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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 159 - (show annotations) (download)
Wed May 2 10:27:18 2007 UTC (6 years, 11 months ago) by ph10
File size: 127245 byte(s)
Add code to configure.ac to remove -g from default CFLAGS and CXXFLAGS.

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

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