/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 170 - (show annotations) (download)
Mon Jun 4 11:21:13 2007 UTC (6 years, 10 months ago) by ph10
File size: 128668 byte(s)
Fix bug in detecting potentially empty groups.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.2 01-May-07
5 ---------------------
6
7 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8 which is apparently normally available under Windows.
9
10 2. Re-jig the pcregrep tests with different newline settings in an attempt
11 to make them independent of the local environment's newline setting.
12
13 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14
15 4. Some of the "internals" tests were previously cut out when the link size
16 was not 2, because the output contained actual offsets. The recent new
17 "Z" feature of pcretest means that these can be cut out, making the tests
18 usable with all link sizes.
19
20 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21 stack recursion. This gives a massive performance boost under BSD, but just
22 a small improvement under Linux. However, it saves one field in the frame
23 in all cases.
24
25 6. Added more features from the forthcoming Perl 5.10:
26
27 (a) (?-n) (where n is a string of digits) is a relative subroutine or
28 recursion call. It refers to the nth most recently opened parentheses.
29
30 (b) (?+n) is also a relative subroutine call; it refers to the nth next
31 to be opened parentheses.
32
33 (c) Conditions that refer to capturing parentheses can be specified
34 relatively, for example, (?(-2)... or (?(+3)...
35
36 (d) \K resets the start of the current match so that everything before
37 is not part of it.
38
39 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
40 PCRE_INFO_JCHANGED.
41
42 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
43 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
44 for detecting groups that can match an empty string.
45
46
47 Version 7.1 24-Apr-07
48 ---------------------
49
50 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
51 that is more "standard", making use of automake and other Autotools. There
52 is some re-arrangement of the files and adjustment of comments consequent
53 on this.
54
55 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
56 for recursive directory scanning broke on some systems because the files
57 are not scanned in any specific order and on different systems the order
58 was different. A call to "sort" has been inserted into RunGrepTest for the
59 approprate test as a short-term fix. In the longer term there may be an
60 alternative.
61
62 3. I had an email from Eric Raymond about problems translating some of PCRE's
63 man pages to HTML (despite the fact that I distribute HTML pages, some
64 people do their own conversions for various reasons). The problems
65 concerned the use of low-level troff macros .br and .in. I have therefore
66 removed all such uses from the man pages (some were redundant, some could
67 be replaced by .nf/.fi pairs). The 132html script that I use to generate
68 HTML has been updated to handle .nf/.fi and to complain if it encounters
69 .br or .in.
70
71 4. Updated comments in configure.ac that get placed in config.h.in and also
72 arranged for config.h to be included in the distribution, with the name
73 config.h.generic, for the benefit of those who have to compile without
74 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
75
76 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
77 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
78 makevp.bat for the current PCRE, using the additional files
79 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
80
81 6. A Windows user reported a minor discrepancy with test 2, which turned out
82 to be caused by a trailing space on an input line that had got lost in his
83 copy. The trailing space was an accident, so I've just removed it.
84
85 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
86 that is needed.
87
88 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
89 as "const" (a) because they are and (b) because it helps the PHP
90 maintainers who have recently made a script to detect big data structures
91 in the php code that should be moved to the .rodata section. I remembered
92 to update Builducptable as well, so it won't revert if ucptable.h is ever
93 re-created.
94
95 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
96 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
97 order to be able to cut out the UTF-8 tables in the latter when UTF-8
98 support is not required. This saves 1.5-2K of code, which is important in
99 some applications.
100
101 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
102 so as not to refer to the tables, even though these functions will never be
103 called when UTF-8 support is disabled. Otherwise there are problems with a
104 shared library.
105
106 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
107
108 (a) It was defining its arguments as char * instead of void *.
109
110 (b) It was assuming that all moves were upwards in memory; this was true
111 a long time ago when I wrote it, but is no longer the case.
112
113 The emulated memove() is provided for those environments that have neither
114 memmove() nor bcopy(). I didn't think anyone used it these days, but that
115 is clearly not the case, as these two bugs were recently reported.
116
117 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
118 and Detrail to create the HTML documentation, the .txt form of the man
119 pages, and it removes trailing spaces from listed files. It also creates
120 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
121 case, it wraps all the #defines with #ifndefs. This script should be run
122 before "make dist".
123
124 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
125 with Unicode property support.
126
127 (a) For a maximizing quantifier, if the two different cases of the
128 character were of different lengths in their UTF-8 codings (there are
129 some cases like this - I found 11), and the matching function had to
130 back up over a mixture of the two cases, it incorrectly assumed they
131 were both the same length.
132
133 (b) When PCRE was configured to use the heap rather than the stack for
134 recursion during matching, it was not correctly preserving the data for
135 the other case of a UTF-8 character when checking ahead for a match
136 while processing a minimizing repeat. If the check also involved
137 matching a wide character, but failed, corruption could cause an
138 erroneous result when trying to check for a repeat of the original
139 character.
140
141 13. Some tidying changes to the testing mechanism:
142
143 (a) The RunTest script now detects the internal link size and whether there
144 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
145 values substituted by "configure". (The RunGrepTest script already did
146 this for UTF-8.) The configure.ac script no longer substitutes the
147 relevant variables.
148
149 (b) The debugging options /B and /D in pcretest show the compiled bytecode
150 with length and offset values. This means that the output is different
151 for different internal link sizes. Test 2 is skipped for link sizes
152 other than 2 because of this, bypassing the problem. Unfortunately,
153 there was also a test in test 3 (the locale tests) that used /B and
154 failed for link sizes other than 2. Rather than cut the whole test out,
155 I have added a new /Z option to pcretest that replaces the length and
156 offset values with spaces. This is now used to make test 3 independent
157 of link size. (Test 2 will be tidied up later.)
158
159 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
160 segmentation fault instead of returning the appropriate error message.
161
162 15. In multiline mode when the newline sequence was set to "any", the pattern
163 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
164 This doesn't seem right; it now treats the CRLF combination as the line
165 ending, and so does not match in that case. It's only a pattern such as ^$
166 that would hit this one: something like ^ABC$ would have failed after \r
167 and then tried again after \r\n.
168
169 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
170 in an attempt to make files that differ only in their line terminators
171 compare equal. This works on Linux.
172
173 17. Under certain error circumstances pcregrep might try to free random memory
174 as it exited. This is now fixed, thanks to valgrind.
175
176 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
177 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
178 was because its rules for how to advance for /g after matching an empty
179 string at the end of a line did not allow for this case. They now check for
180 it specially.
181
182 20. pcretest is supposed to handle patterns and data of any length, by
183 extending its buffers when necessary. It was getting this wrong when the
184 buffer for a data line had to be extended.
185
186 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
187 CRLF as a newline sequence.
188
189 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
190 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
191 I have nevertheless tidied it up.
192
193 23. Added some casts to kill warnings from HP-UX ia64 compiler.
194
195 24. Added a man page for pcre-config.
196
197
198 Version 7.0 19-Dec-06
199 ---------------------
200
201 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
202 moving to gcc 4.1.1.
203
204 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
205 sys/time.h, which is documented as needed for this function. It doesn't
206 seem to matter on Linux, but it showed up on some releases of OS X.
207
208 3. It seems that there are systems where bytes whose values are greater than
209 127 match isprint() in the "C" locale. The "C" locale should be the
210 default when a C program starts up. In most systems, only ASCII printing
211 characters match isprint(). This difference caused the output from pcretest
212 to vary, making some of the tests fail. I have changed pcretest so that:
213
214 (a) When it is outputting text in the compiled version of a pattern, bytes
215 other than 32-126 are always shown as hex escapes.
216
217 (b) When it is outputting text that is a matched part of a subject string,
218 it does the same, unless a different locale has been set for the match
219 (using the /L modifier). In this case, it uses isprint() to decide.
220
221 4. Fixed a major bug that caused incorrect computation of the amount of memory
222 required for a compiled pattern when options that changed within the
223 pattern affected the logic of the preliminary scan that determines the
224 length. The relevant options are -x, and -i in UTF-8 mode. The result was
225 that the computed length was too small. The symptoms of this bug were
226 either the PCRE error "internal error: code overflow" from pcre_compile(),
227 or a glibc crash with a message such as "pcretest: free(): invalid next
228 size (fast)". Examples of patterns that provoked this bug (shown in
229 pcretest format) are:
230
231 /(?-x: )/x
232 /(?x)(?-x: \s*#\s*)/
233 /((?i)[\x{c0}])/8
234 /(?i:[\x{c0}])/8
235
236 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
237 is now done differently.
238
239 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
240 wrapper classes; (b) implement a new function in the C++ scanner that is
241 more efficient than the old way of doing things because it avoids levels of
242 recursion in the regex matching; (c) add a paragraph to the documentation
243 for the FullMatch() function.
244
245 6. The escape sequence \n was being treated as whatever was defined as
246 "newline". Not only was this contrary to the documentation, which states
247 that \n is character 10 (hex 0A), but it also went horribly wrong when
248 "newline" was defined as CRLF. This has been fixed.
249
250 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
251 was being set to -1 for the "end of line" case (supposedly a value that no
252 character can have). Though this value is never used (the check for end of
253 line is "zero bytes in current character"), it caused compiler complaints.
254 I've changed it to 0xffffffff.
255
256 8. In pcre_version.c, the version string was being built by a sequence of
257 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
258 string (as it is for production releases) called a macro with an empty
259 argument. The C standard says the result of this is undefined. The gcc
260 compiler treats it as an empty string (which was what was wanted) but it is
261 reported that Visual C gives an error. The source has been hacked around to
262 avoid this problem.
263
264 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
265 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
266 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
267 of them did).
268
269 10. Originally, pcretest opened its input and output without "b"; then I was
270 told that "b" was needed in some environments, so it was added for release
271 5.0 to both the input and output. (It makes no difference on Unix-like
272 systems.) Later I was told that it is wrong for the input on Windows. I've
273 now abstracted the modes into two macros, to make it easier to fiddle with
274 them, and removed "b" from the input mode under Windows.
275
276 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
277
278 12. Added -help and --help to pcretest as an official way of being reminded
279 of the options.
280
281 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
282 and pcrecpp.cc because they annoy compilers at high warning levels.
283
284 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
285
286 15. Fixed an occurrence of == in configure.ac that should have been = (shell
287 scripts are not C programs :-) and which was not noticed because it works
288 on Linux.
289
290 16. pcretest is supposed to handle any length of pattern and data line (as one
291 line or as a continued sequence of lines) by extending its input buffer if
292 necessary. This feature was broken for very long pattern lines, leading to
293 a string of junk being passed to pcre_compile() if the pattern was longer
294 than about 50K.
295
296 17. I have done a major re-factoring of the way pcre_compile() computes the
297 amount of memory needed for a compiled pattern. Previously, there was code
298 that made a preliminary scan of the pattern in order to do this. That was
299 OK when PCRE was new, but as the facilities have expanded, it has become
300 harder and harder to keep it in step with the real compile phase, and there
301 have been a number of bugs (see for example, 4 above). I have now found a
302 cunning way of running the real compile function in a "fake" mode that
303 enables it to compute how much memory it would need, while actually only
304 ever using a few hundred bytes of working memory and without too many
305 tests of the mode. This should make future maintenance and development
306 easier. A side effect of this work is that the limit of 200 on the nesting
307 depth of parentheses has been removed (though this was never a serious
308 limitation, I suspect). However, there is a downside: pcre_compile() now
309 runs more slowly than before (30% or more, depending on the pattern). I
310 hope this isn't a big issue. There is no effect on runtime performance.
311
312 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
313 newline (only possible for the last line of a file) and it was a
314 pattern that set a locale (followed by /Lsomething), pcretest crashed.
315
316 19. Added additional timing features to pcretest. (1) The -tm option now times
317 matching only, not compiling. (2) Both -t and -tm can be followed, as a
318 separate command line item, by a number that specifies the number of
319 repeats to use when timing. The default is 50000; this gives better
320 precision, but takes uncomfortably long for very large patterns.
321
322 20. Extended pcre_study() to be more clever in cases where a branch of a
323 subpattern has no definite first character. For example, (a*|b*)[cd] would
324 previously give no result from pcre_study(). Now it recognizes that the
325 first character must be a, b, c, or d.
326
327 21. There was an incorrect error "recursive call could loop indefinitely" if
328 a subpattern (or the entire pattern) that was being tested for matching an
329 empty string contained only one non-empty item after a nested subpattern.
330 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
331 incorrectly, because the \d was being skipped in the check.
332
333 22. The pcretest program now has a new pattern option /B and a command line
334 option -b, which is equivalent to adding /B to every pattern. This causes
335 it to show the compiled bytecode, without the additional information that
336 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
337 is the same as /B/I).
338
339 23. A new optimization is now able automatically to treat some sequences such
340 as a*b as a*+b. More specifically, if something simple (such as a character
341 or a simple class like \d) has an unlimited quantifier, and is followed by
342 something that cannot possibly match the quantified thing, the quantifier
343 is automatically "possessified".
344
345 24. A recursive reference to a subpattern whose number was greater than 39
346 went wrong under certain circumstances in UTF-8 mode. This bug could also
347 have affected the operation of pcre_study().
348
349 25. Realized that a little bit of performance could be had by replacing
350 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
351
352 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
353
354 27. Possessive quantifiers such as a++ were previously implemented by turning
355 them into atomic groups such as ($>a+). Now they have their own opcodes,
356 which improves performance. This includes the automatically created ones
357 from 23 above.
358
359 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
360 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
361 the first matched character to be a colon. This applied both to named and
362 numbered groups.
363
364 29. The ucpinternal.h header file was missing its idempotency #ifdef.
365
366 30. I was sent a "project" file called libpcre.a.dev which I understand makes
367 building PCRE on Windows easier, so I have included it in the distribution.
368
369 31. There is now a check in pcretest against a ridiculously large number being
370 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
371 loop, the loop is abandoned.
372
373 32. Forward references to subpatterns in conditions such as (?(2)...) where
374 subpattern 2 is defined later cause pcre_compile() to search forwards in
375 the pattern for the relevant set of parentheses. This search went wrong
376 when there were unescaped parentheses in a character class, parentheses
377 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
378
379 33. "Subroutine" calls and backreferences were previously restricted to
380 referencing subpatterns earlier in the regex. This restriction has now
381 been removed.
382
383 34. Added a number of extra features that are going to be in Perl 5.10. On the
384 whole, these are just syntactic alternatives for features that PCRE had
385 previously implemented using the Python syntax or my own invention. The
386 other formats are all retained for compatibility.
387
388 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
389 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
390 also .NET compatible.
391
392 (b) A recursion or subroutine call to a named group can now be defined as
393 (?&name) as well as (?P>name).
394
395 (c) A backreference to a named group can now be defined as \k<name> or
396 \k'name' as well as (?P=name). The new forms, as well as being in Perl
397 5.10, are also .NET compatible.
398
399 (d) A conditional reference to a named group can now use the syntax
400 (?(<name>) or (?('name') as well as (?(name).
401
402 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
403 groups (named and numbered) that are never evaluated inline, but can be
404 called as "subroutines" from elsewhere. In effect, the DEFINE condition
405 is always false. There may be only one alternative in such a group.
406
407 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
408 as the simple (?(R). The condition is true only if the most recent
409 recursion is that of the given number or name. It does not search out
410 through the entire recursion stack.
411
412 (g) The escape \gN or \g{N} has been added, where N is a positive or
413 negative number, specifying an absolute or relative reference.
414
415 35. Tidied to get rid of some further signed/unsigned compiler warnings and
416 some "unreachable code" warnings.
417
418 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
419 things, this adds five new scripts.
420
421 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
422 There were also incompatibilities regarding the handling of \Q..\E inside
423 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
424 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
425
426 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
427 matches an empty string, and forcibly breaks the loop. There were bugs in
428 this code in non-simple cases. For a pattern such as ^(a()*)* matched
429 against aaaa the result was just "a" rather than "aaaa", for example. Two
430 separate and independent bugs (that affected different cases) have been
431 fixed.
432
433 39. Refactored the code to abolish the use of different opcodes for small
434 capturing bracket numbers. This is a tidy that I avoided doing when I
435 removed the limit on the number of capturing brackets for 3.5 back in 2001.
436 The new approach is not only tidier, it makes it possible to reduce the
437 memory needed to fix the previous bug (38).
438
439 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
440 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
441 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
442 mode.
443
444 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
445 report.
446
447 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
448 copy construction and assignment in the C++ wrapper.
449
450 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
451 couple of bugs that could have given wrong results in the "--newline=crlf"
452 case.
453
454 44. Added a number of casts and did some reorganization of signed/unsigned int
455 variables following suggestions from Dair Grant. Also renamed the variable
456 "this" as "item" because it is a C++ keyword.
457
458 45. Arranged for dftables to add
459
460 #include "pcre_internal.h"
461
462 to pcre_chartables.c because without it, gcc 4.x may remove the array
463 definition from the final binary if PCRE is built into a static library and
464 dead code stripping is activated.
465
466 46. For an unanchored pattern, if a match attempt fails at the start of a
467 newline sequence, and the newline setting is CRLF or ANY, and the next two
468 characters are CRLF, advance by two characters instead of one.
469
470
471 Version 6.7 04-Jul-06
472 ---------------------
473
474 1. In order to handle tests when input lines are enormously long, pcretest has
475 been re-factored so that it automatically extends its buffers when
476 necessary. The code is crude, but this _is_ just a test program. The
477 default size has been increased from 32K to 50K.
478
479 2. The code in pcre_study() was using the value of the re argument before
480 testing it for NULL. (Of course, in any sensible call of the function, it
481 won't be NULL.)
482
483 3. The memmove() emulation function in pcre_internal.h, which is used on
484 systems that lack both memmove() and bcopy() - that is, hardly ever -
485 was missing a "static" storage class specifier.
486
487 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
488 containing an extended class (one that cannot be represented by a bitmap
489 because it contains high-valued characters or Unicode property items, e.g.
490 [\pZ]). Almost always one would set UTF-8 mode when processing such a
491 pattern, but PCRE should not loop if you do not (it no longer does).
492 [Detail: two cases were found: (a) a repeated subpattern containing an
493 extended class; (b) a recursive reference to a subpattern that followed a
494 previous extended class. It wasn't skipping over the extended class
495 correctly when UTF-8 mode was not set.]
496
497 5. A negated single-character class was not being recognized as fixed-length
498 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
499 compile error "lookbehind assertion is not fixed length".
500
501 6. The RunPerlTest auxiliary script was showing an unexpected difference
502 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
503 write a Perl script that can interpret lines of an input file either as
504 byte characters or as UTF-8, which is what "perltest" was being required to
505 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
506 can't do is switch easily at run time between having the "use utf8;" pragma
507 or not. In the end, I fudged it by using the RunPerlTest script to insert
508 "use utf8;" explicitly for the UTF-8 tests.
509
510 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
511 the end of the subject string, contrary to the documentation and to what
512 Perl does. This was true of both matching functions. Now it matches only at
513 the start of the subject and immediately after *internal* newlines.
514
515 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
516 a pointer to an int instead of a pointer to an unsigned long int. This
517 caused problems on 64-bit systems.
518
519 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
520 instance of the 'standard' template library not being so standard".
521
522 10. There was no check on the number of named subpatterns nor the maximum
523 length of a subpattern name. The product of these values is used to compute
524 the size of the memory block for a compiled pattern. By supplying a very
525 long subpattern name and a large number of named subpatterns, the size
526 computation could be caused to overflow. This is now prevented by limiting
527 the length of names to 32 characters, and the number of named subpatterns
528 to 10,000.
529
530 11. Subpatterns that are repeated with specific counts have to be replicated in
531 the compiled pattern. The size of memory for this was computed from the
532 length of the subpattern and the repeat count. The latter is limited to
533 65535, but there was no limit on the former, meaning that integer overflow
534 could in principle occur. The compiled length of a repeated subpattern is
535 now limited to 30,000 bytes in order to prevent this.
536
537 12. Added the optional facility to have named substrings with the same name.
538
539 13. Added the ability to use a named substring as a condition, using the
540 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
541 are numbers (not recommended). Forward references are permitted.
542
543 14. Added forward references in named backreferences (if you see what I mean).
544
545 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
546 pattern could run off the end of the subject. For example, the pattern
547 "(?s)(.{1,5})"8 did this with the subject "ab".
548
549 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
550 PCRE_CASELESS was set when matching characters that were quantified with ?
551 or *.
552
553 17. A character class other than a single negated character that had a minimum
554 but no maximum quantifier - for example [ab]{6,} - was not handled
555 correctly by pce_dfa_exec(). It would match only one character.
556
557 18. A valid (though odd) pattern that looked like a POSIX character
558 class but used an invalid character after [ (for example [[,abc,]]) caused
559 pcre_compile() to give the error "Failed: internal error: code overflow" or
560 in some cases to crash with a glibc free() error. This could even happen if
561 the pattern terminated after [[ but there just happened to be a sequence of
562 letters, a binary zero, and a closing ] in the memory that followed.
563
564 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
565 over the years. Originally (before any Unicode support), just the bottom 8
566 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
567 output from "man perlunicode" includes this:
568
569 The regular expression compiler produces polymorphic opcodes. That
570 is, the pattern adapts to the data and automatically switches to
571 the Unicode character scheme when presented with Unicode data--or
572 instead uses a traditional byte scheme when presented with byte
573 data.
574
575 Sadly, a wide octal escape does not cause a switch, and in a string with
576 no other multibyte characters, these octal escapes are treated as before.
577 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
578 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
579 Unicode string.
580
581 I have not perpetrated such confusion in PCRE. Up till now, it took just
582 the bottom 8 bits, as in old Perl. I have now made octal escapes with
583 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
584 translate to the appropriate multibyte character.
585
586 29. Applied some refactoring to reduce the number of warnings from Microsoft
587 and Borland compilers. This has included removing the fudge introduced
588 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
589 a warning about an unused variable.
590
591 21. PCRE has not included VT (character 0x0b) in the set of whitespace
592 characters since release 4.0, because Perl (from release 5.004) does not.
593 [Or at least, is documented not to: some releases seem to be in conflict
594 with the documentation.] However, when a pattern was studied with
595 pcre_study() and all its branches started with \s, PCRE still included VT
596 as a possible starting character. Of course, this did no harm; it just
597 caused an unnecessary match attempt.
598
599 22. Removed a now-redundant internal flag bit that recorded the fact that case
600 dependency changed within the pattern. This was once needed for "required
601 byte" processing, but is no longer used. This recovers a now-scarce options
602 bit. Also moved the least significant internal flag bit to the most-
603 significant bit of the word, which was not previously used (hangover from
604 the days when it was an int rather than a uint) to free up another bit for
605 the future.
606
607 23. Added support for CRLF line endings as well as CR and LF. As well as the
608 default being selectable at build time, it can now be changed at runtime
609 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
610 specify that it is scanning data with non-default line endings.
611
612 24. Changed the definition of CXXLINK to make it agree with the definition of
613 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
614
615 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
616 recursions. This makes a big different to stack usage for some patterns.
617
618 26. If a subpattern containing a named recursion or subroutine reference such
619 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
620 the space required for the compiled pattern went wrong and gave too small a
621 value. Depending on the environment, this could lead to "Failed: internal
622 error: code overflow at offset 49" or "glibc detected double free or
623 corruption" errors.
624
625 27. Applied patches from Google (a) to support the new newline modes and (b) to
626 advance over multibyte UTF-8 characters in GlobalReplace.
627
628 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
629 difference for some implementation of PCRE in some Windows version.
630
631 29. Added some extra testing facilities to pcretest:
632
633 \q<number> in a data line sets the "match limit" value
634 \Q<number> in a data line sets the "match recursion limt" value
635 -S <number> sets the stack size, where <number> is in megabytes
636
637 The -S option isn't available for Windows.
638
639
640 Version 6.6 06-Feb-06
641 ---------------------
642
643 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
644 in pcreposix.h. I have copied the definition from pcre.h.
645
646 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
647 because pcre.h is no longer a built file.
648
649 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
650 not normally included in the compiled code.
651
652
653 Version 6.5 01-Feb-06
654 ---------------------
655
656 1. When using the partial match feature with pcre_dfa_exec(), it was not
657 anchoring the second and subsequent partial matches at the new starting
658 point. This could lead to incorrect results. For example, with the pattern
659 /1234/, partially matching against "123" and then "a4" gave a match.
660
661 2. Changes to pcregrep:
662
663 (a) All non-match returns from pcre_exec() were being treated as failures
664 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
665 error message is output. Some extra information is given for the
666 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
667 probably the only errors that are likely to be caused by users (by
668 specifying a regex that has nested indefinite repeats, for instance).
669 If there are more than 20 of these errors, pcregrep is abandoned.
670
671 (b) A binary zero was treated as data while matching, but terminated the
672 output line if it was written out. This has been fixed: binary zeroes
673 are now no different to any other data bytes.
674
675 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
676 used to set a locale for matching. The --locale=xxxx long option has
677 been added (no short equivalent) to specify a locale explicitly on the
678 pcregrep command, overriding the environment variables.
679
680 (d) When -B was used with -n, some line numbers in the output were one less
681 than they should have been.
682
683 (e) Added the -o (--only-matching) option.
684
685 (f) If -A or -C was used with -c (count only), some lines of context were
686 accidentally printed for the final match.
687
688 (g) Added the -H (--with-filename) option.
689
690 (h) The combination of options -rh failed to suppress file names for files
691 that were found from directory arguments.
692
693 (i) Added the -D (--devices) and -d (--directories) options.
694
695 (j) Added the -F (--fixed-strings) option.
696
697 (k) Allow "-" to be used as a file name for -f as well as for a data file.
698
699 (l) Added the --colo(u)r option.
700
701 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
702 is not present by default.
703
704 3. A nasty bug was discovered in the handling of recursive patterns, that is,
705 items such as (?R) or (?1), when the recursion could match a number of
706 alternatives. If it matched one of the alternatives, but subsequently,
707 outside the recursion, there was a failure, the code tried to back up into
708 the recursion. However, because of the way PCRE is implemented, this is not
709 possible, and the result was an incorrect result from the match.
710
711 In order to prevent this happening, the specification of recursion has
712 been changed so that all such subpatterns are automatically treated as
713 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
714
715 4. I had overlooked the fact that, in some locales, there are characters for
716 which isalpha() is true but neither isupper() nor islower() are true. In
717 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
718 and ordfeminine) are like this. This affected the treatment of \w and \W
719 when they appeared in character classes, but not when they appeared outside
720 a character class. The bit map for "word" characters is now created
721 separately from the results of isalnum() instead of just taking it from the
722 upper, lower, and digit maps. (Plus the underscore character, of course.)
723
724 5. The above bug also affected the handling of POSIX character classes such as
725 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
726 permanent tables. Instead, the bit maps for such a class were previously
727 created as the appropriate unions of the upper, lower, and digit bitmaps.
728 Now they are created by subtraction from the [[:word:]] class, which has
729 its own bitmap.
730
731 6. The [[:blank:]] character class matches horizontal, but not vertical space.
732 It is created by subtracting the vertical space characters (\x09, \x0a,
733 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
734 subtraction was done in the overall bitmap for a character class, meaning
735 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
736 be recognized. This bug has been fixed.
737
738 7. Patches from the folks at Google:
739
740 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
741 real life, but is still worth protecting against".
742
743 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
744 regular expressions".
745
746 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
747 have it.
748
749 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
750 "configure" and the latter not, in order to fix a problem somebody had
751 with compiling the Arg class on HP-UX.
752
753 (e) Improve the error-handling of the C++ wrapper a little bit.
754
755 (f) New tests for checking recursion limiting.
756
757 8. The pcre_memmove() function, which is used only if the environment does not
758 have a standard memmove() function (and is therefore rarely compiled),
759 contained two bugs: (a) use of int instead of size_t, and (b) it was not
760 returning a result (though PCRE never actually uses the result).
761
762 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
763 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
764 returned instead of calling malloc() with an overflowing number that would
765 most likely cause subsequent chaos.
766
767 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
768
769 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
770 with this option is matched, the nmatch and pmatch options of regexec() are
771 ignored.
772
773 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
774 provided in case anyone wants to the the POSIX interface with UTF-8
775 strings.
776
777 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
778 C++ linking (needed for some HP-UX environments).
779
780 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
781 (unused parameter) and in the pcre_printint() function (omitted "default"
782 switch label when the default is to do nothing).
783
784 15. Added some code to make it possible, when PCRE is compiled as a C++
785 library, to replace subject pointers for pcre_exec() with a smart pointer
786 class, thus making it possible to process discontinuous strings.
787
788 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
789 much the same function. They were added by different people who were trying
790 to make PCRE easy to compile on non-Unix systems. It has been suggested
791 that PCRE_EXPORT be abolished now that there is more automatic apparatus
792 for compiling on Windows systems. I have therefore replaced it with
793 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
794 defaults to "extern" for C or "extern C" for C++, which works fine on
795 Unix-like systems. It is now possible to override the value of PCRE_DATA_
796 SCOPE with something explicit in config.h. In addition:
797
798 (a) pcreposix.h still had just "extern" instead of either of these macros;
799 I have replaced it with PCRE_DATA_SCOPE.
800
801 (b) Functions such as _pcre_xclass(), which are internal to the library,
802 but external in the C sense, all had PCRE_EXPORT in their definitions.
803 This is apparently wrong for the Windows case, so I have removed it.
804 (It makes no difference on Unix-like systems.)
805
806 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
807 of recursive calls to match(). This is different to MATCH_LIMIT because
808 that limits the total number of calls to match(), not all of which increase
809 the depth of recursion. Limiting the recursion depth limits the amount of
810 stack (or heap if NO_RECURSE is set) that is used. The default can be set
811 when PCRE is compiled, and changed at run time. A patch from Google adds
812 this functionality to the C++ interface.
813
814 18. Changes to the handling of Unicode character properties:
815
816 (a) Updated the table to Unicode 4.1.0.
817
818 (b) Recognize characters that are not in the table as "Cn" (undefined).
819
820 (c) I revised the way the table is implemented to a much improved format
821 which includes recognition of ranges. It now supports the ranges that
822 are defined in UnicodeData.txt, and it also amalgamates other
823 characters into ranges. This has reduced the number of entries in the
824 table from around 16,000 to around 3,000, thus reducing its size
825 considerably. I realized I did not need to use a tree structure after
826 all - a binary chop search is just as efficient. Having reduced the
827 number of entries, I extended their size from 6 bytes to 8 bytes to
828 allow for more data.
829
830 (d) Added support for Unicode script names via properties such as \p{Han}.
831
832 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
833 matching that character.
834
835 20. When matching a repeated Unicode property with a minimum greater than zero,
836 (for example \pL{2,}), PCRE could look past the end of the subject if it
837 reached it while seeking the minimum number of characters. This could
838 happen only if some of the characters were more than one byte long, because
839 there is a check for at least the minimum number of bytes.
840
841 21. Refactored the implementation of \p and \P so as to be more general, to
842 allow for more different types of property in future. This has changed the
843 compiled form incompatibly. Anybody with saved compiled patterns that use
844 \p or \P will have to recompile them.
845
846 22. Added "Any" and "L&" to the supported property types.
847
848 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
849 but give a compile time error if the value is greater than 0xff.
850
851 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
852 accidentally not being installed or uninstalled.
853
854 25. The pcre.h file was built from pcre.h.in, but the only changes that were
855 made were to insert the current release number. This seemed silly, because
856 it made things harder for people building PCRE on systems that don't run
857 "configure". I have turned pcre.h into a distributed file, no longer built
858 by "configure", with the version identification directly included. There is
859 no longer a pcre.h.in file.
860
861 However, this change necessitated a change to the pcre-config script as
862 well. It is built from pcre-config.in, and one of the substitutions was the
863 release number. I have updated configure.ac so that ./configure now finds
864 the release number by grepping pcre.h.
865
866 26. Added the ability to run the tests under valgrind.
867
868
869 Version 6.4 05-Sep-05
870 ---------------------
871
872 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
873 "--" to be printed when multiple files were scanned, even when none of the
874 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
875 consider it to be a bug, and have restored the previous behaviour.
876
877 2. A couple of code tidies to get rid of compiler warnings.
878
879 3. The pcretest program used to cheat by referring to symbols in the library
880 whose names begin with _pcre_. These are internal symbols that are not
881 really supposed to be visible externally, and in some environments it is
882 possible to suppress them. The cheating is now confined to including
883 certain files from the library's source, which is a bit cleaner.
884
885 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
886 file's purpose clearer.
887
888 5. Reorganized pcre_ucp_findchar().
889
890
891 Version 6.3 15-Aug-05
892 ---------------------
893
894 1. The file libpcre.pc.in did not have general read permission in the tarball.
895
896 2. There were some problems when building without C++ support:
897
898 (a) If C++ support was not built, "make install" and "make test" still
899 tried to test it.
900
901 (b) There were problems when the value of CXX was explicitly set. Some
902 changes have been made to try to fix these, and ...
903
904 (c) --disable-cpp can now be used to explicitly disable C++ support.
905
906 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
907 backslash in a target when C++ was disabled. This confuses some
908 versions of "make", apparently. Using an intermediate variable solves
909 this. (Same for CPP_LOBJ.)
910
911 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
912 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
913 necessary on certain architectures.
914
915 4. Added a setting of -export-symbols-regex to the link command to remove
916 those symbols that are exported in the C sense, but actually are local
917 within the library, and not documented. Their names all begin with
918 "_pcre_". This is not a perfect job, because (a) we have to except some
919 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
920 available (and never for static libraries). I have made a note to try to
921 find a way round (a) in the future.
922
923
924 Version 6.2 01-Aug-05
925 ---------------------
926
927 1. There was no test for integer overflow of quantifier values. A construction
928 such as {1111111111111111} would give undefined results. What is worse, if
929 a minimum quantifier for a parenthesized subpattern overflowed and became
930 negative, the calculation of the memory size went wrong. This could have
931 led to memory overwriting.
932
933 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
934
935 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
936 operating environments where this matters.
937
938 4. Applied Giuseppe Maxia's patch to add additional features for controlling
939 PCRE options from within the C++ wrapper.
940
941 5. Named capturing subpatterns were not being correctly counted when a pattern
942 was compiled. This caused two problems: (a) If there were more than 100
943 such subpatterns, the calculation of the memory needed for the whole
944 compiled pattern went wrong, leading to an overflow error. (b) Numerical
945 back references of the form \12, where the number was greater than 9, were
946 not recognized as back references, even though there were sufficient
947 previous subpatterns.
948
949 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
950 versions of gcc, e.g. 2.95.4.
951
952
953 Version 6.1 21-Jun-05
954 ---------------------
955
956 1. There was one reference to the variable "posix" in pcretest.c that was not
957 surrounded by "#if !defined NOPOSIX".
958
959 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
960 the cross-check on the old pcre_info() function, for the benefit of the
961 cut-down version of PCRE that is currently imported into Exim.
962
963 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
964 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
965 patterns that start (?i) but I don't think that matters. The use of (?i) is
966 just an example; this all applies to the other options as well.
967
968 4. Since libtool seems to echo the compile commands it is issuing, the output
969 from "make" can be reduced a bit by putting "@" in front of each libtool
970 compile command.
971
972 5. Patch from the folks at Google for configure.in to be a bit more thorough
973 in checking for a suitable C++ installation before trying to compile the
974 C++ stuff. This should fix a reported problem when a compiler was present,
975 but no suitable headers.
976
977 6. The man pages all had just "PCRE" as their title. I have changed them to
978 be the relevant file name. I have also arranged that these names are
979 retained in the file doc/pcre.txt, which is a concatenation in text format
980 of all the man pages except the little individual ones for each function.
981
982 7. The NON-UNIX-USE file had not been updated for the different set of source
983 files that come with release 6. I also added a few comments about the C++
984 wrapper.
985
986
987 Version 6.0 07-Jun-05
988 ---------------------
989
990 1. Some minor internal re-organization to help with my DFA experiments.
991
992 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
993 didn't matter for the library itself when fully configured, but did matter
994 when compiling without UCP support, or within Exim, where the ucp files are
995 not imported.
996
997 3. Refactoring of the library code to split up the various functions into
998 different source modules. The addition of the new DFA matching code (see
999 below) to a single monolithic source would have made it really too
1000 unwieldy, quite apart from causing all the code to be include in a
1001 statically linked application, when only some functions are used. This is
1002 relevant even without the DFA addition now that patterns can be compiled in
1003 one application and matched in another.
1004
1005 The downside of splitting up is that there have to be some external
1006 functions and data tables that are used internally in different modules of
1007 the library but which are not part of the API. These have all had their
1008 names changed to start with "_pcre_" so that they are unlikely to clash
1009 with other external names.
1010
1011 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1012 a different (DFA) algorithm. Although it is slower than the original
1013 function, it does have some advantages for certain types of matching
1014 problem.
1015
1016 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1017 including restarting after a partial match.
1018
1019 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1020 defined when compiling for Windows was sent to me. I have put it into the
1021 code, though I have no means of testing or verifying it.
1022
1023 7. Added the pcre_refcount() auxiliary function.
1024
1025 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1026 match before or at the first newline in the subject string. In pcretest,
1027 the /f option on a pattern can be used to set this.
1028
1029 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1030 would behave wrongly. This has been present in PCRE since release 4.0.
1031
1032 10. A number of changes to the pcregrep command:
1033
1034 (a) Refactored how -x works; insert ^(...)$ instead of setting
1035 PCRE_ANCHORED and checking the length, in preparation for adding
1036 something similar for -w.
1037
1038 (b) Added the -w (match as a word) option.
1039
1040 (c) Refactored the way lines are read and buffered so as to have more
1041 than one at a time available.
1042
1043 (d) Implemented a pcregrep test script.
1044
1045 (e) Added the -M (multiline match) option. This allows patterns to match
1046 over several lines of the subject. The buffering ensures that at least
1047 8K, or the rest of the document (whichever is the shorter) is available
1048 for matching (and similarly the previous 8K for lookbehind assertions).
1049
1050 (f) Changed the --help output so that it now says
1051
1052 -w, --word-regex(p)
1053
1054 instead of two lines, one with "regex" and the other with "regexp"
1055 because that confused at least one person since the short forms are the
1056 same. (This required a bit of code, as the output is generated
1057 automatically from a table. It wasn't just a text change.)
1058
1059 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1060 option but starts with a hyphen. Could be a pattern or a path name
1061 starting with a hyphen, for instance.
1062
1063 (h) "-" can be given as a file name to represent stdin.
1064
1065 (i) When file names are being printed, "(standard input)" is used for
1066 the standard input, for compatibility with GNU grep. Previously
1067 "<stdin>" was used.
1068
1069 (j) The option --label=xxx can be used to supply a name to be used for
1070 stdin when file names are being printed. There is no short form.
1071
1072 (k) Re-factored the options decoding logic because we are going to add
1073 two more options that take data. Such options can now be given in four
1074 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1075
1076 (l) Added the -A, -B, and -C options for requesting that lines of context
1077 around matches be printed.
1078
1079 (m) Added the -L option to print the names of files that do not contain
1080 any matching lines, that is, the complement of -l.
1081
1082 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1083 continue to scan other files.
1084
1085 (o) The -s option was incorrectly implemented. For compatibility with other
1086 greps, it now suppresses the error message for a non-existent or non-
1087 accessible file (but not the return code). There is a new option called
1088 -q that suppresses the output of matching lines, which was what -s was
1089 previously doing.
1090
1091 (p) Added --include and --exclude options to specify files for inclusion
1092 and exclusion when recursing.
1093
1094 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1095 Hopefully, it now does.
1096
1097 12. Missing cast in pcre_study().
1098
1099 13. Added an "uninstall" target to the makefile.
1100
1101 14. Replaced "extern" in the function prototypes in Makefile.in with
1102 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1103 world, but is set differently for Windows.
1104
1105 15. Added a second compiling function called pcre_compile2(). The only
1106 difference is that it has an extra argument, which is a pointer to an
1107 integer error code. When there is a compile-time failure, this is set
1108 non-zero, in addition to the error test pointer being set to point to an
1109 error message. The new argument may be NULL if no error number is required
1110 (but then you may as well call pcre_compile(), which is now just a
1111 wrapper). This facility is provided because some applications need a
1112 numeric error indication, but it has also enabled me to tidy up the way
1113 compile-time errors are handled in the POSIX wrapper.
1114
1115 16. Added VPATH=.libs to the makefile; this should help when building with one
1116 prefix path and installing with another. (Or so I'm told by someone who
1117 knows more about this stuff than I do.)
1118
1119 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1120 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1121 match everything, including newlines. This is not POSIX-compatible, but
1122 somebody wanted the feature. From pcretest it can be activated by using
1123 both the P and the s flags.
1124
1125 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1126
1127 19. libpcre.pc was being incorrectly installed as executable.
1128
1129 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1130 it now also looks for '\r' so that it will work unmodified on Windows.
1131
1132 21. Added Google's contributed C++ wrapper to the distribution.
1133
1134 22. Added some untidy missing memory free() calls in pcretest, to keep
1135 Electric Fence happy when testing.
1136
1137
1138
1139 Version 5.0 13-Sep-04
1140 ---------------------
1141
1142 1. Internal change: literal characters are no longer packed up into items
1143 containing multiple characters in a single byte-string. Each character
1144 is now matched using a separate opcode. However, there may be more than one
1145 byte in the character in UTF-8 mode.
1146
1147 2. The pcre_callout_block structure has two new fields: pattern_position and
1148 next_item_length. These contain the offset in the pattern to the next match
1149 item, and its length, respectively.
1150
1151 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1152 insertion of callouts before each pattern item. Added the /C option to
1153 pcretest to make use of this.
1154
1155 4. On the advice of a Windows user, the lines
1156
1157 #if defined(_WIN32) || defined(WIN32)
1158 _setmode( _fileno( stdout ), 0x8000 );
1159 #endif /* defined(_WIN32) || defined(WIN32) */
1160
1161 have been added to the source of pcretest. This apparently does useful
1162 magic in relation to line terminators.
1163
1164 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1165 for the benefit of those environments where the "b" makes a difference.
1166
1167 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1168 to know about it. I have put a hack into configure.in that adds in code
1169 to set GCC=yes if CC=icc. This seems to end up at a point in the
1170 generated configure script that is early enough to affect the setting of
1171 compiler options, which is what is needed, but I have no means of testing
1172 whether it really works. (The user who reported this had patched the
1173 generated configure script, which of course I cannot do.)
1174
1175 LATER: After change 22 below (new libtool files), the configure script
1176 seems to know about icc (and also ecc). Therefore, I have commented out
1177 this hack in configure.in.
1178
1179 7. Added support for pkg-config (2 patches were sent in).
1180
1181 8. Negated POSIX character classes that used a combination of internal tables
1182 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1183 [[:^ascii]]. Typically, they would match almost any characters. The other
1184 POSIX classes were not broken in this way.
1185
1186 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1187 to find the match, as PCRE was deluded into thinking that the match had to
1188 start at the start point or following a newline. The same bug applied to
1189 patterns with negative forward assertions or any backward assertions
1190 preceding ".*" at the start, unless the pattern required a fixed first
1191 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1192
1193 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1194 starting at the last subject character, bytes beyond the end of the subject
1195 string were read.
1196
1197 11. Renamed the variable "class" as "classbits" to make life easier for C++
1198 users. (Previously there was a macro definition, but it apparently wasn't
1199 enough.)
1200
1201 12. Added the new field "tables" to the extra data so that tables can be passed
1202 in at exec time, or the internal tables can be re-selected. This allows
1203 a compiled regex to be saved and re-used at a later time by a different
1204 program that might have everything at different addresses.
1205
1206 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1207 -R library as well as a -L library.
1208
1209 14. The debugging options of pcretest (-d on the command line or D on a
1210 pattern) showed incorrect output for anything following an extended class
1211 that contained multibyte characters and which was followed by a quantifier.
1212
1213 15. Added optional support for general category Unicode character properties
1214 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1215 support. It adds about 90K to the size of the library. The meanings of the
1216 inbuilt class escapes such as \d and \s have NOT been changed.
1217
1218 16. Updated pcredemo.c to include calls to free() to release the memory for the
1219 compiled pattern.
1220
1221 17. The generated file chartables.c was being created in the source directory
1222 instead of in the building directory. This caused the build to fail if the
1223 source directory was different from the building directory, and was
1224 read-only.
1225
1226 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1227 file. No doubt somebody will tell me if they don't make sense... Also added
1228 Dan Mooney's comments about building on OpenVMS.
1229
1230 19. Added support for partial matching via the PCRE_PARTIAL option for
1231 pcre_exec() and the \P data escape in pcretest.
1232
1233 20. Extended pcretest with 3 new pattern features:
1234
1235 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1236 write the compiled pattern to the file whose name is "rest-of-line".
1237 This is a straight binary dump of the data, with the saved pointer to
1238 the character tables forced to be NULL. The study data, if any, is
1239 written too. After writing, pcretest reads a new pattern.
1240
1241 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1242 compiled pattern from the given file. There must not be any
1243 occurrences of "<" in the file name (pretty unlikely); if there are,
1244 pcretest will instead treat the initial "<" as a pattern delimiter.
1245 After reading in the pattern, pcretest goes on to read data lines as
1246 usual.
1247
1248 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1249 and 16-bit fields in a compiled pattern, to simulate a pattern that
1250 was compiled on a host of opposite endianness.
1251
1252 21. The pcre-exec() function can now cope with patterns that were compiled on
1253 hosts of opposite endianness, with this restriction:
1254
1255 As for any compiled expression that is saved and used later, the tables
1256 pointer field cannot be preserved; the extra_data field in the arguments
1257 to pcre_exec() should be used to pass in a tables address if a value
1258 other than the default internal tables were used at compile time.
1259
1260 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1261 now diagnosed as an error. Previously, most of the time, a negative number
1262 would have been treated as zero, but if in addition "ovector" was passed as
1263 NULL, a crash could occur.
1264
1265 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1266 new versions from the libtool 1.5 distribution (the last one is a copy of
1267 a file called libtool.m4). This seems to have fixed the need to patch
1268 "configure" to support Darwin 1.3 (which I used to do). However, I still
1269 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1270 workstation).
1271
1272 24. Changed the PCRE licence to be the more standard "BSD" licence.
1273
1274
1275 Version 4.5 01-Dec-03
1276 ---------------------
1277
1278 1. There has been some re-arrangement of the code for the match() function so
1279 that it can be compiled in a version that does not call itself recursively.
1280 Instead, it keeps those local variables that need separate instances for
1281 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1282 needs to "recurse". Keeping track of where control must go is done by means
1283 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1284 hide most of the details from the main code, and operates only if
1285 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1286 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1287 operating.
1288
1289 To make it easier for callers to provide specially tailored get/free
1290 functions for this usage, two new functions, pcre_stack_malloc, and
1291 pcre_stack_free, are used. They are always called in strict stacking order,
1292 and the size of block requested is always the same.
1293
1294 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1295 PCRE has been compiled to use the stack or the heap for recursion. The
1296 -C option of pcretest uses this to show which version is compiled.
1297
1298 A new data escape \S, is added to pcretest; it causes the amounts of store
1299 obtained and freed by both kinds of malloc/free at match time to be added
1300 to the output.
1301
1302 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1303 what's available on my current Linux desktop machine.
1304
1305 3. When matching a UTF-8 string, the test for a valid string at the start has
1306 been extended. If start_offset is not zero, PCRE now checks that it points
1307 to a byte that is the start of a UTF-8 character. If not, it returns
1308 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1309 this is necessary because there may be backward assertions in the pattern.
1310 When matching the same subject several times, it may save resources to use
1311 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1312
1313 4. The code for checking the validity of UTF-8 strings has been tightened so
1314 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1315 containing "overlong sequences".
1316
1317 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1318 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1319 should have been "&", but it just so happened that all the cases this let
1320 through by mistake were picked up later in the function.
1321
1322 6. I had used a variable called "isblank" - this is a C99 function, causing
1323 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1324
1325 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1326 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1327 so the version is shown; (c) stop "make test" echoing "./RunTest".
1328
1329 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1330
1331 9. The prototype for memmove() for systems that don't have it was using
1332 size_t, but the inclusion of the header that defines size_t was later. I've
1333 moved the #includes for the C headers earlier to avoid this.
1334
1335 10. Added some adjustments to the code to make it easier to compiler on certain
1336 special systems:
1337
1338 (a) Some "const" qualifiers were missing.
1339 (b) Added the macro EXPORT before all exported functions; by default this
1340 is defined to be empty.
1341 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1342 that it reads its output file name as an argument instead of writing
1343 to the standard output and assuming this can be redirected.
1344
1345 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1346 class containing characters with values greater than 255, PCRE compilation
1347 went into a loop.
1348
1349 12. A recursive reference to a subpattern that was within another subpattern
1350 that had a minimum quantifier of zero caused PCRE to crash. For example,
1351 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1352 recursion. If the recursively-called subpattern itself had a zero repeat,
1353 that was OK.
1354
1355 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1356 buffer into which it was copied (for escape processing) was still set at
1357 1024, so long lines caused crashes.
1358
1359 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1360 "internal error: code overflow...". This applied to any character class
1361 that was followed by a possessive quantifier.
1362
1363 15. Modified the Makefile to add libpcre.la as a prerequisite for
1364 libpcreposix.la because I was told this is needed for a parallel build to
1365 work.
1366
1367 16. If a pattern that contained .* following optional items at the start was
1368 studied, the wrong optimizing data was generated, leading to matching
1369 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1370 matching string must start with a or b or c. The correct conclusion for
1371 this pattern is that a match can start with any character.
1372
1373
1374 Version 4.4 13-Aug-03
1375 ---------------------
1376
1377 1. In UTF-8 mode, a character class containing characters with values between
1378 127 and 255 was not handled correctly if the compiled pattern was studied.
1379 In fixing this, I have also improved the studying algorithm for such
1380 classes (slightly).
1381
1382 2. Three internal functions had redundant arguments passed to them. Removal
1383 might give a very teeny performance improvement.
1384
1385 3. Documentation bug: the value of the capture_top field in a callout is *one
1386 more than* the number of the hightest numbered captured substring.
1387
1388 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1389 in incorrectly linking with a previously installed version. They now link
1390 explicitly with libpcre.la.
1391
1392 5. configure.in no longer needs to recognize Cygwin specially.
1393
1394 6. A problem in pcre.in for Windows platforms is fixed.
1395
1396 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1397 pcretest, it used to include the size of the study block as part of its
1398 output. Unfortunately, the structure contains a field that has a different
1399 size on different hardware architectures. This meant that the tests that
1400 showed this size failed. As the block is currently always of a fixed size,
1401 this information isn't actually particularly useful in pcretest output, so
1402 I have just removed it.
1403
1404 8. Three pre-processor statements accidentally did not start in column 1.
1405 Sadly, there are *still* compilers around that complain, even though
1406 standard C has not required this for well over a decade. Sigh.
1407
1408 9. In pcretest, the code for checking callouts passed small integers in the
1409 callout_data field, which is a void * field. However, some picky compilers
1410 complained about the casts involved for this on 64-bit systems. Now
1411 pcretest passes the address of the small integer instead, which should get
1412 rid of the warnings.
1413
1414 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1415 both compile and run time, and gives an error if an invalid UTF-8 sequence
1416 is found. There is a option for disabling this check in cases where the
1417 string is known to be correct and/or the maximum performance is wanted.
1418
1419 11. In response to a bug report, I changed one line in Makefile.in from
1420
1421 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1422 to
1423 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1424
1425 to look similar to other lines, but I have no way of telling whether this
1426 is the right thing to do, as I do not use Windows. No doubt I'll get told
1427 if it's wrong...
1428
1429
1430 Version 4.3 21-May-03
1431 ---------------------
1432
1433 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1434 Makefile.
1435
1436 2. Some refactoring to improve the quality of the code:
1437
1438 (i) The utf8_table... variables are now declared "const".
1439
1440 (ii) The code for \cx, which used the "case flipping" table to upper case
1441 lower case letters, now just substracts 32. This is ASCII-specific,
1442 but the whole concept of \cx is ASCII-specific, so it seems
1443 reasonable.
1444
1445 (iii) PCRE was using its character types table to recognize decimal and
1446 hexadecimal digits in the pattern. This is silly, because it handles
1447 only 0-9, a-f, and A-F, but the character types table is locale-
1448 specific, which means strange things might happen. A private
1449 table is now used for this - though it costs 256 bytes, a table is
1450 much faster than multiple explicit tests. Of course, the standard
1451 character types table is still used for matching digits in subject
1452 strings against \d.
1453
1454 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1455 ending in _t are). So I've renamed it as ESC_tee.
1456
1457 3. The first argument for regexec() in the POSIX wrapper should have been
1458 defined as "const".
1459
1460 4. Changed pcretest to use malloc() for its buffers so that they can be
1461 Electric Fenced for debugging.
1462
1463 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1464 to read one or more bytes before the start of the subject string. Often this
1465 had no effect on PCRE's behaviour, but in some circumstances it could
1466 provoke a segmentation fault.
1467
1468 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1469 to try to read one or more bytes before the start of the subject string.
1470
1471 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1472 UTF-8 support could misbehave in various ways if the subject string
1473 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1474 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1475 back over UTF-8 characters.)
1476
1477
1478 Version 4.2 14-Apr-03
1479 ---------------------
1480
1481 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1482
1483 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1484 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1485 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1486 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1487 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1488 and BUILD_EXEEXT
1489 Note: automatic setting of the BUILD variables is not yet working
1490 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1491 compile-time but not at link-time
1492 [LINK]: use for linking executables only
1493 make different versions for Windows and non-Windows
1494 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1495 libraries
1496 [LINK_FOR_BUILD]: new variable
1497 [OBJEXT]: use throughout
1498 [EXEEXT]: use throughout
1499 <winshared>: new target
1500 <wininstall>: new target
1501 <dftables.o>: use native compiler
1502 <dftables>: use native linker
1503 <install>: handle Windows platform correctly
1504 <clean>: ditto
1505 <check>: ditto
1506 copy DLL to top builddir before testing
1507
1508 As part of these changes, -no-undefined was removed again. This was reported
1509 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1510 in any case.
1511
1512 3. Some tidies to get rid of compiler warnings:
1513
1514 . In the match_data structure, match_limit was an unsigned long int, whereas
1515 match_call_count was an int. I've made them both unsigned long ints.
1516
1517 . In pcretest the fact that a const uschar * doesn't automatically cast to
1518 a void * provoked a warning.
1519
1520 . Turning on some more compiler warnings threw up some "shadow" variables
1521 and a few more missing casts.
1522
1523 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1524 option, a class that contained a single character with a value between 128
1525 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1526
1527 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1528 option, a class that contained several characters, but with at least one
1529 whose value was between 128 and 255 caused PCRE to crash.
1530
1531
1532 Version 4.1 12-Mar-03
1533 ---------------------
1534
1535 1. Compiling with gcc -pedantic found a couple of places where casts were
1536 needed, and a string in dftables.c that was longer than standard compilers are
1537 required to support.
1538
1539 2. Compiling with Sun's compiler found a few more places where the code could
1540 be tidied up in order to avoid warnings.
1541
1542 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1543 first of these names is deprecated in the latest Autoconf in favour of the name
1544 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1545 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1546 analogy I have changed it to CFLAGS_FOR_BUILD.
1547
1548 4. Added -no-undefined to the linking command in the Makefile, because this is
1549 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1550 linking step for the pcreposix library.
1551
1552 5. PCRE was failing to diagnose the case of two named groups with the same
1553 name.
1554
1555 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1556 literal character that is needed in the subject for a match, and scans along to
1557 ensure that it is present before embarking on the full matching process. This
1558 saves time in cases of nested unlimited repeats that are never going to match.
1559 Problem: the scan can take a lot of time if the subject is very long (e.g.
1560 megabytes), thus penalizing straightforward matches. It is now done only if the
1561 amount of subject to be scanned is less than 1000 bytes.
1562
1563 7. A lesser problem with the same optimization is that it was recording the
1564 first character of an anchored pattern as "needed", thus provoking a search
1565 right along the subject, even when the first match of the pattern was going to
1566 fail. The "needed" character is now not set for anchored patterns, unless it
1567 follows something in the pattern that is of non-fixed length. Thus, it still
1568 fulfils its original purpose of finding quick non-matches in cases of nested
1569 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1570
1571
1572 Version 4.0 17-Feb-03
1573 ---------------------
1574
1575 1. If a comment in an extended regex that started immediately after a meta-item
1576 extended to the end of string, PCRE compiled incorrect data. This could lead to
1577 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1578
1579 2. Moved to autoconf 2.53 and libtool 1.4.2.
1580
1581 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1582 the special perltest8 script is no longer needed - all the tests can be run
1583 from a single perltest script.
1584
1585 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1586 by \s. It has now been removed in PCRE. This means it isn't recognized as
1587 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1588 class [:space:] *does* include VT, thereby creating a mess.
1589
1590 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1591 space and tab.
1592
1593 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1594 its new features into the main test script, reducing the number of scripts.
1595
1596 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1597 were backward compatible, and made the (?i) apply to the whole pattern, as if
1598 /i were given. Now it behaves more logically, and applies the option setting
1599 only to what follows. PCRE has been changed to follow suit. However, if it
1600 finds options settings right at the start of the pattern, it extracts them into
1601 the global options, as before. Thus, they show up in the info data.
1602
1603 8. Added support for the \Q...\E escape sequence. Characters in between are
1604 treated as literals. This is slightly different from Perl in that $ and @ are
1605 also handled as literals inside the quotes. In Perl, they will cause variable
1606 interpolation. Note the following examples:
1607
1608 Pattern PCRE matches Perl matches
1609
1610 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1611 \Qabc\$xyz\E abc\$xyz abc\$xyz
1612 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1613
1614 For compatibility with Perl, \Q...\E sequences are recognized inside character
1615 classes as well as outside them.
1616
1617 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1618 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1619 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1620 signed/unsigned warnings.
1621
1622 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1623 option for pcretest, so I've replaced it by a simple function that does just
1624 that job.
1625
1626 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1627 "pcregrep -".
1628
1629 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1630 Java package. This provides some syntactic sugar for simple cases of what my
1631 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1632 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1633 item, you can use this simplified notation. Note that only makes sense with
1634 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1635 greediness, whatever the setting of the PCRE_UNGREEDY option.
1636
1637 13. A change of greediness default within a pattern was not taking effect at
1638 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1639 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1640 was abstracted outside.
1641
1642 14. PCRE now supports the \G assertion. It is true when the current matching
1643 position is at the start point of the match. This differs from \A when the
1644 starting offset is non-zero. Used with the /g option of pcretest (or similar
1645 code), it works in the same way as it does for Perl's /g option. If all
1646 alternatives of a regex begin with \G, the expression is anchored to the start
1647 match position, and the "anchored" flag is set in the compiled expression.
1648
1649 15. Some bugs concerning the handling of certain option changes within patterns
1650 have been fixed. These applied to options other than (?ims). For example,
1651 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1652 been the other way round. Some of this was related to change 7 above.
1653
1654 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1655 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1656 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1657 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1658
1659 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1660 mode. Unlike ".", it always matches newline, whatever the setting of
1661 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1662 assertions. Perl allows it, but it doesn't (in general) work because it can't
1663 calculate the length of the lookbehind. At least, that's the case for Perl
1664 5.8.0 - I've been told they are going to document that it doesn't work in
1665 future.
1666
1667 18. Added an error diagnosis for escapes that PCRE does not support: these are
1668 \L, \l, \N, \P, \p, \U, \u, and \X.
1669
1670 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1671 reading past the end of the pattern in cases such as /[abcd/.
1672
1673 20. PCRE was getting more memory than necessary for patterns with classes that
1674 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1675
1676 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1677 compiling PCRE for use with Virtual Pascal.
1678
1679 22. Small fix to the Makefile to make it work properly if the build is done
1680 outside the source tree.
1681
1682 23. Added a new extension: a condition to go with recursion. If a conditional
1683 subpattern starts with (?(R) the "true" branch is used if recursion has
1684 happened, whereas the "false" branch is used only at the top level.
1685
1686 24. When there was a very long string of literal characters (over 255 bytes
1687 without UTF support, over 250 bytes with UTF support), the computation of how
1688 much memory was required could be incorrect, leading to segfaults or other
1689 strange effects.
1690
1691 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1692 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1693 there was a subsequent back reference to those brackets. This meant that, for
1694 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1695 possible to check for precisely this case. All we can do is abandon the
1696 optimization if .* occurs inside capturing brackets when there are any back
1697 references whatsoever. (See below for a better fix that came later.)
1698
1699 26. The handling of the optimization for finding the first character of a
1700 non-anchored pattern, and for finding a character that is required later in the
1701 match were failing in some cases. This didn't break the matching; it just
1702 failed to optimize when it could. The way this is done has been re-implemented.
1703
1704 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1705
1706 28. Added a new feature that provides some of the functionality that Perl
1707 provides with (?{...}). The facility is termed a "callout". The way it is done
1708 in PCRE is for the caller to provide an optional function, by setting
1709 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1710 global variable. By default it is unset, which disables all calling out. To get
1711 the function called, the regex must include (?C) at appropriate points. This
1712 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1713 This provides a means of identifying different callout points. When PCRE
1714 reaches such a point in the regex, if pcre_callout has been set, the external
1715 function is called. It is provided with data in a structure called
1716 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1717 matching continues; if it returns a non-zero value, the match at the current
1718 point fails. However, backtracking will occur if possible. [This was changed
1719 later and other features added - see item 49 below.]
1720
1721 29. pcretest is upgraded to test the callout functionality. It provides a
1722 callout function that displays information. By default, it shows the start of
1723 the match and the current position in the text. There are some new data escapes
1724 to vary what happens:
1725
1726 \C+ in addition, show current contents of captured substrings
1727 \C- do not supply a callout function
1728 \C!n return 1 when callout number n is reached
1729 \C!n!m return 1 when callout number n is reached for the mth time
1730
1731 30. If pcregrep was called with the -l option and just a single file name, it
1732 output "<stdin>" if a match was found, instead of the file name.
1733
1734 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1735 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1736 pcre_exec(). This saves a malloc/free per call. The default value of
1737 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1738 when configuring.
1739
1740 32. The default maximum size of a compiled pattern is 64K. There have been a
1741 few cases of people hitting this limit. The code now uses macros to handle the
1742 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1743 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1744 configuring. Tests 2 and 5 work only with 2-byte links because they output
1745 debugging information about compiled patterns.
1746
1747 33. Internal code re-arrangements:
1748
1749 (a) Moved the debugging function for printing out a compiled regex into
1750 its own source file (printint.c) and used #include to pull it into
1751 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1752 separate copies.
1753
1754 (b) Defined the list of op-code names for debugging as a macro in
1755 internal.h so that it is next to the definition of the opcodes.
1756
1757 (c) Defined a table of op-code lengths for simpler skipping along compiled
1758 code. This is again a macro in internal.h so that it is next to the
1759 definition of the opcodes.
1760
1761 34. Added support for recursive calls to individual subpatterns, along the
1762 lines of Robin Houston's patch (but implemented somewhat differently).
1763
1764 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1765 allow it to read and process whole directories in Win32. This code was
1766 contributed by Lionel Fourquaux; it has not been tested by me.
1767
1768 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1769 used to name a group. Names consist of alphanumerics and underscores, and must
1770 be unique. Back references use the syntax (?P=name) and recursive calls use
1771 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1772 numbers. The function pcre_fullinfo() can be used after compilation to extract
1773 a name/number map. There are three relevant calls:
1774
1775 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1776 PCRE_INFO_NAMECOUNT yields the number of entries
1777 PCRE_INFO_NAMETABLE yields a pointer to the map.
1778
1779 The map is a vector of fixed-size entries. The size of each entry depends on
1780 the length of the longest name used. The first two bytes of each entry are the
1781 group number, most significant byte first. There follows the corresponding
1782 name, zero terminated. The names are in alphabetical order.
1783
1784 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1785 case instead of 255. Making it the same both with and without UTF-8 support
1786 means that the same test output works with both.
1787
1788 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1789 calling malloc() with a zero argument.
1790
1791 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1792 optimization. I've improved things by keeping a bitmap of backreferences with
1793 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1794 fact referenced, the optimization can be applied. It is unlikely that a
1795 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1796 the match to follow \n) will appear inside brackets with a number greater than
1797 31, but if it does, any back reference > 31 suppresses the optimization.
1798
1799 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1800 of disabling numbered capturing parentheses. Any opening parenthesis that is
1801 not followed by ? behaves as if it were followed by ?: but named parentheses
1802 can still be used for capturing (and they will acquire numbers in the usual
1803 way).
1804
1805 41. Redesigned the return codes from the match() function into yes/no/error so
1806 that errors can be passed back from deep inside the nested calls. A malloc
1807 failure while inside a recursive subpattern call now causes the
1808 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1809
1810 42. It is now possible to set a limit on the number of times the match()
1811 function is called in a call to pcre_exec(). This facility makes it possible to
1812 limit the amount of recursion and backtracking, though not in a directly
1813 obvious way, because the match() function is used in a number of different
1814 circumstances. The count starts from zero for each position in the subject
1815 string (for non-anchored patterns). The default limit is, for compatibility, a
1816 large number, namely 10 000 000. You can change this in two ways:
1817
1818 (a) When configuring PCRE before making, you can use --with-match-limit=n
1819 to set a default value for the compiled library.
1820
1821 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1822 a different value is set. See 45 below.
1823
1824 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1825
1826 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1827 of things that can be changed at compile time. The first argument specifies
1828 what is wanted and the second points to where the information is to be placed.
1829 The current list of available information is:
1830
1831 PCRE_CONFIG_UTF8
1832
1833 The output is an integer that is set to one if UTF-8 support is available;
1834 otherwise it is set to zero.
1835
1836 PCRE_CONFIG_NEWLINE
1837
1838 The output is an integer that it set to the value of the code that is used for
1839 newline. It is either LF (10) or CR (13).
1840
1841 PCRE_CONFIG_LINK_SIZE
1842
1843 The output is an integer that contains the number of bytes used for internal
1844 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1845
1846 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1847
1848 The output is an integer that contains the threshold above which the POSIX
1849 interface uses malloc() for output vectors. See item 31 above.
1850
1851 PCRE_CONFIG_MATCH_LIMIT
1852
1853 The output is an unsigned integer that contains the default limit of the number
1854 of match() calls in a pcre_exec() execution. See 42 above.
1855
1856 44. pcretest has been upgraded by the addition of the -C option. This causes it
1857 to extract all the available output from the new pcre_config() function, and to
1858 output it. The program then exits immediately.
1859
1860 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1861 order to support additional features. One way would have been to define
1862 pcre_exec2() (for example) with extra arguments, but this would not have been
1863 extensible, and would also have required all calls to the original function to
1864 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1865 is used for passing in "extra" data from pcre_study().
1866
1867 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1868 contains the following fields:
1869
1870 flags a bitmap indicating which of the following fields are set
1871 study_data opaque data from pcre_study()
1872 match_limit a way of specifying a limit on match() calls for a specific
1873 call to pcre_exec()
1874 callout_data data for callouts (see 49 below)
1875
1876 The flag bits are also defined in pcre.h, and are
1877
1878 PCRE_EXTRA_STUDY_DATA
1879 PCRE_EXTRA_MATCH_LIMIT
1880 PCRE_EXTRA_CALLOUT_DATA
1881
1882 The pcre_study() function now returns one of these new pcre_extra blocks, with
1883 the actual study data pointed to by the study_data field, and the
1884 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1885 before. That is, this change is entirely upwards-compatible and requires no
1886 change to existing code.
1887
1888 If you want to pass in additional data to pcre_exec(), you can either place it
1889 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1890 block.
1891
1892 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1893 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1894 times with different match limits, until it finds the minimum value needed for
1895 pcre_exec() to complete. The value is then output. This can be instructive; for
1896 most simple matches the number is quite small, but for pathological cases it
1897 gets very large very quickly.
1898
1899 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1900 returns the size of the data block pointed to by the study_data field in a
1901 pcre_extra block, that is, the value that was passed as the argument to
1902 pcre_malloc() when PCRE was getting memory in which to place the information
1903 created by pcre_study(). The fourth argument should point to a size_t variable.
1904 pcretest has been extended so that this information is shown after a successful
1905 pcre_study() call when information about the compiled regex is being displayed.
1906
1907 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1908 because what follows is always an absolute path. (Later: it turns out that this
1909 is more than cosmetic for MinGW, because it doesn't like empty path
1910 components.)
1911
1912 49. Some changes have been made to the callout feature (see 28 above):
1913
1914 (i) A callout function now has three choices for what it returns:
1915
1916 0 => success, carry on matching
1917 > 0 => failure at this point, but backtrack if possible
1918 < 0 => serious error, return this value from pcre_exec()
1919
1920 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1921 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1922 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1923 use by callout functions. It will never be used by PCRE itself.
1924
1925 (ii) The pcre_extra structure (see 45 above) has a void * field called
1926 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1927 pcre_callout_block structure has a field of the same name. The contents of
1928 the field passed in the pcre_extra structure are passed to the callout
1929 function in the corresponding field in the callout block. This makes it
1930 easier to use the same callout-containing regex from multiple threads. For
1931 testing, the pcretest program has a new data escape
1932
1933 \C*n pass the number n (may be negative) as callout_data
1934
1935 If the callout function in pcretest receives a non-zero value as
1936 callout_data, it returns that value.
1937
1938 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1939 there were some redundant $(CFLAGS) in commands that are now specified as
1940 $(LINK), which already includes $(CFLAGS).
1941
1942 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1943 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1944 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1945 one-byte characters throughout. Note that case-insensitive matching applies
1946 only to characters whose values are less than 256. PCRE doesn't support the
1947 notion of cases for higher-valued characters.
1948
1949 (i) A character class whose characters are all within 0-255 is handled as
1950 a bit map, and the map is inverted for negative classes. Previously, a
1951 character > 255 always failed to match such a class; however it should
1952 match if the class was a negative one (e.g. [^ab]). This has been fixed.
1953
1954 (ii) A negated character class with a single character < 255 is coded as
1955 "not this character" (OP_NOT). This wasn't working properly when the test
1956 character was multibyte, either singly or repeated.
1957
1958 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1959 mode, for example: \x{100}{2,3}.
1960
1961 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1962 singly or repeated) now correctly test multibyte characters. However,
1963 PCRE doesn't recognize any characters with values greater than 255 as
1964 digits, spaces, or word characters. Such characters always match \D, \S,
1965 and \W, and never match \d, \s, or \w.
1966
1967 (v) Classes may now contain characters and character ranges with values
1968 greater than 255. For example: [ab\x{100}-\x{400}].
1969
1970 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1971 PCRE in UTF-8 mode.
1972
1973 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1974 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1975 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1976 value.)
1977
1978 53. The single man page has become too large. I have therefore split it up into
1979 a number of separate man pages. These also give rise to individual HTML pages;
1980 these are now put in a separate directory, and there is an index.html page that
1981 lists them all. Some hyperlinking between the pages has been installed.
1982
1983 54. Added convenience functions for handling named capturing parentheses.
1984
1985 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1986 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1987 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1988 are faulted.
1989
1990 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1991 calling configure. These values are used when compiling the dftables.c program
1992 which is run to generate the source of the default character tables. They
1993 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1994 you will need to set these values.
1995
1996 57. Updated the building process for Windows DLL, as provided by Fred Cox.
1997
1998
1999 Version 3.9 02-Jan-02
2000 ---------------------
2001
2002 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2003
2004 2. If --disable-static was given, the building process failed when trying to
2005 build pcretest and pcregrep. (For some reason it was using libtool to compile
2006 them, which is not right, as they aren't part of the library.)
2007
2008
2009 Version 3.8 18-Dec-01
2010 ---------------------
2011
2012 1. The experimental UTF-8 code was completely screwed up. It was packing the
2013 bytes in the wrong order. How dumb can you get?
2014
2015
2016 Version 3.7 29-Oct-01
2017 ---------------------
2018
2019 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2020 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2021 this didn't happen under Solaris 8, where I normally test things.
2022
2023 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2024 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2025 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2026 this fix an hour or so after the initial 3.7 release.)
2027
2028
2029 Version 3.6 23-Oct-01
2030 ---------------------
2031
2032 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2033 offsets passed as NULL with zero offset count.
2034
2035 2. The config.guess and config.sub files had not been updated when I moved to
2036 the latest autoconf.
2037
2038
2039 Version 3.5 15-Aug-01
2040 ---------------------
2041
2042 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2043 had been forgotten.
2044
2045 2. By using declared but undefined structures, we can avoid using "void"
2046 definitions in pcre.h while keeping the internal definitions of the structures
2047 private.
2048
2049 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2050 user point of view, this means that both static and shared libraries are built
2051 by default, but this can be individually controlled. More of the work of
2052 handling this static/shared cases is now inside libtool instead of PCRE's make
2053 file.
2054
2055 4. The pcretest utility is now installed along with pcregrep because it is
2056 useful for users (to test regexs) and by doing this, it automatically gets
2057 relinked by libtool. The documentation has been turned into a man page, so
2058 there are now .1, .txt, and .html versions in /doc.
2059
2060 5. Upgrades to pcregrep:
2061 (i) Added long-form option names like gnu grep.
2062 (ii) Added --help to list all options with an explanatory phrase.
2063 (iii) Added -r, --recursive to recurse into sub-directories.
2064 (iv) Added -f, --file to read patterns from a file.
2065
2066 6. pcre_exec() was referring to its "code" argument before testing that
2067 argument for NULL (and giving an error if it was NULL).
2068
2069 7. Upgraded Makefile.in to allow for compiling in a different directory from
2070 the source directory.
2071
2072 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2073 options bits, the pointer it was passed was to an int instead of to an unsigned
2074 long int. This mattered only on 64-bit systems.
2075
2076 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2077 generated) instead of pcre.in, which it its source. Also made the same change
2078 in several of the .c files.
2079
2080 10. A new release of gcc defines printf() as a macro, which broke pcretest
2081 because it had an ifdef in the middle of a string argument for printf(). Fixed
2082 by using separate calls to printf().
2083
2084 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2085 script, to force use of CR or LF instead of \n in the source. On non-Unix
2086 systems, the value can be set in config.h.
2087
2088 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2089 absolute limit. Changed the text of the error message to make this clear, and
2090 likewise updated the man page.
2091
2092 13. The limit of 99 on the number of capturing subpatterns has been removed.
2093 The new limit is 65535, which I hope will not be a "real" limit.
2094
2095
2096 Version 3.4 22-Aug-00
2097 ---------------------
2098
2099 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2100
2101 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2102
2103
2104 Version 3.3 01-Aug-00
2105 ---------------------
2106
2107 1. If an octal character was given, but the value was greater than \377, it
2108 was not getting masked to the least significant bits, as documented. This could
2109 lead to crashes in some systems.
2110
2111 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2112 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2113
2114 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2115 These just pass their arguments on to (pcre_free)(), but they are provided
2116 because some uses of PCRE bind it to non-C systems that can call its functions,
2117 but cannot call free() or pcre_free() directly.
2118
2119 4. Add "make test" as a synonym for "make check". Corrected some comments in
2120 the Makefile.
2121
2122 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2123 Makefile.
2124
2125 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2126 command called pgrep for grepping around the active processes.
2127
2128 7. Added the beginnings of support for UTF-8 character strings.
2129
2130 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2131 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2132 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2133 out for the ar command.)
2134
2135
2136 Version 3.2 12-May-00
2137 ---------------------
2138
2139 This is purely a bug fixing release.
2140
2141 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2142 of ZA. This was just one example of several cases that could provoke this bug,
2143 which was introduced by change 9 of version 2.00. The code for breaking
2144 infinite loops after an iteration that matches an empty string was't working
2145 correctly.
2146
2147 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2148 when matched against abbab (for example). After matching an empty string, it
2149 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2150 caused it to match further down the string than it should.
2151
2152 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2153 was there because it doesn't seem to be needed, and it causes trouble on some
2154 systems, as it is not a Standard C header. It has been removed.
2155
2156 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2157 were reported on the Macintosh. The changes were from
2158
2159 while ((c = *(++ptr)) != 0 && c != '\n');
2160 to
2161 while ((c = *(++ptr)) != 0 && c != '\n') ;
2162
2163 Totally extraordinary, but if that's what it takes...
2164
2165 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2166 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2167 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2168 assumes the way PCRE uses memmove() (always moving upwards).
2169
2170 6. PCRE is being used in one environment where strchr() is not available. There
2171 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2172 faster code anyway.
2173
2174
2175 Version 3.1 09-Feb-00
2176 ---------------------
2177
2178 The only change in this release is the fixing of some bugs in Makefile.in for
2179 the "install" target:
2180
2181 (1) It was failing to install pcreposix.h.
2182
2183 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2184
2185
2186 Version 3.0 01-Feb-00
2187 ---------------------
2188
2189 1. Add support for the /+ modifier to perltest (to output $` like it does in
2190 pcretest).
2191
2192 2. Add support for the /g modifier to perltest.
2193
2194 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2195 matches null strings.
2196
2197 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2198 pattern. Perl treats empty patterns specially - it reuses the most recent
2199 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2200 effect.
2201
2202 5. The POSIX interface was broken in that it was just handing over the POSIX
2203 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2204 required a bigger vector, with some working space on the end. This means that
2205 the POSIX wrapper now has to get and free some memory, and copy the results.
2206
2207 6. Added some simple autoconf support, placing the test data and the
2208 documentation in separate directories, re-organizing some of the
2209 information files, and making it build pcre-config (a GNU standard). Also added
2210 libtool support for building PCRE as a shared library, which is now the
2211 default.
2212
2213 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2214 09 are not valid octal constants. Single digits will be used for minor values
2215 less than 10.
2216
2217 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2218 existing programs that set these in the POSIX interface can use PCRE without
2219 modification.
2220
2221 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2222 return all that pcre_info() returns, plus additional data. The pcre_info()
2223 function is retained for compatibility, but is considered to be obsolete.
2224
2225 10. Added experimental recursion feature (?R) to handle one common case that
2226 Perl 5.6 will be able to do with (?p{...}).
2227
2228 11. Added support for POSIX character classes like [:alpha:], which Perl is
2229 adopting.
2230
2231
2232 Version 2.08 31-Aug-99
2233 ----------------------
2234
2235 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2236 trying to match at the startoffset position, but instead was moving forward to
2237 the next newline as if a previous match had failed.
2238
2239 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2240 and could get into a loop if a null string was matched other than at the start
2241 of the subject.
2242
2243 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2244 be distinguished at compile time, and for completeness also added PCRE_DATE.
2245
2246 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2247 in GnuWin32 environments.
2248
2249
2250 Version 2.07 29-Jul-99
2251 ----------------------
2252
2253 1. The documentation is now supplied in plain text form and HTML as well as in
2254 the form of man page sources.
2255
2256 2. C++ compilers don't like assigning (void *) values to other pointer types.
2257 In particular this affects malloc(). Although there is no problem in Standard
2258 C, I've put in casts to keep C++ compilers happy.
2259
2260 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2261 should be (const char *).
2262
2263 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2264 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2265 However, I haven't made this a standard facility. The documentation doesn't
2266 mention it, and the Makefile doesn't support it.
2267
2268 5. The Makefile now contains an "install" target, with editable destinations at
2269 the top of the file. The pcretest program is not installed.
2270
2271 6. pgrep -V now gives the PCRE version number and date.
2272
2273 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2274 causing the entire string to be ignored, instead of just the last character.
2275
2276 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2277 non-matching string, it can take a very, very long time, even for strings of
2278 quite modest length, because of the nested recursion. PCRE now does better in
2279 some of these cases. It does this by remembering the last required literal
2280 character in the pattern, and pre-searching the subject to ensure it is present
2281 before running the real match. In other words, it applies a heuristic to detect
2282 some types of certain failure quickly, and in the above example, if presented
2283 with a string that has no trailing " it gives "no match" very quickly.
2284
2285 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2286 other alternatives are tried instead.
2287
2288
2289 Version 2.06 09-Jun-99
2290 ----------------------
2291
2292 1. Change pcretest's output for amount of store used to show just the code
2293 space, because the remainder (the data block) varies in size between 32-bit and
2294 64-bit systems.
2295
2296 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2297 start matching at. This allows lookbehinds to work when searching for multiple
2298 occurrences in a string.
2299
2300 3. Added additional options to pcretest for testing multiple occurrences:
2301
2302 /+ outputs the rest of the string that follows a match
2303 /g loops for multiple occurrences, using the new startoffset argument
2304 /G loops for multiple occurrences by passing an incremented pointer
2305
2306 4. PCRE wasn't doing the "first character" optimization for patterns starting
2307 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2308 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2309 the letter 'x'. On long subject strings, this gives a significant speed-up.
2310
2311
2312 Version 2.05 21-Apr-99
2313 ----------------------
2314
2315 1. Changed the type of magic_number from int to long int so that it works
2316 properly on 16-bit systems.
2317
2318 2. Fixed a bug which caused patterns starting with .* not to work correctly
2319 when the subject string contained newline characters. PCRE was assuming
2320 anchoring for such patterns in all cases, which is not correct because .* will
2321 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2322 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2323 must be retried after every newline in the subject.
2324
2325
2326 Version 2.04 18-Feb-99
2327 ----------------------
2328
2329 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2330 computation of the store needed to hold the pattern was incorrect (too large).
2331 If such patterns were nested a few deep, this could multiply and become a real
2332 problem.
2333
2334 2. Added /M option to pcretest to show the memory requirement of a specific
2335 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2336
2337 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2338 compiled in such a way that the backtracking after subsequent failure was
2339 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2340 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2341
2342
2343 Version 2.03 02-Feb-99
2344 ----------------------
2345
2346 1. Fixed typo and small mistake in man page.
2347
2348 2. Added 4th condition (GPL supersedes if conflict) and created separate
2349 LICENCE file containing the conditions.
2350
2351 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2352 Perl, that is the internal \ allows the delimiter to be included in the
2353 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2354 the final delimiter, add \ to the end of the pattern (to test the error).
2355
2356 4. Added the convenience functions for extracting substrings after a successful
2357 match. Updated pcretest to make it able to test these functions.
2358
2359
2360 Version 2.02 14-Jan-99
2361 ----------------------
2362
2363 1. Initialized the working variables associated with each extraction so that
2364 their saving and restoring doesn't refer to uninitialized store.
2365
2366 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2367 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2368 fix the problem.
2369
2370 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2371 calls, and wasn't printing the correct value for compiling calls. Increased the
2372 default value of LOOPREPEAT, and the number of significant figures in the
2373 times.
2374
2375 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2376
2377 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2378 a building problem on Windows NT with a FAT file system.
2379
2380
2381 Version 2.01 21-Oct-98
2382 ----------------------
2383
2384 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2385 to character tables built by pcre_maketables() in the current locale. If NULL
2386 is passed, the default tables are used.
2387
2388
2389 Version 2.00 24-Sep-98
2390 ----------------------
2391
2392 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2393 it any more.
2394
2395 2. Allow quantification of (?>) groups, and make it work correctly.
2396
2397 3. The first character computation wasn't working for (?>) groups.
2398
2399 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2400 end of the subject) and add 5.005's \z, which really does match only at the
2401 very end of the subject.
2402
2403 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2404
2405 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2406 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2407 localized options. All options to pcre_study() were also removed.
2408
2409 7. Add other new features from 5.005:
2410
2411 $(?<= positive lookbehind
2412 $(?<! negative lookbehind
2413 (?imsx-imsx) added the unsetting capability
2414 such a setting is global if at outer level; local otherwise
2415 (?imsx-imsx:) non-capturing groups with option setting
2416 (?(cond)re|re) conditional pattern matching
2417
2418 A backreference to itself in a repeated group matches the previous
2419 captured string.
2420
2421 8. General tidying up of studying (both automatic and via "study")
2422 consequential on the addition of new assertions.
2423
2424 9. As in 5.005, unlimited repeated groups that could match an empty substring
2425 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2426 runtime if any iteration does actually match an empty substring.
2427
2428 10. Include the RunTest script in the distribution.
2429
2430 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2431 discrepancies, some of which were old and were also with respect to 5.004. They
2432 have now been fixed.
2433
2434
2435 Version 1.09 28-Apr-98
2436 ----------------------
2437
2438 1. A negated single character class followed by a quantifier with a minimum
2439 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2440 program crashes, or just wrong answers. This did not apply to negated classes
2441 containing more than one character, or to minima other than one.
2442
2443
2444 Version 1.08 27-Mar-98
2445 ----------------------
2446
2447 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2448
2449 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2450 latter must appear before anything that relies on it in the pattern.
2451
2452
2453 Version 1.07 16-Feb-98
2454 ----------------------
2455
2456 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2457 repeat of a potentially empty string).
2458
2459
2460 Version 1.06 23-Jan-98
2461 ----------------------
2462
2463 1. Added Markus Oberhumer's little patches for C++.
2464
2465 2. Literal strings longer than 255 characters were broken.
2466
2467
2468 Version 1.05 23-Dec-97
2469 ----------------------
2470
2471 1. Negated character classes containing more than one character were failing if
2472 PCRE_CASELESS was set at run time.
2473
2474
2475 Version 1.04 19-Dec-97
2476 ----------------------
2477
2478 1. Corrected the man page, where some "const" qualifiers had been omitted.
2479
2480 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2481 input syntax.
2482
2483 3. Fixed memory leak which occurred when a regex with back references was
2484 matched with an offsets vector that wasn't big enough. The temporary memory
2485 that is used in this case wasn't being freed if the match failed.
2486
2487 4. Tidied pcretest to ensure it frees memory that it gets.
2488
2489 5. Temporary memory was being obtained in the case where the passed offsets
2490 vector was exactly big enough.
2491
2492 6. Corrected definition of offsetof() from change 5 below.
2493
2494 7. I had screwed up change 6 below and broken the rules for the use of
2495 setjmp(). Now fixed.
2496
2497
2498 Version 1.03 18-Dec-97
2499 ----------------------
2500
2501 1. A erroneous regex with a missing opening parenthesis was correctly
2502 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2503 on some systems.
2504
2505 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2506 it was reported that one broken compiler failed on the former because "code" is
2507 also an independent variable.
2508
2509 3. The erroneous regex a[]b caused an array overrun reference.
2510
2511 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2512 fail on data ending with that character. (It was going on too far, and checking
2513 the next character, typically a binary zero.) This was specific to the
2514 optimized code for single-character negative classes.
2515
2516 5. Added a contributed patch from the TIN world which does the following:
2517
2518 + Add an undef for memmove, in case the the system defines a macro for it.
2519
2520 + Add a definition of offsetof(), in case there isn't one. (I don't know
2521 the reason behind this - offsetof() is part of the ANSI standard - but
2522 it does no harm).
2523
2524 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2525 most of the places where whitespace preceded '#'. I have given up and
2526 allowed the remaining 2 cases to be at the margin.
2527
2528 + Rename some variables in pcre to eliminate shadowing. This seems very
2529 pedantic, but does no harm, of course.
2530
2531 6. Moved the call to setjmp() into its own function, to get rid of warnings
2532 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2533
2534 7. Constructs such as \d{8,} were compiling into the equivalent of
2535 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2536 outcome, but in this particular case used more store than had been allocated,
2537 which caused the bug to be discovered because it threw up an internal error.
2538
2539 8. The debugging code in both pcre and pcretest for outputting the compiled
2540 form of a regex was going wrong in the case of back references followed by
2541 curly-bracketed repeats.
2542
2543
2544 Version 1.02 12-Dec-97
2545 ----------------------
2546
2547 1. Typos in pcre.3 and comments in the source fixed.
2548
2549 2. Applied a contributed patch to get rid of places where it used to remove
2550 'const' from variables, and fixed some signed/unsigned and uninitialized
2551 variable warnings.
2552
2553 3. Added the "runtest" target to Makefile.
2554
2555 4. Set default compiler flag to -O2 rather than just -O.
2556
2557
2558 Version 1.01 19-Nov-97
2559 ----------------------
2560
2561 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2562 like /([ab]*)*/, that is, for classes with more than one character in them.
2563
2564 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2565 as /((?>a*))*/ (a PCRE_EXTRA facility).
2566
2567
2568 Version 1.00 18-Nov-97
2569 ----------------------
2570
2571 1. Added compile-time macros to support systems such as SunOS4 which don't have
2572 memmove() or strerror() but have other things that can be used instead.
2573
2574 2. Arranged that "make clean" removes the executables.
2575
2576
2577 Version 0.99 27-Oct-97
2578 ----------------------
2579
2580 1. Fixed bug in code for optimizing classes with only one character. It was
2581 initializing a 32-byte map regardless, which could cause it to run off the end
2582 of the memory it had got.
2583
2584 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2585
2586
2587 Version 0.98 22-Oct-97
2588 ----------------------
2589
2590 1. Fixed bug in code for handling temporary memory usage when there are more
2591 back references than supplied space in the ovector. This could cause segfaults.
2592
2593
2594 Version 0.97 21-Oct-97
2595 ----------------------
2596
2597 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2598
2599 2. Optimized negated single characters not to use a bit map.
2600
2601 3. Brought error texts together as macro definitions; clarified some of them;
2602 fixed one that was wrong - it said "range out of order" when it meant "invalid
2603 escape sequence".
2604
2605 4. Changed some char * arguments to const char *.
2606
2607 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2608
2609 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2610 pcretest.
2611
2612
2613 Version 0.96 16-Oct-97
2614 ----------------------
2615
2616 1. Added a simple "pgrep" utility to the distribution.
2617
2618 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2619 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2620 where "ddd" means "one or more decimal digits".
2621
2622 3. Fixed serious bug. If a pattern had a back reference, but the call to
2623 pcre_exec() didn't supply a large enough ovector to record the related
2624 identifying subpattern, the match always failed. PCRE now remembers the number
2625 of the largest back reference, and gets some temporary memory in which to save
2626 the offsets during matching if necessary, in order to ensure that
2627 backreferences always work.
2628
2629 4. Increased the compatibility with Perl in a number of ways:
2630
2631 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2632 to request this handling. The option can be set at compile or exec time.
2633
2634 (b) $ matches before a terminating newline by default; an option
2635 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2636 mode). The option can be set at compile or exec time.
2637
2638 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2639 the same as Perl's. If the decimal number it represents is less than 10
2640 or there aren't that many previous left capturing parentheses, an octal
2641 escape is read. Inside a character class, it's always an octal escape,
2642 even if it is a single digit.
2643
2644 (d) An escaped but undefined alphabetic character is taken as a literal,
2645 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2646 escapes.
2647
2648 (e) {0} is now permitted. (The previous item is removed from the compiled
2649 pattern).
2650
2651 5. Changed all the names of code files so that the basic parts are no longer
2652 than 10 characters, and abolished the teeny "globals.c" file.
2653
2654 6. Changed the handling of character classes; they are now done with a 32-byte
2655 bit map always.
2656
2657 7. Added the -d and /D options to pcretest to make it possible to look at the
2658 internals of compilation without having to recompile pcre.
2659
2660
2661 Version 0.95 23-Sep-97
2662 ----------------------
2663
2664 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2665 \x20 at the start of a run of normal characters. These were being treated as
2666 real characters, instead of the source characters being re-checked.
2667
2668
2669 Version 0.94 18-Sep-97
2670 ----------------------
2671
2672 1. The functions are now thread-safe, with the caveat that the global variables
2673 containing pointers to malloc() and free() or alternative functions are the
2674 same for all threads.
2675
2676 2. Get pcre_study() to generate a bitmap of initial characters for non-
2677 anchored patterns when this is possible, and use it if passed to pcre_exec().
2678
2679
2680 Version 0.93 15-Sep-97
2681 ----------------------
2682
2683 1. /(b)|(:+)/ was computing an incorrect first character.
2684
2685 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2686 but not actually doing anything yet.
2687
2688 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2689 as Perl does (e.g. [-az] or [az-]).
2690
2691 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2692 all possible positions.
2693
2694 5. Split up into different modules to avoid including unneeded functions in a
2695 compiled binary. However, compile and exec are still in one module. The "study"
2696 function is split off.
2697
2698 6. The character tables are now in a separate module whose source is generated
2699 by an auxiliary program - but can then be edited by hand if required. There are
2700 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2701 toupper() in the code.
2702
2703 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2704 make them global. Abolish the function for setting them, as the caller can now
2705 set them directly.
2706
2707
2708 Version 0.92 11-Sep-97
2709 ----------------------
2710
2711 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2712 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2713
2714 2. Caseless matching was not working in character classes if the characters in
2715 the pattern were in upper case.
2716
2717 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2718
2719 4. Make PCRE_ANCHORED public and accept as a compile option.
2720
2721 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2722 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2723 pass them.
2724
2725 6. Give an error if bad option bits passed at compile or run time.
2726
2727 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2728 pcretest to cause it to pass that flag.
2729
2730 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2731 options, and the first character, if set.
2732
2733 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2734
2735
2736 Version 0.91 10-Sep-97
2737 ----------------------
2738
2739 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2740 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2741
2742 2. PCRE was looping on encountering an indefinitely repeated back reference to
2743 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2744 Perl does - treats the match as successful.
2745
2746 ****

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12