number spec that we are willing to parse
562 static const int kMaxNumberLength = 32;
563
564 // REQUIRES "buf" must have length at least kMaxNumberLength+1
565 // REQUIRES "n > 0"
566 // Copies "str" into "buf" and null-terminates if necessary.
567 // Returns one of:
568 // a. "str" if no termination is needed
569 // b. "buf" if the string was copied and null-terminated
570 // c. "" if the input was invalid and has no hope of being parsed
571 static const char* TerminateNumber(char* buf, const char* str, int n) {
572 if ((n > 0) && isspace(*str)) {
573 // We are less forgiving than the strtoxxx() routines and do not
574 // allow leading spaces.
575 return "";
576 }
577
578 // See if the character right after the input text may potentially
579 // look like a digit.
580 if (isdigit(str[n]) ||
581 ((str[n] >= 'a') && (str[n] <= 'f')) ||
582 ((str[n] >= 'A') && (str[n] <= 'F'))) {
583 if (n > kMaxNumberLength) return ""; // Input too big to be a valid number
584 memcpy(buf, str, n);
585 buf[n] = '\0';
586 return buf;
587 } else {
588 // We can parse right out of the supplied string, so return it.
589 return str;
590 }
591 }
592
593 bool Arg::parse_long_radix(const char* str,
594 int n,
595 void* dest,
596 int radix) {
597 if (n == 0) return false;
598 char buf[kMaxNumberLength+1];
599 str = TerminateNumber(buf, str, n);
600 char* end;
601 errno = 0;
602 long r = strtol(str, &end, radix);
603 if (end != str + n) return false; // Leftover junk
604 if (errno) return false;
605 *(reinterpret_cast<long*>(dest)) = r;
606 return true;
607 }
608
609 bool Arg::parse_ulong_radix(const char* str,
610 int n,
611 void* dest,
612 int radix) {
613 if (n == 0) return false;
614 char buf[kMaxNumberLength+1];
615 str = TerminateNumber(buf, str, n);
616 char* end;
617 errno = 0;
618 unsigned long r = strtoul(str, &end, radix);
619 if (end != str + n) return false; // Leftover junk
620 if (errno) return false;
621 *(reinterpret_cast<unsigned long*>(dest)) = r;
622 return true;
623 }
624
625 bool Arg::parse_short_radix(const char* str,
626 int n,
627 void* dest,
628 int radix) {
629 long r;
630 if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
631 if (r < SHRT_MIN || r > SHRT_MAX) return false; // Out of range
632 *(reinterpret_cast<short*>(dest)) = r;
633 return true;
634 }
635
636 bool Arg::parse_ushort_radix(const char* str,
637 int n,
638 void* dest,
639 int radix) {
640 unsigned long r;
641 if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
642 if (r > USHRT_MAX) return false; // Out of range
643 *(reinterpret_cast<unsigned short*>(dest)) = r;
644 return true;
645 }
646
647 bool Arg::parse_int_radix(const char* str,
648 int n,
649 void* dest,
650 int radix) {
651 long r;
652 if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
653 if (r < INT_MIN || r > INT_MAX) return false; // Out of range
654 *(reinterpret_cast<int*>(dest)) = r;
655 return true;
656 }
657
658 bool Arg::parse_uint_radix(const char* str,
659 int n,
660 void* dest,
661 int radix) {
662 unsigned long r;
663 if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
664 if (r > UINT_MAX) return false; // Out of range
665 *(reinterpret_cast<unsigned int*>(dest)) = r;
666 return true;
667 }
668
669 bool Arg::parse_longlong_radix(const char* str,
670 int n,
671 void* dest,
672 int radix) {
673 #ifndef HAVE_LONG_LONG
674 return false;
675 #else
676 if (n == 0) return false;
677 char buf[kMaxNumberLength+1];
678 str = TerminateNumber(buf, str, n);
679 char* end;
680 errno = 0;
681 #if defined HAVE_STRTOQ
682 long long r = strtoq(str, &end, radix);
683 #elif defined HAVE_STRTOLL
684 long long r = strtoll(str, &end, radix);
685 #else
686 #error parse_longlong_radix: cannot convert input to a long-long
687 #endif
688 if (end != str + n) return false; // Leftover junk
689 if (errno) return false;
690 *(reinterpret_cast<long long*>(dest)) = r;
691 return true;
692 #endif /* HAVE_LONG_LONG */
693 }
694
695 bool Arg::parse_ulonglong_radix(const char* str,
696 int n,
697 void* dest,
698 int radix) {
699 #ifndef HAVE_UNSIGNED_LONG_LONG
700 return false;
701 #else
702 if (n == 0) return false;
703 char buf[kMaxNumberLength+1];
704 str = TerminateNumber(buf, str, n);
705 char* end;
706 errno = 0;
707 #if defined HAVE_STRTOQ
708 unsigned long long r = strtouq(str, &end, radix);
709 #elif defined HAVE_STRTOLL
710 unsigned long long r = strtoull(str, &end, radix);
711 #else
712 #error parse_ulonglong_radix: cannot convert input to a long-long
713 #endif
714 if (end != str + n) return false; // Leftover junk
715 if (errno) return false;
716 *(reinterpret_cast<unsigned long long*>(dest)) = r;
717 return true;
718 #endif /* HAVE_UNSIGNED_LONG_LONG */
719 }
720
721 bool Arg::parse_double(const char* str, int n, void* dest) {
722 if (n == 0) return false;
723 static const int kMaxLength = 200;
724 char buf[kMaxLength];
725 if (n >= kMaxLength) return false;
726 memcpy(buf, str, n);
727 buf[n] = '\0';
728 errno = 0;
729 char* end;
730 double r = strtod(buf, &end);
731 if (end != buf + n) return false; // Leftover junk
732 if (errno) return false;
733 *(reinterpret_cast<double*>(dest)) = r;
734 return true;
735 }
736
737 bool Arg::parse_float(const char* str, int n, void* dest) {
738 double r;
739 if (!parse_double(str, n, &r)) return false;
740 *(reinterpret_cast<float*>(dest)) = static_cast<float>(r);
741 return true;
742 }
743
744
745 #define DEFINE_INTEGER_PARSERS(name) \
746 bool Arg::parse_##name(const char* str, int n, void* dest) { \
747 return parse_##name##_radix(str, n, dest, 10); \
748 } \
749 bool Arg::parse_##name##_hex(const char* str, int n, void* dest) { \
750 return parse_##name##_radix(str, n, dest, 16); \
751 } \
752 bool Arg::parse_##name##_octal(const char* str, int n, void* dest) { \
753 return parse_##name##_radix(str, n, dest, 8); \
754 } \
755 bool Arg::parse_##name##_cradix(const char* str, int n, void* dest) { \
756 return parse_##name##_radix(str, n, dest, 0); \
757 }
758
759 DEFINE_INTEGER_PARSERS(short);
760 DEFINE_INTEGER_PARSERS(ushort);
761 DEFINE_INTEGER_PARSERS(int);
762 DEFINE_INTEGER_PARSERS(uint);
763 DEFINE_INTEGER_PARSERS(long);
764 DEFINE_INTEGER_PARSERS(ulong);
765 DEFINE_INTEGER_PARSERS(longlong);
766 DEFINE_INTEGER_PARSERS(ulonglong);
767
768 #undef DEFINE_INTEGER_PARSERS
769
770 } // namespace pcrecpp

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12