1 |
// Copyright (c) 2005, Google Inc. |
// -*- coding: utf-8 -*- |
2 |
|
// |
3 |
|
// Copyright (c) 2005 - 2006, Google Inc. |
4 |
// All rights reserved. |
// All rights reserved. |
5 |
// |
// |
6 |
// Redistribution and use in source and binary forms, with or without |
// Redistribution and use in source and binary forms, with or without |
33 |
// |
// |
34 |
// TODO: Test extractions for PartialMatch/Consume |
// TODO: Test extractions for PartialMatch/Consume |
35 |
|
|
36 |
|
#ifdef HAVE_CONFIG_H |
37 |
|
#include <config.h> |
38 |
|
#endif |
39 |
|
|
40 |
|
#ifdef _WIN32 |
41 |
|
#define snprintf _snprintf |
42 |
|
#endif |
43 |
|
|
44 |
#include <stdio.h> |
#include <stdio.h> |
45 |
|
#include <cassert> |
46 |
#include <vector> |
#include <vector> |
|
#include "config.h" |
|
47 |
#include "pcrecpp.h" |
#include "pcrecpp.h" |
48 |
|
|
49 |
using pcrecpp::StringPiece; |
using pcrecpp::StringPiece; |
53 |
using pcrecpp::Octal; |
using pcrecpp::Octal; |
54 |
using pcrecpp::CRadix; |
using pcrecpp::CRadix; |
55 |
|
|
56 |
|
static bool VERBOSE_TEST = false; |
57 |
|
|
58 |
// CHECK dies with a fatal error if condition is not true. It is *not* |
// CHECK dies with a fatal error if condition is not true. It is *not* |
59 |
// controlled by NDEBUG, so the check will be executed regardless of |
// controlled by NDEBUG, so the check will be executed regardless of |
60 |
// compilation mode. Therefore, it is safe to do things like: |
// compilation mode. Therefore, it is safe to do things like: |
269 |
"aaaaa", |
"aaaaa", |
270 |
"bbaaaaa", |
"bbaaaaa", |
271 |
"bbabbabbabbabbabb" }, |
"bbabbabbabbabbabb" }, |
272 |
|
{ "b*", |
273 |
|
"bb", |
274 |
|
"aa\naa\n", |
275 |
|
"bbaa\naa\n", |
276 |
|
"bbabbabb\nbbabbabb\nbb" }, |
277 |
|
{ "b*", |
278 |
|
"bb", |
279 |
|
"aa\raa\r", |
280 |
|
"bbaa\raa\r", |
281 |
|
"bbabbabb\rbbabbabb\rbb" }, |
282 |
|
{ "b*", |
283 |
|
"bb", |
284 |
|
"aa\r\naa\r\n", |
285 |
|
"bbaa\r\naa\r\n", |
286 |
|
"bbabbabb\r\nbbabbabb\r\nbb" }, |
287 |
|
#ifdef SUPPORT_UTF8 |
288 |
|
{ "b*", |
289 |
|
"bb", |
290 |
|
"\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8", // utf8 |
291 |
|
"bb\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8", |
292 |
|
"bb\xE3\x83\x9B""bb""\xE3\x83\xBC""bb""\xE3\x83\xA0""bb""\xE3\x81\xB8""bb" }, |
293 |
|
{ "b*", |
294 |
|
"bb", |
295 |
|
"\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n", // utf8 |
296 |
|
"bb\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n", |
297 |
|
("bb\xE3\x83\x9B""bb\r\nbb""\xE3\x83\xBC""bb\rbb""\xE3\x83\xA0" |
298 |
|
"bb\nbb""\xE3\x81\xB8""bb\r\nbb") }, |
299 |
|
#endif |
300 |
{ "", NULL, NULL, NULL, NULL } |
{ "", NULL, NULL, NULL, NULL } |
301 |
}; |
}; |
302 |
|
|
303 |
|
#ifdef SUPPORT_UTF8 |
304 |
|
const bool support_utf8 = true; |
305 |
|
#else |
306 |
|
const bool support_utf8 = false; |
307 |
|
#endif |
308 |
|
|
309 |
for (const ReplaceTest *t = tests; t->original != NULL; ++t) { |
for (const ReplaceTest *t = tests; t->original != NULL; ++t) { |
310 |
|
RE re(t->regexp, RE_Options(PCRE_NEWLINE_CRLF).set_utf8(support_utf8)); |
311 |
|
assert(re.error().empty()); |
312 |
string one(t->original); |
string one(t->original); |
313 |
CHECK(RE(t->regexp).Replace(t->rewrite, &one)); |
CHECK(re.Replace(t->rewrite, &one)); |
314 |
CHECK_EQ(one, t->single); |
CHECK_EQ(one, t->single); |
315 |
string all(t->original); |
string all(t->original); |
316 |
CHECK(RE(t->regexp).GlobalReplace(t->rewrite, &all) > 0); |
CHECK(re.GlobalReplace(t->rewrite, &all) > 0); |
317 |
CHECK_EQ(all, t->global); |
CHECK_EQ(all, t->global); |
318 |
} |
} |
319 |
|
|
320 |
|
// One final test: test \r\n replacement when we're not in CRLF mode |
321 |
|
{ |
322 |
|
RE re("b*", RE_Options(PCRE_NEWLINE_CR).set_utf8(support_utf8)); |
323 |
|
assert(re.error().empty()); |
324 |
|
string all("aa\r\naa\r\n"); |
325 |
|
CHECK(re.GlobalReplace("bb", &all) > 0); |
326 |
|
CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb")); |
327 |
|
} |
328 |
|
{ |
329 |
|
RE re("b*", RE_Options(PCRE_NEWLINE_LF).set_utf8(support_utf8)); |
330 |
|
assert(re.error().empty()); |
331 |
|
string all("aa\r\naa\r\n"); |
332 |
|
CHECK(re.GlobalReplace("bb", &all) > 0); |
333 |
|
CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb")); |
334 |
|
} |
335 |
|
// TODO: test what happens when no PCRE_NEWLINE_* flag is set. |
336 |
|
// Alas, the answer depends on how pcre was compiled. |
337 |
} |
} |
338 |
|
|
339 |
static void TestExtract() { |
static void TestExtract() { |
412 |
CHECK_EQ(a, ""); |
CHECK_EQ(a, ""); |
413 |
} |
} |
414 |
|
|
415 |
static void TestRecursion(int size, const char *pattern, int match_limit) { |
static void TestRecursion() { |
416 |
printf("Testing recursion\n"); |
printf("Testing recursion\n"); |
417 |
|
|
418 |
// Fill up a string repeating the pattern given |
// Get one string that passes (sometimes), one that never does. |
419 |
string domain; |
string text_good("abcdefghijk"); |
420 |
domain.resize(size); |
string text_bad("acdefghijkl"); |
421 |
int patlen = strlen(pattern); |
|
422 |
for (int i = 0; i < size; ++i) { |
// According to pcretest, matching text_good against (\w+)*b |
423 |
domain[i] = pattern[i % patlen]; |
// requires match_limit of at least 8192, and match_recursion_limit |
424 |
|
// of at least 37. |
425 |
|
|
426 |
|
RE_Options options_ml; |
427 |
|
options_ml.set_match_limit(8192); |
428 |
|
RE re("(\\w+)*b", options_ml); |
429 |
|
CHECK(re.PartialMatch(text_good) == true); |
430 |
|
CHECK(re.PartialMatch(text_bad) == false); |
431 |
|
CHECK(re.FullMatch(text_good) == false); |
432 |
|
CHECK(re.FullMatch(text_bad) == false); |
433 |
|
|
434 |
|
options_ml.set_match_limit(1024); |
435 |
|
RE re2("(\\w+)*b", options_ml); |
436 |
|
CHECK(re2.PartialMatch(text_good) == false); // because of match_limit |
437 |
|
CHECK(re2.PartialMatch(text_bad) == false); |
438 |
|
CHECK(re2.FullMatch(text_good) == false); |
439 |
|
CHECK(re2.FullMatch(text_bad) == false); |
440 |
|
|
441 |
|
RE_Options options_mlr; |
442 |
|
options_mlr.set_match_limit_recursion(50); |
443 |
|
RE re3("(\\w+)*b", options_mlr); |
444 |
|
CHECK(re3.PartialMatch(text_good) == true); |
445 |
|
CHECK(re3.PartialMatch(text_bad) == false); |
446 |
|
CHECK(re3.FullMatch(text_good) == false); |
447 |
|
CHECK(re3.FullMatch(text_bad) == false); |
448 |
|
|
449 |
|
options_mlr.set_match_limit_recursion(10); |
450 |
|
RE re4("(\\w+)*b", options_mlr); |
451 |
|
CHECK(re4.PartialMatch(text_good) == false); |
452 |
|
CHECK(re4.PartialMatch(text_bad) == false); |
453 |
|
CHECK(re4.FullMatch(text_good) == false); |
454 |
|
CHECK(re4.FullMatch(text_bad) == false); |
455 |
|
} |
456 |
|
|
457 |
|
// A meta-quoted string, interpreted as a pattern, should always match |
458 |
|
// the original unquoted string. |
459 |
|
static void TestQuoteMeta(string unquoted, RE_Options options = RE_Options()) { |
460 |
|
string quoted = RE::QuoteMeta(unquoted); |
461 |
|
RE re(quoted, options); |
462 |
|
CHECK(re.FullMatch(unquoted)); |
463 |
|
} |
464 |
|
|
465 |
|
// A string containing meaningful regexp characters, which is then meta- |
466 |
|
// quoted, should not generally match a string the unquoted string does. |
467 |
|
static void NegativeTestQuoteMeta(string unquoted, string should_not_match, |
468 |
|
RE_Options options = RE_Options()) { |
469 |
|
string quoted = RE::QuoteMeta(unquoted); |
470 |
|
RE re(quoted, options); |
471 |
|
CHECK(!re.FullMatch(should_not_match)); |
472 |
|
} |
473 |
|
|
474 |
|
// Tests that quoted meta characters match their original strings, |
475 |
|
// and that a few things that shouldn't match indeed do not. |
476 |
|
static void TestQuotaMetaSimple() { |
477 |
|
TestQuoteMeta("foo"); |
478 |
|
TestQuoteMeta("foo.bar"); |
479 |
|
TestQuoteMeta("foo\\.bar"); |
480 |
|
TestQuoteMeta("[1-9]"); |
481 |
|
TestQuoteMeta("1.5-2.0?"); |
482 |
|
TestQuoteMeta("\\d"); |
483 |
|
TestQuoteMeta("Who doesn't like ice cream?"); |
484 |
|
TestQuoteMeta("((a|b)c?d*e+[f-h]i)"); |
485 |
|
TestQuoteMeta("((?!)xxx).*yyy"); |
486 |
|
TestQuoteMeta("(["); |
487 |
|
} |
488 |
|
|
489 |
|
static void TestQuoteMetaSimpleNegative() { |
490 |
|
NegativeTestQuoteMeta("foo", "bar"); |
491 |
|
NegativeTestQuoteMeta("...", "bar"); |
492 |
|
NegativeTestQuoteMeta("\\.", "."); |
493 |
|
NegativeTestQuoteMeta("\\.", ".."); |
494 |
|
NegativeTestQuoteMeta("(a)", "a"); |
495 |
|
NegativeTestQuoteMeta("(a|b)", "a"); |
496 |
|
NegativeTestQuoteMeta("(a|b)", "(a)"); |
497 |
|
NegativeTestQuoteMeta("(a|b)", "a|b"); |
498 |
|
NegativeTestQuoteMeta("[0-9]", "0"); |
499 |
|
NegativeTestQuoteMeta("[0-9]", "0-9"); |
500 |
|
NegativeTestQuoteMeta("[0-9]", "[9]"); |
501 |
|
NegativeTestQuoteMeta("((?!)xxx)", "xxx"); |
502 |
|
} |
503 |
|
|
504 |
|
static void TestQuoteMetaLatin1() { |
505 |
|
TestQuoteMeta("3\xb2 = 9"); |
506 |
|
} |
507 |
|
|
508 |
|
static void TestQuoteMetaUtf8() { |
509 |
|
#ifdef SUPPORT_UTF8 |
510 |
|
TestQuoteMeta("Pl\xc3\xa1\x63ido Domingo", pcrecpp::UTF8()); |
511 |
|
TestQuoteMeta("xyz", pcrecpp::UTF8()); // No fancy utf8 |
512 |
|
TestQuoteMeta("\xc2\xb0", pcrecpp::UTF8()); // 2-byte utf8 (degree symbol) |
513 |
|
TestQuoteMeta("27\xc2\xb0 degrees", pcrecpp::UTF8()); // As a middle character |
514 |
|
TestQuoteMeta("\xe2\x80\xb3", pcrecpp::UTF8()); // 3-byte utf8 (double prime) |
515 |
|
TestQuoteMeta("\xf0\x9d\x85\x9f", pcrecpp::UTF8()); // 4-byte utf8 (music note) |
516 |
|
TestQuoteMeta("27\xc2\xb0"); // Interpreted as Latin-1, but should still work |
517 |
|
NegativeTestQuoteMeta("27\xc2\xb0", // 2-byte utf (degree symbol) |
518 |
|
"27\\\xc2\\\xb0", |
519 |
|
pcrecpp::UTF8()); |
520 |
|
#endif |
521 |
|
} |
522 |
|
|
523 |
|
static void TestQuoteMetaAll() { |
524 |
|
printf("Testing QuoteMeta\n"); |
525 |
|
TestQuotaMetaSimple(); |
526 |
|
TestQuoteMetaSimpleNegative(); |
527 |
|
TestQuoteMetaLatin1(); |
528 |
|
TestQuoteMetaUtf8(); |
529 |
|
} |
530 |
|
|
531 |
|
// |
532 |
|
// Options tests contributed by |
533 |
|
// Giuseppe Maxia, CTO, Stardata s.r.l. |
534 |
|
// July 2005 |
535 |
|
// |
536 |
|
static void GetOneOptionResult( |
537 |
|
const char *option_name, |
538 |
|
const char *regex, |
539 |
|
const char *str, |
540 |
|
RE_Options options, |
541 |
|
bool full, |
542 |
|
string expected) { |
543 |
|
|
544 |
|
printf("Testing Option <%s>\n", option_name); |
545 |
|
if(VERBOSE_TEST) |
546 |
|
printf("/%s/ finds \"%s\" within \"%s\" \n", |
547 |
|
regex, |
548 |
|
expected.c_str(), |
549 |
|
str); |
550 |
|
string captured(""); |
551 |
|
if (full) |
552 |
|
RE(regex,options).FullMatch(str, &captured); |
553 |
|
else |
554 |
|
RE(regex,options).PartialMatch(str, &captured); |
555 |
|
CHECK_EQ(captured, expected); |
556 |
|
} |
557 |
|
|
558 |
|
static void TestOneOption( |
559 |
|
const char *option_name, |
560 |
|
const char *regex, |
561 |
|
const char *str, |
562 |
|
RE_Options options, |
563 |
|
bool full, |
564 |
|
bool assertive = true) { |
565 |
|
|
566 |
|
printf("Testing Option <%s>\n", option_name); |
567 |
|
if (VERBOSE_TEST) |
568 |
|
printf("'%s' %s /%s/ \n", |
569 |
|
str, |
570 |
|
(assertive? "matches" : "doesn't match"), |
571 |
|
regex); |
572 |
|
if (assertive) { |
573 |
|
if (full) |
574 |
|
CHECK(RE(regex,options).FullMatch(str)); |
575 |
|
else |
576 |
|
CHECK(RE(regex,options).PartialMatch(str)); |
577 |
|
} else { |
578 |
|
if (full) |
579 |
|
CHECK(!RE(regex,options).FullMatch(str)); |
580 |
|
else |
581 |
|
CHECK(!RE(regex,options).PartialMatch(str)); |
582 |
} |
} |
583 |
// Just make sure it doesn't crash due to too much recursion. |
} |
584 |
|
|
585 |
|
static void Test_CASELESS() { |
586 |
RE_Options options; |
RE_Options options; |
587 |
options.set_match_limit(match_limit); |
RE_Options options2; |
588 |
RE re("([a-zA-Z0-9]|-)+(\\.([a-zA-Z0-9]|-)+)*(\\.)?", options); |
|
589 |
re.FullMatch(domain); |
options.set_caseless(true); |
590 |
|
TestOneOption("CASELESS (class)", "HELLO", "hello", options, false); |
591 |
|
TestOneOption("CASELESS (class2)", "HELLO", "hello", options2.set_caseless(true), false); |
592 |
|
TestOneOption("CASELESS (class)", "^[A-Z]+$", "Hello", options, false); |
593 |
|
|
594 |
|
TestOneOption("CASELESS (function)", "HELLO", "hello", pcrecpp::CASELESS(), false); |
595 |
|
TestOneOption("CASELESS (function)", "^[A-Z]+$", "Hello", pcrecpp::CASELESS(), false); |
596 |
|
options.set_caseless(false); |
597 |
|
TestOneOption("no CASELESS", "HELLO", "hello", options, false, false); |
598 |
|
} |
599 |
|
|
600 |
|
static void Test_MULTILINE() { |
601 |
|
RE_Options options; |
602 |
|
RE_Options options2; |
603 |
|
const char *str = "HELLO\n" "cruel\n" "world\n"; |
604 |
|
|
605 |
|
options.set_multiline(true); |
606 |
|
TestOneOption("MULTILINE (class)", "^cruel$", str, options, false); |
607 |
|
TestOneOption("MULTILINE (class2)", "^cruel$", str, options2.set_multiline(true), false); |
608 |
|
TestOneOption("MULTILINE (function)", "^cruel$", str, pcrecpp::MULTILINE(), false); |
609 |
|
options.set_multiline(false); |
610 |
|
TestOneOption("no MULTILINE", "^cruel$", str, options, false, false); |
611 |
} |
} |
612 |
|
|
613 |
|
static void Test_DOTALL() { |
614 |
|
RE_Options options; |
615 |
|
RE_Options options2; |
616 |
|
const char *str = "HELLO\n" "cruel\n" "world"; |
617 |
|
|
618 |
|
options.set_dotall(true); |
619 |
|
TestOneOption("DOTALL (class)", "HELLO.*world", str, options, true); |
620 |
|
TestOneOption("DOTALL (class2)", "HELLO.*world", str, options2.set_dotall(true), true); |
621 |
|
TestOneOption("DOTALL (function)", "HELLO.*world", str, pcrecpp::DOTALL(), true); |
622 |
|
options.set_dotall(false); |
623 |
|
TestOneOption("no DOTALL", "HELLO.*world", str, options, true, false); |
624 |
|
} |
625 |
|
|
626 |
|
static void Test_DOLLAR_ENDONLY() { |
627 |
|
RE_Options options; |
628 |
|
RE_Options options2; |
629 |
|
const char *str = "HELLO world\n"; |
630 |
|
|
631 |
|
TestOneOption("no DOLLAR_ENDONLY", "world$", str, options, false); |
632 |
|
options.set_dollar_endonly(true); |
633 |
|
TestOneOption("DOLLAR_ENDONLY 1", "world$", str, options, false, false); |
634 |
|
TestOneOption("DOLLAR_ENDONLY 2", "world$", str, options2.set_dollar_endonly(true), false, false); |
635 |
|
} |
636 |
|
|
637 |
|
static void Test_EXTRA() { |
638 |
|
RE_Options options; |
639 |
|
const char *str = "HELLO"; |
640 |
|
|
641 |
|
options.set_extra(true); |
642 |
|
TestOneOption("EXTRA 1", "\\HELL\\O", str, options, true, false ); |
643 |
|
TestOneOption("EXTRA 2", "\\HELL\\O", str, RE_Options().set_extra(true), true, false ); |
644 |
|
options.set_extra(false); |
645 |
|
TestOneOption("no EXTRA", "\\HELL\\O", str, options, true ); |
646 |
|
} |
647 |
|
|
648 |
|
static void Test_EXTENDED() { |
649 |
|
RE_Options options; |
650 |
|
RE_Options options2; |
651 |
|
const char *str = "HELLO world"; |
652 |
|
|
653 |
|
options.set_extended(true); |
654 |
|
TestOneOption("EXTENDED (class)", "HELLO world", str, options, false, false); |
655 |
|
TestOneOption("EXTENDED (class2)", "HELLO world", str, options2.set_extended(true), false, false); |
656 |
|
TestOneOption("EXTENDED (class)", |
657 |
|
"^ HE L{2} O " |
658 |
|
"\\s+ " |
659 |
|
"\\w+ $ ", |
660 |
|
str, |
661 |
|
options, |
662 |
|
false); |
663 |
|
|
664 |
|
TestOneOption("EXTENDED (function)", "HELLO world", str, pcrecpp::EXTENDED(), false, false); |
665 |
|
TestOneOption("EXTENDED (function)", |
666 |
|
"^ HE L{2} O " |
667 |
|
"\\s+ " |
668 |
|
"\\w+ $ ", |
669 |
|
str, |
670 |
|
pcrecpp::EXTENDED(), |
671 |
|
false); |
672 |
|
|
673 |
|
options.set_extended(false); |
674 |
|
TestOneOption("no EXTENDED", "HELLO world", str, options, false); |
675 |
|
} |
676 |
|
|
677 |
|
static void Test_NO_AUTO_CAPTURE() { |
678 |
|
RE_Options options; |
679 |
|
const char *str = "HELLO world"; |
680 |
|
string captured; |
681 |
|
|
682 |
|
printf("Testing Option <no NO_AUTO_CAPTURE>\n"); |
683 |
|
if (VERBOSE_TEST) |
684 |
|
printf("parentheses capture text\n"); |
685 |
|
RE re("(world|universe)$", options); |
686 |
|
CHECK(re.Extract("\\1", str , &captured)); |
687 |
|
CHECK_EQ(captured, "world"); |
688 |
|
options.set_no_auto_capture(true); |
689 |
|
printf("testing Option <NO_AUTO_CAPTURE>\n"); |
690 |
|
if (VERBOSE_TEST) |
691 |
|
printf("parentheses do not capture text\n"); |
692 |
|
re.Extract("\\1",str, &captured ); |
693 |
|
CHECK_EQ(captured, "world"); |
694 |
|
} |
695 |
|
|
696 |
|
static void Test_UNGREEDY() { |
697 |
|
RE_Options options; |
698 |
|
const char *str = "HELLO, 'this' is the 'world'"; |
699 |
|
|
700 |
|
options.set_ungreedy(true); |
701 |
|
GetOneOptionResult("UNGREEDY 1", "('.*')", str, options, false, "'this'" ); |
702 |
|
GetOneOptionResult("UNGREEDY 2", "('.*')", str, RE_Options().set_ungreedy(true), false, "'this'" ); |
703 |
|
GetOneOptionResult("UNGREEDY", "('.*?')", str, options, false, "'this' is the 'world'" ); |
704 |
|
|
705 |
|
options.set_ungreedy(false); |
706 |
|
GetOneOptionResult("no UNGREEDY", "('.*')", str, options, false, "'this' is the 'world'" ); |
707 |
|
GetOneOptionResult("no UNGREEDY", "('.*?')", str, options, false, "'this'" ); |
708 |
|
} |
709 |
|
|
710 |
|
static void Test_all_options() { |
711 |
|
const char *str = "HELLO\n" "cruel\n" "world"; |
712 |
|
RE_Options options; |
713 |
|
options.set_all_options(PCRE_CASELESS | PCRE_DOTALL); |
714 |
|
|
715 |
|
TestOneOption("all_options (CASELESS|DOTALL)", "^hello.*WORLD", str , options, false); |
716 |
|
options.set_all_options(0); |
717 |
|
TestOneOption("all_options (0)", "^hello.*WORLD", str , options, false, false); |
718 |
|
options.set_all_options(PCRE_MULTILINE | PCRE_EXTENDED); |
719 |
|
|
720 |
|
TestOneOption("all_options (MULTILINE|EXTENDED)", " ^ c r u e l $ ", str, options, false); |
721 |
|
TestOneOption("all_options (MULTILINE|EXTENDED) with constructor", |
722 |
|
" ^ c r u e l $ ", |
723 |
|
str, |
724 |
|
RE_Options(PCRE_MULTILINE | PCRE_EXTENDED), |
725 |
|
false); |
726 |
|
|
727 |
|
TestOneOption("all_options (MULTILINE|EXTENDED) with concatenation", |
728 |
|
" ^ c r u e l $ ", |
729 |
|
str, |
730 |
|
RE_Options() |
731 |
|
.set_multiline(true) |
732 |
|
.set_extended(true), |
733 |
|
false); |
734 |
|
|
735 |
|
options.set_all_options(0); |
736 |
|
TestOneOption("all_options (0)", "^ c r u e l $", str, options, false, false); |
737 |
|
|
738 |
|
} |
739 |
|
|
740 |
|
static void TestOptions() { |
741 |
|
printf("Testing Options\n"); |
742 |
|
Test_CASELESS(); |
743 |
|
Test_MULTILINE(); |
744 |
|
Test_DOTALL(); |
745 |
|
Test_DOLLAR_ENDONLY(); |
746 |
|
Test_EXTENDED(); |
747 |
|
Test_NO_AUTO_CAPTURE(); |
748 |
|
Test_UNGREEDY(); |
749 |
|
Test_EXTRA(); |
750 |
|
Test_all_options(); |
751 |
|
} |
752 |
|
|
753 |
|
static void TestConstructors() { |
754 |
|
printf("Testing constructors\n"); |
755 |
|
|
756 |
|
RE_Options options; |
757 |
|
options.set_dotall(true); |
758 |
|
const char *str = "HELLO\n" "cruel\n" "world"; |
759 |
|
|
760 |
|
RE orig("HELLO.*world", options); |
761 |
|
CHECK(orig.FullMatch(str)); |
762 |
|
|
763 |
|
RE copy1(orig); |
764 |
|
CHECK(copy1.FullMatch(str)); |
765 |
|
|
766 |
|
RE copy2("not a match"); |
767 |
|
CHECK(!copy2.FullMatch(str)); |
768 |
|
copy2 = copy1; |
769 |
|
CHECK(copy2.FullMatch(str)); |
770 |
|
copy2 = orig; |
771 |
|
CHECK(copy2.FullMatch(str)); |
772 |
|
|
773 |
|
// Make sure when we assign to ourselves, nothing bad happens |
774 |
|
orig = orig; |
775 |
|
copy1 = copy1; |
776 |
|
copy2 = copy2; |
777 |
|
CHECK(orig.FullMatch(str)); |
778 |
|
CHECK(copy1.FullMatch(str)); |
779 |
|
CHECK(copy2.FullMatch(str)); |
780 |
|
} |
781 |
|
|
782 |
int main(int argc, char** argv) { |
int main(int argc, char** argv) { |
783 |
// Treat any flag as --help |
// Treat any flag as --help |
813 |
/***** FullMatch with no args *****/ |
/***** FullMatch with no args *****/ |
814 |
|
|
815 |
CHECK(RE("h.*o").FullMatch("hello")); |
CHECK(RE("h.*o").FullMatch("hello")); |
816 |
CHECK(!RE("h.*o").FullMatch("othello")); |
CHECK(!RE("h.*o").FullMatch("othello")); // Must be anchored at front |
817 |
CHECK(!RE("h.*o").FullMatch("hello!")); |
CHECK(!RE("h.*o").FullMatch("hello!")); // Must be anchored at end |
818 |
|
CHECK(RE("a*").FullMatch("aaaa")); // Fullmatch with normal op |
819 |
|
CHECK(RE("a*?").FullMatch("aaaa")); // Fullmatch with nongreedy op |
820 |
|
CHECK(RE("a*?\\z").FullMatch("aaaa")); // Two unusual ops |
821 |
|
|
822 |
/***** FullMatch with args *****/ |
/***** FullMatch with args *****/ |
823 |
|
|
912 |
CHECK(!RE("(\\d+)").FullMatch("4294967296", &v)); |
CHECK(!RE("(\\d+)").FullMatch("4294967296", &v)); |
913 |
} |
} |
914 |
#ifdef HAVE_LONG_LONG |
#ifdef HAVE_LONG_LONG |
915 |
|
# if defined(__MINGW__) || defined(__MINGW32__) |
916 |
|
# define LLD "%I64d" |
917 |
|
# define LLU "%I64u" |
918 |
|
# else |
919 |
|
# define LLD "%lld" |
920 |
|
# define LLU "%llu" |
921 |
|
# endif |
922 |
{ |
{ |
923 |
long long v; |
long long v; |
924 |
static const long long max_value = 0x7fffffffffffffffLL; |
static const long long max_value = 0x7fffffffffffffffLL; |
928 |
CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100); |
CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100); |
929 |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v)); CHECK_EQ(v, -100); |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v)); CHECK_EQ(v, -100); |
930 |
|
|
931 |
snprintf(buf, sizeof(buf), "%lld", max_value); |
snprintf(buf, sizeof(buf), LLD, max_value); |
932 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value); |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value); |
933 |
|
|
934 |
snprintf(buf, sizeof(buf), "%lld", min_value); |
snprintf(buf, sizeof(buf), LLD, min_value); |
935 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, min_value); |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, min_value); |
936 |
|
|
937 |
snprintf(buf, sizeof(buf), "%lld", max_value); |
snprintf(buf, sizeof(buf), LLD, max_value); |
938 |
assert(buf[strlen(buf)-1] != '9'); |
assert(buf[strlen(buf)-1] != '9'); |
939 |
buf[strlen(buf)-1]++; |
buf[strlen(buf)-1]++; |
940 |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v)); |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v)); |
941 |
|
|
942 |
snprintf(buf, sizeof(buf), "%lld", min_value); |
snprintf(buf, sizeof(buf), LLD, min_value); |
943 |
assert(buf[strlen(buf)-1] != '9'); |
assert(buf[strlen(buf)-1] != '9'); |
944 |
buf[strlen(buf)-1]++; |
buf[strlen(buf)-1]++; |
945 |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v)); |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v)); |
955 |
CHECK(RE("(-?\\d+)").FullMatch("100",&v)); CHECK_EQ(v, 100); |
CHECK(RE("(-?\\d+)").FullMatch("100",&v)); CHECK_EQ(v, 100); |
956 |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v2)); CHECK_EQ(v2, -100); |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v2)); CHECK_EQ(v2, -100); |
957 |
|
|
958 |
snprintf(buf, sizeof(buf), "%llu", max_value); |
snprintf(buf, sizeof(buf), LLU, max_value); |
959 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value); |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value); |
960 |
|
|
961 |
assert(buf[strlen(buf)-1] != '9'); |
assert(buf[strlen(buf)-1] != '9'); |
1107 |
CHECK(RE("h.*o").PartialMatch("hello!")); |
CHECK(RE("h.*o").PartialMatch("hello!")); |
1108 |
CHECK(RE("((((((((((((((((((((x))))))))))))))))))))").PartialMatch("x")); |
CHECK(RE("((((((((((((((((((((x))))))))))))))))))))").PartialMatch("x")); |
1109 |
|
|
1110 |
|
/***** other tests *****/ |
1111 |
|
|
1112 |
RadixTests(); |
RadixTests(); |
1113 |
TestReplace(); |
TestReplace(); |
1114 |
TestExtract(); |
TestExtract(); |
1115 |
TestConsume(); |
TestConsume(); |
1116 |
TestFindAndConsume(); |
TestFindAndConsume(); |
1117 |
|
TestQuoteMetaAll(); |
1118 |
TestMatchNumberPeculiarity(); |
TestMatchNumberPeculiarity(); |
1119 |
|
|
1120 |
// Check the pattern() accessor |
// Check the pattern() accessor |
1226 |
CHECK(!re.error().empty()); |
CHECK(!re.error().empty()); |
1227 |
} |
} |
1228 |
|
|
1229 |
// Test that recursion is stopped: there will be some errors reported |
// Test that recursion is stopped |
1230 |
int matchlimit = 5000; |
TestRecursion(); |
1231 |
int bytes = 15 * 1024; // enough to crash if there was no match limit |
|
1232 |
TestRecursion(bytes, ".", matchlimit); |
// Test Options |
1233 |
TestRecursion(bytes, "a", matchlimit); |
if (getenv("VERBOSE_TEST") != NULL) |
1234 |
TestRecursion(bytes, "a.", matchlimit); |
VERBOSE_TEST = true; |
1235 |
TestRecursion(bytes, "ab.", matchlimit); |
TestOptions(); |
1236 |
TestRecursion(bytes, "abc.", matchlimit); |
|
1237 |
|
// Test the constructors |
1238 |
|
TestConstructors(); |
1239 |
|
|
1240 |
// Done |
// Done |
1241 |
printf("OK\n"); |
printf("OK\n"); |