ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 176 - (show annotations)
Mon Jun 11 13:48:37 2007 UTC (8 years, 4 months ago) by ph10
File size: 129457 byte(s)
Inserted some (unsigned int) casts to kill compiler warnings.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.2 05-June-07
5 ---------------------
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.
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.
13 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
15 4. Some of the "internals" tests were previously cut out when the link size
16 was not 2, because the output contained actual offsets. The recent new
17 "Z" feature of pcretest means that these can be cut out, making the tests
18 usable with all link sizes.
20 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21 stack recursion. This gives a massive performance boost under BSD, but just
22 a small improvement under Linux. However, it saves one field in the frame
23 in all cases.
25 6. Added more features from the forthcoming Perl 5.10:
27 (a) (?-n) (where n is a string of digits) is a relative subroutine or
28 recursion call. It refers to the nth most recently opened parentheses.
30 (b) (?+n) is also a relative subroutine call; it refers to the nth next
31 to be opened parentheses.
33 (c) Conditions that refer to capturing parentheses can be specified
34 relatively, for example, (?(-2)... or (?(+3)...
36 (d) \K resets the start of the current match so that everything before
37 is not part of it.
39 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
41 (f) \g{name} is another synonym - part of Perl 5.10's unification of
42 reference syntax.
44 (g) (?| introduces a group in which the numbering of parentheses in each
45 alternative starts with the same number.
47 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
50 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
51 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
52 for detecting groups that can match an empty string.
54 9. A pattern with a very large number of alternatives (more than several
55 hundred) was running out of internal workspace during the pre-compile
56 phase, where pcre_compile() figures out how much memory will be needed. A
57 bit of new cunning has reduced the workspace needed for groups with
58 alternatives. The 1000-alternative test pattern now uses 12 bytes of
59 workspace instead of running out of the 4096 that are available.
61 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
64 Version 7.1 24-Apr-07
65 ---------------------
67 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
68 that is more "standard", making use of automake and other Autotools. There
69 is some re-arrangement of the files and adjustment of comments consequent
70 on this.
72 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
73 for recursive directory scanning broke on some systems because the files
74 are not scanned in any specific order and on different systems the order
75 was different. A call to "sort" has been inserted into RunGrepTest for the
76 approprate test as a short-term fix. In the longer term there may be an
77 alternative.
79 3. I had an email from Eric Raymond about problems translating some of PCRE's
80 man pages to HTML (despite the fact that I distribute HTML pages, some
81 people do their own conversions for various reasons). The problems
82 concerned the use of low-level troff macros .br and .in. I have therefore
83 removed all such uses from the man pages (some were redundant, some could
84 be replaced by .nf/.fi pairs). The 132html script that I use to generate
85 HTML has been updated to handle .nf/.fi and to complain if it encounters
86 .br or .in.
88 4. Updated comments in configure.ac that get placed in config.h.in and also
89 arranged for config.h to be included in the distribution, with the name
90 config.h.generic, for the benefit of those who have to compile without
91 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
93 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
94 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
95 makevp.bat for the current PCRE, using the additional files
96 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
98 6. A Windows user reported a minor discrepancy with test 2, which turned out
99 to be caused by a trailing space on an input line that had got lost in his
100 copy. The trailing space was an accident, so I've just removed it.
102 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
103 that is needed.
105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
106 as "const" (a) because they are and (b) because it helps the PHP
107 maintainers who have recently made a script to detect big data structures
108 in the php code that should be moved to the .rodata section. I remembered
109 to update Builducptable as well, so it won't revert if ucptable.h is ever
110 re-created.
112 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
113 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
114 order to be able to cut out the UTF-8 tables in the latter when UTF-8
115 support is not required. This saves 1.5-2K of code, which is important in
116 some applications.
118 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
119 so as not to refer to the tables, even though these functions will never be
120 called when UTF-8 support is disabled. Otherwise there are problems with a
121 shared library.
123 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
125 (a) It was defining its arguments as char * instead of void *.
127 (b) It was assuming that all moves were upwards in memory; this was true
128 a long time ago when I wrote it, but is no longer the case.
130 The emulated memove() is provided for those environments that have neither
131 memmove() nor bcopy(). I didn't think anyone used it these days, but that
132 is clearly not the case, as these two bugs were recently reported.
134 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
135 and Detrail to create the HTML documentation, the .txt form of the man
136 pages, and it removes trailing spaces from listed files. It also creates
137 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
138 case, it wraps all the #defines with #ifndefs. This script should be run
139 before "make dist".
141 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
142 with Unicode property support.
144 (a) For a maximizing quantifier, if the two different cases of the
145 character were of different lengths in their UTF-8 codings (there are
146 some cases like this - I found 11), and the matching function had to
147 back up over a mixture of the two cases, it incorrectly assumed they
148 were both the same length.
150 (b) When PCRE was configured to use the heap rather than the stack for
151 recursion during matching, it was not correctly preserving the data for
152 the other case of a UTF-8 character when checking ahead for a match
153 while processing a minimizing repeat. If the check also involved
154 matching a wide character, but failed, corruption could cause an
155 erroneous result when trying to check for a repeat of the original
156 character.
158 13. Some tidying changes to the testing mechanism:
160 (a) The RunTest script now detects the internal link size and whether there
161 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
162 values substituted by "configure". (The RunGrepTest script already did
163 this for UTF-8.) The configure.ac script no longer substitutes the
164 relevant variables.
166 (b) The debugging options /B and /D in pcretest show the compiled bytecode
167 with length and offset values. This means that the output is different
168 for different internal link sizes. Test 2 is skipped for link sizes
169 other than 2 because of this, bypassing the problem. Unfortunately,
170 there was also a test in test 3 (the locale tests) that used /B and
171 failed for link sizes other than 2. Rather than cut the whole test out,
172 I have added a new /Z option to pcretest that replaces the length and
173 offset values with spaces. This is now used to make test 3 independent
174 of link size. (Test 2 will be tidied up later.)
176 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
177 segmentation fault instead of returning the appropriate error message.
179 15. In multiline mode when the newline sequence was set to "any", the pattern
180 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
181 This doesn't seem right; it now treats the CRLF combination as the line
182 ending, and so does not match in that case. It's only a pattern such as ^$
183 that would hit this one: something like ^ABC$ would have failed after \r
184 and then tried again after \r\n.
186 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
187 in an attempt to make files that differ only in their line terminators
188 compare equal. This works on Linux.
190 17. Under certain error circumstances pcregrep might try to free random memory
191 as it exited. This is now fixed, thanks to valgrind.
193 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
194 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
195 was because its rules for how to advance for /g after matching an empty
196 string at the end of a line did not allow for this case. They now check for
197 it specially.
199 20. pcretest is supposed to handle patterns and data of any length, by
200 extending its buffers when necessary. It was getting this wrong when the
201 buffer for a data line had to be extended.
203 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
204 CRLF as a newline sequence.
206 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
207 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
208 I have nevertheless tidied it up.
210 23. Added some casts to kill warnings from HP-UX ia64 compiler.
212 24. Added a man page for pcre-config.
215 Version 7.0 19-Dec-06
216 ---------------------
218 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
219 moving to gcc 4.1.1.
221 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
222 sys/time.h, which is documented as needed for this function. It doesn't
223 seem to matter on Linux, but it showed up on some releases of OS X.
225 3. It seems that there are systems where bytes whose values are greater than
226 127 match isprint() in the "C" locale. The "C" locale should be the
227 default when a C program starts up. In most systems, only ASCII printing
228 characters match isprint(). This difference caused the output from pcretest
229 to vary, making some of the tests fail. I have changed pcretest so that:
231 (a) When it is outputting text in the compiled version of a pattern, bytes
232 other than 32-126 are always shown as hex escapes.
234 (b) When it is outputting text that is a matched part of a subject string,
235 it does the same, unless a different locale has been set for the match
236 (using the /L modifier). In this case, it uses isprint() to decide.
238 4. Fixed a major bug that caused incorrect computation of the amount of memory
239 required for a compiled pattern when options that changed within the
240 pattern affected the logic of the preliminary scan that determines the
241 length. The relevant options are -x, and -i in UTF-8 mode. The result was
242 that the computed length was too small. The symptoms of this bug were
243 either the PCRE error "internal error: code overflow" from pcre_compile(),
244 or a glibc crash with a message such as "pcretest: free(): invalid next
245 size (fast)". Examples of patterns that provoked this bug (shown in
246 pcretest format) are:
248 /(?-x: )/x
249 /(?x)(?-x: \s*#\s*)/
250 /((?i)[\x{c0}])/8
251 /(?i:[\x{c0}])/8
253 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
254 is now done differently.
256 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
257 wrapper classes; (b) implement a new function in the C++ scanner that is
258 more efficient than the old way of doing things because it avoids levels of
259 recursion in the regex matching; (c) add a paragraph to the documentation
260 for the FullMatch() function.
262 6. The escape sequence \n was being treated as whatever was defined as
263 "newline". Not only was this contrary to the documentation, which states
264 that \n is character 10 (hex 0A), but it also went horribly wrong when
265 "newline" was defined as CRLF. This has been fixed.
267 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
268 was being set to -1 for the "end of line" case (supposedly a value that no
269 character can have). Though this value is never used (the check for end of
270 line is "zero bytes in current character"), it caused compiler complaints.
271 I've changed it to 0xffffffff.
273 8. In pcre_version.c, the version string was being built by a sequence of
274 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
275 string (as it is for production releases) called a macro with an empty
276 argument. The C standard says the result of this is undefined. The gcc
277 compiler treats it as an empty string (which was what was wanted) but it is
278 reported that Visual C gives an error. The source has been hacked around to
279 avoid this problem.
281 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
282 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
283 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
284 of them did).
286 10. Originally, pcretest opened its input and output without "b"; then I was
287 told that "b" was needed in some environments, so it was added for release
288 5.0 to both the input and output. (It makes no difference on Unix-like
289 systems.) Later I was told that it is wrong for the input on Windows. I've
290 now abstracted the modes into two macros, to make it easier to fiddle with
291 them, and removed "b" from the input mode under Windows.
293 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
295 12. Added -help and --help to pcretest as an official way of being reminded
296 of the options.
298 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
299 and pcrecpp.cc because they annoy compilers at high warning levels.
301 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
303 15. Fixed an occurrence of == in configure.ac that should have been = (shell
304 scripts are not C programs :-) and which was not noticed because it works
305 on Linux.
307 16. pcretest is supposed to handle any length of pattern and data line (as one
308 line or as a continued sequence of lines) by extending its input buffer if
309 necessary. This feature was broken for very long pattern lines, leading to
310 a string of junk being passed to pcre_compile() if the pattern was longer
311 than about 50K.
313 17. I have done a major re-factoring of the way pcre_compile() computes the
314 amount of memory needed for a compiled pattern. Previously, there was code
315 that made a preliminary scan of the pattern in order to do this. That was
316 OK when PCRE was new, but as the facilities have expanded, it has become
317 harder and harder to keep it in step with the real compile phase, and there
318 have been a number of bugs (see for example, 4 above). I have now found a
319 cunning way of running the real compile function in a "fake" mode that
320 enables it to compute how much memory it would need, while actually only
321 ever using a few hundred bytes of working memory and without too many
322 tests of the mode. This should make future maintenance and development
323 easier. A side effect of this work is that the limit of 200 on the nesting
324 depth of parentheses has been removed (though this was never a serious
325 limitation, I suspect). However, there is a downside: pcre_compile() now
326 runs more slowly than before (30% or more, depending on the pattern). I
327 hope this isn't a big issue. There is no effect on runtime performance.
329 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
330 newline (only possible for the last line of a file) and it was a
331 pattern that set a locale (followed by /Lsomething), pcretest crashed.
333 19. Added additional timing features to pcretest. (1) The -tm option now times
334 matching only, not compiling. (2) Both -t and -tm can be followed, as a
335 separate command line item, by a number that specifies the number of
336 repeats to use when timing. The default is 50000; this gives better
337 precision, but takes uncomfortably long for very large patterns.
339 20. Extended pcre_study() to be more clever in cases where a branch of a
340 subpattern has no definite first character. For example, (a*|b*)[cd] would
341 previously give no result from pcre_study(). Now it recognizes that the
342 first character must be a, b, c, or d.
344 21. There was an incorrect error "recursive call could loop indefinitely" if
345 a subpattern (or the entire pattern) that was being tested for matching an
346 empty string contained only one non-empty item after a nested subpattern.
347 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
348 incorrectly, because the \d was being skipped in the check.
350 22. The pcretest program now has a new pattern option /B and a command line
351 option -b, which is equivalent to adding /B to every pattern. This causes
352 it to show the compiled bytecode, without the additional information that
353 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
354 is the same as /B/I).
356 23. A new optimization is now able automatically to treat some sequences such
357 as a*b as a*+b. More specifically, if something simple (such as a character
358 or a simple class like \d) has an unlimited quantifier, and is followed by
359 something that cannot possibly match the quantified thing, the quantifier
360 is automatically "possessified".
362 24. A recursive reference to a subpattern whose number was greater than 39
363 went wrong under certain circumstances in UTF-8 mode. This bug could also
364 have affected the operation of pcre_study().
366 25. Realized that a little bit of performance could be had by replacing
367 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
369 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
371 27. Possessive quantifiers such as a++ were previously implemented by turning
372 them into atomic groups such as ($>a+). Now they have their own opcodes,
373 which improves performance. This includes the automatically created ones
374 from 23 above.
376 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
377 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
378 the first matched character to be a colon. This applied both to named and
379 numbered groups.
381 29. The ucpinternal.h header file was missing its idempotency #ifdef.
383 30. I was sent a "project" file called libpcre.a.dev which I understand makes
384 building PCRE on Windows easier, so I have included it in the distribution.
386 31. There is now a check in pcretest against a ridiculously large number being
387 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
388 loop, the loop is abandoned.
390 32. Forward references to subpatterns in conditions such as (?(2)...) where
391 subpattern 2 is defined later cause pcre_compile() to search forwards in
392 the pattern for the relevant set of parentheses. This search went wrong
393 when there were unescaped parentheses in a character class, parentheses
394 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
396 33. "Subroutine" calls and backreferences were previously restricted to
397 referencing subpatterns earlier in the regex. This restriction has now
398 been removed.
400 34. Added a number of extra features that are going to be in Perl 5.10. On the
401 whole, these are just syntactic alternatives for features that PCRE had
402 previously implemented using the Python syntax or my own invention. The
403 other formats are all retained for compatibility.
405 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
406 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
407 also .NET compatible.
409 (b) A recursion or subroutine call to a named group can now be defined as
410 (?&name) as well as (?P>name).
412 (c) A backreference to a named group can now be defined as \k<name> or
413 \k'name' as well as (?P=name). The new forms, as well as being in Perl
414 5.10, are also .NET compatible.
416 (d) A conditional reference to a named group can now use the syntax
417 (?(<name>) or (?('name') as well as (?(name).
419 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
420 groups (named and numbered) that are never evaluated inline, but can be
421 called as "subroutines" from elsewhere. In effect, the DEFINE condition
422 is always false. There may be only one alternative in such a group.
424 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
425 as the simple (?(R). The condition is true only if the most recent
426 recursion is that of the given number or name. It does not search out
427 through the entire recursion stack.
429 (g) The escape \gN or \g{N} has been added, where N is a positive or
430 negative number, specifying an absolute or relative reference.
432 35. Tidied to get rid of some further signed/unsigned compiler warnings and
433 some "unreachable code" warnings.
435 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
436 things, this adds five new scripts.
438 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
439 There were also incompatibilities regarding the handling of \Q..\E inside
440 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
441 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
443 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
444 matches an empty string, and forcibly breaks the loop. There were bugs in
445 this code in non-simple cases. For a pattern such as ^(a()*)* matched
446 against aaaa the result was just "a" rather than "aaaa", for example. Two
447 separate and independent bugs (that affected different cases) have been
448 fixed.
450 39. Refactored the code to abolish the use of different opcodes for small
451 capturing bracket numbers. This is a tidy that I avoided doing when I
452 removed the limit on the number of capturing brackets for 3.5 back in 2001.
453 The new approach is not only tidier, it makes it possible to reduce the
454 memory needed to fix the previous bug (38).
456 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
457 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
458 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
459 mode.
461 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
462 report.
464 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
465 copy construction and assignment in the C++ wrapper.
467 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
468 couple of bugs that could have given wrong results in the "--newline=crlf"
469 case.
471 44. Added a number of casts and did some reorganization of signed/unsigned int
472 variables following suggestions from Dair Grant. Also renamed the variable
473 "this" as "item" because it is a C++ keyword.
475 45. Arranged for dftables to add
477 #include "pcre_internal.h"
479 to pcre_chartables.c because without it, gcc 4.x may remove the array
480 definition from the final binary if PCRE is built into a static library and
481 dead code stripping is activated.
483 46. For an unanchored pattern, if a match attempt fails at the start of a
484 newline sequence, and the newline setting is CRLF or ANY, and the next two
485 characters are CRLF, advance by two characters instead of one.
488 Version 6.7 04-Jul-06
489 ---------------------
491 1. In order to handle tests when input lines are enormously long, pcretest has
492 been re-factored so that it automatically extends its buffers when
493 necessary. The code is crude, but this _is_ just a test program. The
494 default size has been increased from 32K to 50K.
496 2. The code in pcre_study() was using the value of the re argument before
497 testing it for NULL. (Of course, in any sensible call of the function, it
498 won't be NULL.)
500 3. The memmove() emulation function in pcre_internal.h, which is used on
501 systems that lack both memmove() and bcopy() - that is, hardly ever -
502 was missing a "static" storage class specifier.
504 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
505 containing an extended class (one that cannot be represented by a bitmap
506 because it contains high-valued characters or Unicode property items, e.g.
507 [\pZ]). Almost always one would set UTF-8 mode when processing such a
508 pattern, but PCRE should not loop if you do not (it no longer does).
509 [Detail: two cases were found: (a) a repeated subpattern containing an
510 extended class; (b) a recursive reference to a subpattern that followed a
511 previous extended class. It wasn't skipping over the extended class
512 correctly when UTF-8 mode was not set.]
514 5. A negated single-character class was not being recognized as fixed-length
515 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
516 compile error "lookbehind assertion is not fixed length".
518 6. The RunPerlTest auxiliary script was showing an unexpected difference
519 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
520 write a Perl script that can interpret lines of an input file either as
521 byte characters or as UTF-8, which is what "perltest" was being required to
522 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
523 can't do is switch easily at run time between having the "use utf8;" pragma
524 or not. In the end, I fudged it by using the RunPerlTest script to insert
525 "use utf8;" explicitly for the UTF-8 tests.
527 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
528 the end of the subject string, contrary to the documentation and to what
529 Perl does. This was true of both matching functions. Now it matches only at
530 the start of the subject and immediately after *internal* newlines.
532 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
533 a pointer to an int instead of a pointer to an unsigned long int. This
534 caused problems on 64-bit systems.
536 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
537 instance of the 'standard' template library not being so standard".
539 10. There was no check on the number of named subpatterns nor the maximum
540 length of a subpattern name. The product of these values is used to compute
541 the size of the memory block for a compiled pattern. By supplying a very
542 long subpattern name and a large number of named subpatterns, the size
543 computation could be caused to overflow. This is now prevented by limiting
544 the length of names to 32 characters, and the number of named subpatterns
545 to 10,000.
547 11. Subpatterns that are repeated with specific counts have to be replicated in
548 the compiled pattern. The size of memory for this was computed from the
549 length of the subpattern and the repeat count. The latter is limited to
550 65535, but there was no limit on the former, meaning that integer overflow
551 could in principle occur. The compiled length of a repeated subpattern is
552 now limited to 30,000 bytes in order to prevent this.
554 12. Added the optional facility to have named substrings with the same name.
556 13. Added the ability to use a named substring as a condition, using the
557 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
558 are numbers (not recommended). Forward references are permitted.
560 14. Added forward references in named backreferences (if you see what I mean).
562 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
563 pattern could run off the end of the subject. For example, the pattern
564 "(?s)(.{1,5})"8 did this with the subject "ab".
566 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
567 PCRE_CASELESS was set when matching characters that were quantified with ?
568 or *.
570 17. A character class other than a single negated character that had a minimum
571 but no maximum quantifier - for example [ab]{6,} - was not handled
572 correctly by pce_dfa_exec(). It would match only one character.
574 18. A valid (though odd) pattern that looked like a POSIX character
575 class but used an invalid character after [ (for example [[,abc,]]) caused
576 pcre_compile() to give the error "Failed: internal error: code overflow" or
577 in some cases to crash with a glibc free() error. This could even happen if
578 the pattern terminated after [[ but there just happened to be a sequence of
579 letters, a binary zero, and a closing ] in the memory that followed.
581 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
582 over the years. Originally (before any Unicode support), just the bottom 8
583 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
584 output from "man perlunicode" includes this:
586 The regular expression compiler produces polymorphic opcodes. That
587 is, the pattern adapts to the data and automatically switches to
588 the Unicode character scheme when presented with Unicode data--or
589 instead uses a traditional byte scheme when presented with byte
590 data.
592 Sadly, a wide octal escape does not cause a switch, and in a string with
593 no other multibyte characters, these octal escapes are treated as before.
594 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
595 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
596 Unicode string.
598 I have not perpetrated such confusion in PCRE. Up till now, it took just
599 the bottom 8 bits, as in old Perl. I have now made octal escapes with
600 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
601 translate to the appropriate multibyte character.
603 29. Applied some refactoring to reduce the number of warnings from Microsoft
604 and Borland compilers. This has included removing the fudge introduced
605 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
606 a warning about an unused variable.
608 21. PCRE has not included VT (character 0x0b) in the set of whitespace
609 characters since release 4.0, because Perl (from release 5.004) does not.
610 [Or at least, is documented not to: some releases seem to be in conflict
611 with the documentation.] However, when a pattern was studied with
612 pcre_study() and all its branches started with \s, PCRE still included VT
613 as a possible starting character. Of course, this did no harm; it just
614 caused an unnecessary match attempt.
616 22. Removed a now-redundant internal flag bit that recorded the fact that case
617 dependency changed within the pattern. This was once needed for "required
618 byte" processing, but is no longer used. This recovers a now-scarce options
619 bit. Also moved the least significant internal flag bit to the most-
620 significant bit of the word, which was not previously used (hangover from
621 the days when it was an int rather than a uint) to free up another bit for
622 the future.
624 23. Added support for CRLF line endings as well as CR and LF. As well as the
625 default being selectable at build time, it can now be changed at runtime
626 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
627 specify that it is scanning data with non-default line endings.
629 24. Changed the definition of CXXLINK to make it agree with the definition of
630 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
632 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
633 recursions. This makes a big different to stack usage for some patterns.
635 26. If a subpattern containing a named recursion or subroutine reference such
636 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
637 the space required for the compiled pattern went wrong and gave too small a
638 value. Depending on the environment, this could lead to "Failed: internal
639 error: code overflow at offset 49" or "glibc detected double free or
640 corruption" errors.
642 27. Applied patches from Google (a) to support the new newline modes and (b) to
643 advance over multibyte UTF-8 characters in GlobalReplace.
645 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
646 difference for some implementation of PCRE in some Windows version.
648 29. Added some extra testing facilities to pcretest:
650 \q<number> in a data line sets the "match limit" value
651 \Q<number> in a data line sets the "match recursion limt" value
652 -S <number> sets the stack size, where <number> is in megabytes
654 The -S option isn't available for Windows.
657 Version 6.6 06-Feb-06
658 ---------------------
660 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
661 in pcreposix.h. I have copied the definition from pcre.h.
663 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
664 because pcre.h is no longer a built file.
666 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
667 not normally included in the compiled code.
670 Version 6.5 01-Feb-06
671 ---------------------
673 1. When using the partial match feature with pcre_dfa_exec(), it was not
674 anchoring the second and subsequent partial matches at the new starting
675 point. This could lead to incorrect results. For example, with the pattern
676 /1234/, partially matching against "123" and then "a4" gave a match.
678 2. Changes to pcregrep:
680 (a) All non-match returns from pcre_exec() were being treated as failures
681 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
682 error message is output. Some extra information is given for the
684 probably the only errors that are likely to be caused by users (by
685 specifying a regex that has nested indefinite repeats, for instance).
686 If there are more than 20 of these errors, pcregrep is abandoned.
688 (b) A binary zero was treated as data while matching, but terminated the
689 output line if it was written out. This has been fixed: binary zeroes
690 are now no different to any other data bytes.
692 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
693 used to set a locale for matching. The --locale=xxxx long option has
694 been added (no short equivalent) to specify a locale explicitly on the
695 pcregrep command, overriding the environment variables.
697 (d) When -B was used with -n, some line numbers in the output were one less
698 than they should have been.
700 (e) Added the -o (--only-matching) option.
702 (f) If -A or -C was used with -c (count only), some lines of context were
703 accidentally printed for the final match.
705 (g) Added the -H (--with-filename) option.
707 (h) The combination of options -rh failed to suppress file names for files
708 that were found from directory arguments.
710 (i) Added the -D (--devices) and -d (--directories) options.
712 (j) Added the -F (--fixed-strings) option.
714 (k) Allow "-" to be used as a file name for -f as well as for a data file.
716 (l) Added the --colo(u)r option.
718 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
719 is not present by default.
721 3. A nasty bug was discovered in the handling of recursive patterns, that is,
722 items such as (?R) or (?1), when the recursion could match a number of
723 alternatives. If it matched one of the alternatives, but subsequently,
724 outside the recursion, there was a failure, the code tried to back up into
725 the recursion. However, because of the way PCRE is implemented, this is not
726 possible, and the result was an incorrect result from the match.
728 In order to prevent this happening, the specification of recursion has
729 been changed so that all such subpatterns are automatically treated as
730 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
732 4. I had overlooked the fact that, in some locales, there are characters for
733 which isalpha() is true but neither isupper() nor islower() are true. In
734 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
735 and ordfeminine) are like this. This affected the treatment of \w and \W
736 when they appeared in character classes, but not when they appeared outside
737 a character class. The bit map for "word" characters is now created
738 separately from the results of isalnum() instead of just taking it from the
739 upper, lower, and digit maps. (Plus the underscore character, of course.)
741 5. The above bug also affected the handling of POSIX character classes such as
742 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
743 permanent tables. Instead, the bit maps for such a class were previously
744 created as the appropriate unions of the upper, lower, and digit bitmaps.
745 Now they are created by subtraction from the [[:word:]] class, which has
746 its own bitmap.
748 6. The [[:blank:]] character class matches horizontal, but not vertical space.
749 It is created by subtracting the vertical space characters (\x09, \x0a,
750 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
751 subtraction was done in the overall bitmap for a character class, meaning
752 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
753 be recognized. This bug has been fixed.
755 7. Patches from the folks at Google:
757 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
758 real life, but is still worth protecting against".
760 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
761 regular expressions".
763 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
764 have it.
766 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
767 "configure" and the latter not, in order to fix a problem somebody had
768 with compiling the Arg class on HP-UX.
770 (e) Improve the error-handling of the C++ wrapper a little bit.
772 (f) New tests for checking recursion limiting.
774 8. The pcre_memmove() function, which is used only if the environment does not
775 have a standard memmove() function (and is therefore rarely compiled),
776 contained two bugs: (a) use of int instead of size_t, and (b) it was not
777 returning a result (though PCRE never actually uses the result).
779 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
780 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
781 returned instead of calling malloc() with an overflowing number that would
782 most likely cause subsequent chaos.
784 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
786 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
787 with this option is matched, the nmatch and pmatch options of regexec() are
788 ignored.
790 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
791 provided in case anyone wants to the the POSIX interface with UTF-8
792 strings.
794 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
795 C++ linking (needed for some HP-UX environments).
797 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
798 (unused parameter) and in the pcre_printint() function (omitted "default"
799 switch label when the default is to do nothing).
801 15. Added some code to make it possible, when PCRE is compiled as a C++
802 library, to replace subject pointers for pcre_exec() with a smart pointer
803 class, thus making it possible to process discontinuous strings.
805 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
806 much the same function. They were added by different people who were trying
807 to make PCRE easy to compile on non-Unix systems. It has been suggested
808 that PCRE_EXPORT be abolished now that there is more automatic apparatus
809 for compiling on Windows systems. I have therefore replaced it with
810 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
811 defaults to "extern" for C or "extern C" for C++, which works fine on
812 Unix-like systems. It is now possible to override the value of PCRE_DATA_
813 SCOPE with something explicit in config.h. In addition:
815 (a) pcreposix.h still had just "extern" instead of either of these macros;
816 I have replaced it with PCRE_DATA_SCOPE.
818 (b) Functions such as _pcre_xclass(), which are internal to the library,
819 but external in the C sense, all had PCRE_EXPORT in their definitions.
820 This is apparently wrong for the Windows case, so I have removed it.
821 (It makes no difference on Unix-like systems.)
823 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
824 of recursive calls to match(). This is different to MATCH_LIMIT because
825 that limits the total number of calls to match(), not all of which increase
826 the depth of recursion. Limiting the recursion depth limits the amount of
827 stack (or heap if NO_RECURSE is set) that is used. The default can be set
828 when PCRE is compiled, and changed at run time. A patch from Google adds
829 this functionality to the C++ interface.
831 18. Changes to the handling of Unicode character properties:
833 (a) Updated the table to Unicode 4.1.0.
835 (b) Recognize characters that are not in the table as "Cn" (undefined).
837 (c) I revised the way the table is implemented to a much improved format
838 which includes recognition of ranges. It now supports the ranges that
839 are defined in UnicodeData.txt, and it also amalgamates other
840 characters into ranges. This has reduced the number of entries in the
841 table from around 16,000 to around 3,000, thus reducing its size
842 considerably. I realized I did not need to use a tree structure after
843 all - a binary chop search is just as efficient. Having reduced the
844 number of entries, I extended their size from 6 bytes to 8 bytes to
845 allow for more data.
847 (d) Added support for Unicode script names via properties such as \p{Han}.
849 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
850 matching that character.
852 20. When matching a repeated Unicode property with a minimum greater than zero,
853 (for example \pL{2,}), PCRE could look past the end of the subject if it
854 reached it while seeking the minimum number of characters. This could
855 happen only if some of the characters were more than one byte long, because
856 there is a check for at least the minimum number of bytes.
858 21. Refactored the implementation of \p and \P so as to be more general, to
859 allow for more different types of property in future. This has changed the
860 compiled form incompatibly. Anybody with saved compiled patterns that use
861 \p or \P will have to recompile them.
863 22. Added "Any" and "L&" to the supported property types.
865 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
866 but give a compile time error if the value is greater than 0xff.
868 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
869 accidentally not being installed or uninstalled.
871 25. The pcre.h file was built from pcre.h.in, but the only changes that were
872 made were to insert the current release number. This seemed silly, because
873 it made things harder for people building PCRE on systems that don't run
874 "configure". I have turned pcre.h into a distributed file, no longer built
875 by "configure", with the version identification directly included. There is
876 no longer a pcre.h.in file.
878 However, this change necessitated a change to the pcre-config script as
879 well. It is built from pcre-config.in, and one of the substitutions was the
880 release number. I have updated configure.ac so that ./configure now finds
881 the release number by grepping pcre.h.
883 26. Added the ability to run the tests under valgrind.
886 Version 6.4 05-Sep-05
887 ---------------------
889 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
890 "--" to be printed when multiple files were scanned, even when none of the
891 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
892 consider it to be a bug, and have restored the previous behaviour.
894 2. A couple of code tidies to get rid of compiler warnings.
896 3. The pcretest program used to cheat by referring to symbols in the library
897 whose names begin with _pcre_. These are internal symbols that are not
898 really supposed to be visible externally, and in some environments it is
899 possible to suppress them. The cheating is now confined to including
900 certain files from the library's source, which is a bit cleaner.
902 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
903 file's purpose clearer.
905 5. Reorganized pcre_ucp_findchar().
908 Version 6.3 15-Aug-05
909 ---------------------
911 1. The file libpcre.pc.in did not have general read permission in the tarball.
913 2. There were some problems when building without C++ support:
915 (a) If C++ support was not built, "make install" and "make test" still
916 tried to test it.
918 (b) There were problems when the value of CXX was explicitly set. Some
919 changes have been made to try to fix these, and ...
921 (c) --disable-cpp can now be used to explicitly disable C++ support.
923 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
924 backslash in a target when C++ was disabled. This confuses some
925 versions of "make", apparently. Using an intermediate variable solves
926 this. (Same for CPP_LOBJ.)
928 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
929 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
930 necessary on certain architectures.
932 4. Added a setting of -export-symbols-regex to the link command to remove
933 those symbols that are exported in the C sense, but actually are local
934 within the library, and not documented. Their names all begin with
935 "_pcre_". This is not a perfect job, because (a) we have to except some
936 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
937 available (and never for static libraries). I have made a note to try to
938 find a way round (a) in the future.
941 Version 6.2 01-Aug-05
942 ---------------------
944 1. There was no test for integer overflow of quantifier values. A construction
945 such as {1111111111111111} would give undefined results. What is worse, if
946 a minimum quantifier for a parenthesized subpattern overflowed and became
947 negative, the calculation of the memory size went wrong. This could have
948 led to memory overwriting.
950 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
952 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
953 operating environments where this matters.
955 4. Applied Giuseppe Maxia's patch to add additional features for controlling
956 PCRE options from within the C++ wrapper.
958 5. Named capturing subpatterns were not being correctly counted when a pattern
959 was compiled. This caused two problems: (a) If there were more than 100
960 such subpatterns, the calculation of the memory needed for the whole
961 compiled pattern went wrong, leading to an overflow error. (b) Numerical
962 back references of the form \12, where the number was greater than 9, were
963 not recognized as back references, even though there were sufficient
964 previous subpatterns.
966 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
967 versions of gcc, e.g. 2.95.4.
970 Version 6.1 21-Jun-05
971 ---------------------
973 1. There was one reference to the variable "posix" in pcretest.c that was not
974 surrounded by "#if !defined NOPOSIX".
976 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
977 the cross-check on the old pcre_info() function, for the benefit of the
978 cut-down version of PCRE that is currently imported into Exim.
980 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
981 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
982 patterns that start (?i) but I don't think that matters. The use of (?i) is
983 just an example; this all applies to the other options as well.
985 4. Since libtool seems to echo the compile commands it is issuing, the output
986 from "make" can be reduced a bit by putting "@" in front of each libtool
987 compile command.
989 5. Patch from the folks at Google for configure.in to be a bit more thorough
990 in checking for a suitable C++ installation before trying to compile the
991 C++ stuff. This should fix a reported problem when a compiler was present,
992 but no suitable headers.
994 6. The man pages all had just "PCRE" as their title. I have changed them to
995 be the relevant file name. I have also arranged that these names are
996 retained in the file doc/pcre.txt, which is a concatenation in text format
997 of all the man pages except the little individual ones for each function.
999 7. The NON-UNIX-USE file had not been updated for the different set of source
1000 files that come with release 6. I also added a few comments about the C++
1001 wrapper.
1004 Version 6.0 07-Jun-05
1005 ---------------------
1007 1. Some minor internal re-organization to help with my DFA experiments.
1009 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1010 didn't matter for the library itself when fully configured, but did matter
1011 when compiling without UCP support, or within Exim, where the ucp files are
1012 not imported.
1014 3. Refactoring of the library code to split up the various functions into
1015 different source modules. The addition of the new DFA matching code (see
1016 below) to a single monolithic source would have made it really too
1017 unwieldy, quite apart from causing all the code to be include in a
1018 statically linked application, when only some functions are used. This is
1019 relevant even without the DFA addition now that patterns can be compiled in
1020 one application and matched in another.
1022 The downside of splitting up is that there have to be some external
1023 functions and data tables that are used internally in different modules of
1024 the library but which are not part of the API. These have all had their
1025 names changed to start with "_pcre_" so that they are unlikely to clash
1026 with other external names.
1028 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1029 a different (DFA) algorithm. Although it is slower than the original
1030 function, it does have some advantages for certain types of matching
1031 problem.
1033 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1034 including restarting after a partial match.
1036 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1037 defined when compiling for Windows was sent to me. I have put it into the
1038 code, though I have no means of testing or verifying it.
1040 7. Added the pcre_refcount() auxiliary function.
1042 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1043 match before or at the first newline in the subject string. In pcretest,
1044 the /f option on a pattern can be used to set this.
1046 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1047 would behave wrongly. This has been present in PCRE since release 4.0.
1049 10. A number of changes to the pcregrep command:
1051 (a) Refactored how -x works; insert ^(...)$ instead of setting
1052 PCRE_ANCHORED and checking the length, in preparation for adding
1053 something similar for -w.
1055 (b) Added the -w (match as a word) option.
1057 (c) Refactored the way lines are read and buffered so as to have more
1058 than one at a time available.
1060 (d) Implemented a pcregrep test script.
1062 (e) Added the -M (multiline match) option. This allows patterns to match
1063 over several lines of the subject. The buffering ensures that at least
1064 8K, or the rest of the document (whichever is the shorter) is available
1065 for matching (and similarly the previous 8K for lookbehind assertions).
1067 (f) Changed the --help output so that it now says
1069 -w, --word-regex(p)
1071 instead of two lines, one with "regex" and the other with "regexp"
1072 because that confused at least one person since the short forms are the
1073 same. (This required a bit of code, as the output is generated
1074 automatically from a table. It wasn't just a text change.)
1076 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1077 option but starts with a hyphen. Could be a pattern or a path name
1078 starting with a hyphen, for instance.
1080 (h) "-" can be given as a file name to represent stdin.
1082 (i) When file names are being printed, "(standard input)" is used for
1083 the standard input, for compatibility with GNU grep. Previously
1084 "<stdin>" was used.
1086 (j) The option --label=xxx can be used to supply a name to be used for
1087 stdin when file names are being printed. There is no short form.
1089 (k) Re-factored the options decoding logic because we are going to add
1090 two more options that take data. Such options can now be given in four
1091 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1093 (l) Added the -A, -B, and -C options for requesting that lines of context
1094 around matches be printed.
1096 (m) Added the -L option to print the names of files that do not contain
1097 any matching lines, that is, the complement of -l.
1099 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1100 continue to scan other files.
1102 (o) The -s option was incorrectly implemented. For compatibility with other
1103 greps, it now suppresses the error message for a non-existent or non-
1104 accessible file (but not the return code). There is a new option called
1105 -q that suppresses the output of matching lines, which was what -s was
1106 previously doing.
1108 (p) Added --include and --exclude options to specify files for inclusion
1109 and exclusion when recursing.
1111 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1112 Hopefully, it now does.
1114 12. Missing cast in pcre_study().
1116 13. Added an "uninstall" target to the makefile.
1118 14. Replaced "extern" in the function prototypes in Makefile.in with
1119 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1120 world, but is set differently for Windows.
1122 15. Added a second compiling function called pcre_compile2(). The only
1123 difference is that it has an extra argument, which is a pointer to an
1124 integer error code. When there is a compile-time failure, this is set
1125 non-zero, in addition to the error test pointer being set to point to an
1126 error message. The new argument may be NULL if no error number is required
1127 (but then you may as well call pcre_compile(), which is now just a
1128 wrapper). This facility is provided because some applications need a
1129 numeric error indication, but it has also enabled me to tidy up the way
1130 compile-time errors are handled in the POSIX wrapper.
1132 16. Added VPATH=.libs to the makefile; this should help when building with one
1133 prefix path and installing with another. (Or so I'm told by someone who
1134 knows more about this stuff than I do.)
1136 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1137 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1138 match everything, including newlines. This is not POSIX-compatible, but
1139 somebody wanted the feature. From pcretest it can be activated by using
1140 both the P and the s flags.
1142 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1144 19. libpcre.pc was being incorrectly installed as executable.
1146 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1147 it now also looks for '\r' so that it will work unmodified on Windows.
1149 21. Added Google's contributed C++ wrapper to the distribution.
1151 22. Added some untidy missing memory free() calls in pcretest, to keep
1152 Electric Fence happy when testing.
1156 Version 5.0 13-Sep-04
1157 ---------------------
1159 1. Internal change: literal characters are no longer packed up into items
1160 containing multiple characters in a single byte-string. Each character
1161 is now matched using a separate opcode. However, there may be more than one
1162 byte in the character in UTF-8 mode.
1164 2. The pcre_callout_block structure has two new fields: pattern_position and
1165 next_item_length. These contain the offset in the pattern to the next match
1166 item, and its length, respectively.
1168 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1169 insertion of callouts before each pattern item. Added the /C option to
1170 pcretest to make use of this.
1172 4. On the advice of a Windows user, the lines
1174 #if defined(_WIN32) || defined(WIN32)
1175 _setmode( _fileno( stdout ), 0x8000 );
1176 #endif /* defined(_WIN32) || defined(WIN32) */
1178 have been added to the source of pcretest. This apparently does useful
1179 magic in relation to line terminators.
1181 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1182 for the benefit of those environments where the "b" makes a difference.
1184 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1185 to know about it. I have put a hack into configure.in that adds in code
1186 to set GCC=yes if CC=icc. This seems to end up at a point in the
1187 generated configure script that is early enough to affect the setting of
1188 compiler options, which is what is needed, but I have no means of testing
1189 whether it really works. (The user who reported this had patched the
1190 generated configure script, which of course I cannot do.)
1192 LATER: After change 22 below (new libtool files), the configure script
1193 seems to know about icc (and also ecc). Therefore, I have commented out
1194 this hack in configure.in.
1196 7. Added support for pkg-config (2 patches were sent in).
1198 8. Negated POSIX character classes that used a combination of internal tables
1199 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1200 [[:^ascii]]. Typically, they would match almost any characters. The other
1201 POSIX classes were not broken in this way.
1203 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1204 to find the match, as PCRE was deluded into thinking that the match had to
1205 start at the start point or following a newline. The same bug applied to
1206 patterns with negative forward assertions or any backward assertions
1207 preceding ".*" at the start, unless the pattern required a fixed first
1208 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1210 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1211 starting at the last subject character, bytes beyond the end of the subject
1212 string were read.
1214 11. Renamed the variable "class" as "classbits" to make life easier for C++
1215 users. (Previously there was a macro definition, but it apparently wasn't
1216 enough.)
1218 12. Added the new field "tables" to the extra data so that tables can be passed
1219 in at exec time, or the internal tables can be re-selected. This allows
1220 a compiled regex to be saved and re-used at a later time by a different
1221 program that might have everything at different addresses.
1223 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1224 -R library as well as a -L library.
1226 14. The debugging options of pcretest (-d on the command line or D on a
1227 pattern) showed incorrect output for anything following an extended class
1228 that contained multibyte characters and which was followed by a quantifier.
1230 15. Added optional support for general category Unicode character properties
1231 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1232 support. It adds about 90K to the size of the library. The meanings of the
1233 inbuilt class escapes such as \d and \s have NOT been changed.
1235 16. Updated pcredemo.c to include calls to free() to release the memory for the
1236 compiled pattern.
1238 17. The generated file chartables.c was being created in the source directory
1239 instead of in the building directory. This caused the build to fail if the
1240 source directory was different from the building directory, and was
1241 read-only.
1243 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1244 file. No doubt somebody will tell me if they don't make sense... Also added
1245 Dan Mooney's comments about building on OpenVMS.
1247 19. Added support for partial matching via the PCRE_PARTIAL option for
1248 pcre_exec() and the \P data escape in pcretest.
1250 20. Extended pcretest with 3 new pattern features:
1252 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1253 write the compiled pattern to the file whose name is "rest-of-line".
1254 This is a straight binary dump of the data, with the saved pointer to
1255 the character tables forced to be NULL. The study data, if any, is
1256 written too. After writing, pcretest reads a new pattern.
1258 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1259 compiled pattern from the given file. There must not be any
1260 occurrences of "<" in the file name (pretty unlikely); if there are,
1261 pcretest will instead treat the initial "<" as a pattern delimiter.
1262 After reading in the pattern, pcretest goes on to read data lines as
1263 usual.
1265 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1266 and 16-bit fields in a compiled pattern, to simulate a pattern that
1267 was compiled on a host of opposite endianness.
1269 21. The pcre-exec() function can now cope with patterns that were compiled on
1270 hosts of opposite endianness, with this restriction:
1272 As for any compiled expression that is saved and used later, the tables
1273 pointer field cannot be preserved; the extra_data field in the arguments
1274 to pcre_exec() should be used to pass in a tables address if a value
1275 other than the default internal tables were used at compile time.
1277 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1278 now diagnosed as an error. Previously, most of the time, a negative number
1279 would have been treated as zero, but if in addition "ovector" was passed as
1280 NULL, a crash could occur.
1282 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1283 new versions from the libtool 1.5 distribution (the last one is a copy of
1284 a file called libtool.m4). This seems to have fixed the need to patch
1285 "configure" to support Darwin 1.3 (which I used to do). However, I still
1286 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1287 workstation).
1289 24. Changed the PCRE licence to be the more standard "BSD" licence.
1292 Version 4.5 01-Dec-03
1293 ---------------------
1295 1. There has been some re-arrangement of the code for the match() function so
1296 that it can be compiled in a version that does not call itself recursively.
1297 Instead, it keeps those local variables that need separate instances for
1298 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1299 needs to "recurse". Keeping track of where control must go is done by means
1300 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1301 hide most of the details from the main code, and operates only if
1302 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1303 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1304 operating.
1306 To make it easier for callers to provide specially tailored get/free
1307 functions for this usage, two new functions, pcre_stack_malloc, and
1308 pcre_stack_free, are used. They are always called in strict stacking order,
1309 and the size of block requested is always the same.
1311 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1312 PCRE has been compiled to use the stack or the heap for recursion. The
1313 -C option of pcretest uses this to show which version is compiled.
1315 A new data escape \S, is added to pcretest; it causes the amounts of store
1316 obtained and freed by both kinds of malloc/free at match time to be added
1317 to the output.
1319 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1320 what's available on my current Linux desktop machine.
1322 3. When matching a UTF-8 string, the test for a valid string at the start has
1323 been extended. If start_offset is not zero, PCRE now checks that it points
1324 to a byte that is the start of a UTF-8 character. If not, it returns
1325 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1326 this is necessary because there may be backward assertions in the pattern.
1327 When matching the same subject several times, it may save resources to use
1328 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1330 4. The code for checking the validity of UTF-8 strings has been tightened so
1331 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1332 containing "overlong sequences".
1334 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1335 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1336 should have been "&", but it just so happened that all the cases this let
1337 through by mistake were picked up later in the function.
1339 6. I had used a variable called "isblank" - this is a C99 function, causing
1340 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1342 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1343 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1344 so the version is shown; (c) stop "make test" echoing "./RunTest".
1346 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1348 9. The prototype for memmove() for systems that don't have it was using
1349 size_t, but the inclusion of the header that defines size_t was later. I've
1350 moved the #includes for the C headers earlier to avoid this.
1352 10. Added some adjustments to the code to make it easier to compiler on certain
1353 special systems:
1355 (a) Some "const" qualifiers were missing.
1356 (b) Added the macro EXPORT before all exported functions; by default this
1357 is defined to be empty.
1358 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1359 that it reads its output file name as an argument instead of writing
1360 to the standard output and assuming this can be redirected.
1362 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1363 class containing characters with values greater than 255, PCRE compilation
1364 went into a loop.
1366 12. A recursive reference to a subpattern that was within another subpattern
1367 that had a minimum quantifier of zero caused PCRE to crash. For example,
1368 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1369 recursion. If the recursively-called subpattern itself had a zero repeat,
1370 that was OK.
1372 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1373 buffer into which it was copied (for escape processing) was still set at
1374 1024, so long lines caused crashes.
1376 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1377 "internal error: code overflow...". This applied to any character class
1378 that was followed by a possessive quantifier.
1380 15. Modified the Makefile to add libpcre.la as a prerequisite for
1381 libpcreposix.la because I was told this is needed for a parallel build to
1382 work.
1384 16. If a pattern that contained .* following optional items at the start was
1385 studied, the wrong optimizing data was generated, leading to matching
1386 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1387 matching string must start with a or b or c. The correct conclusion for
1388 this pattern is that a match can start with any character.
1391 Version 4.4 13-Aug-03
1392 ---------------------
1394 1. In UTF-8 mode, a character class containing characters with values between
1395 127 and 255 was not handled correctly if the compiled pattern was studied.
1396 In fixing this, I have also improved the studying algorithm for such
1397 classes (slightly).
1399 2. Three internal functions had redundant arguments passed to them. Removal
1400 might give a very teeny performance improvement.
1402 3. Documentation bug: the value of the capture_top field in a callout is *one
1403 more than* the number of the hightest numbered captured substring.
1405 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1406 in incorrectly linking with a previously installed version. They now link
1407 explicitly with libpcre.la.
1409 5. configure.in no longer needs to recognize Cygwin specially.
1411 6. A problem in pcre.in for Windows platforms is fixed.
1413 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1414 pcretest, it used to include the size of the study block as part of its
1415 output. Unfortunately, the structure contains a field that has a different
1416 size on different hardware architectures. This meant that the tests that
1417 showed this size failed. As the block is currently always of a fixed size,
1418 this information isn't actually particularly useful in pcretest output, so
1419 I have just removed it.
1421 8. Three pre-processor statements accidentally did not start in column 1.
1422 Sadly, there are *still* compilers around that complain, even though
1423 standard C has not required this for well over a decade. Sigh.
1425 9. In pcretest, the code for checking callouts passed small integers in the
1426 callout_data field, which is a void * field. However, some picky compilers
1427 complained about the casts involved for this on 64-bit systems. Now
1428 pcretest passes the address of the small integer instead, which should get
1429 rid of the warnings.
1431 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1432 both compile and run time, and gives an error if an invalid UTF-8 sequence
1433 is found. There is a option for disabling this check in cases where the
1434 string is known to be correct and/or the maximum performance is wanted.
1436 11. In response to a bug report, I changed one line in Makefile.in from
1438 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1439 to
1440 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1442 to look similar to other lines, but I have no way of telling whether this
1443 is the right thing to do, as I do not use Windows. No doubt I'll get told
1444 if it's wrong...
1447 Version 4.3 21-May-03
1448 ---------------------
1450 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1451 Makefile.
1453 2. Some refactoring to improve the quality of the code:
1455 (i) The utf8_table... variables are now declared "const".
1457 (ii) The code for \cx, which used the "case flipping" table to upper case
1458 lower case letters, now just substracts 32. This is ASCII-specific,
1459 but the whole concept of \cx is ASCII-specific, so it seems
1460 reasonable.
1462 (iii) PCRE was using its character types table to recognize decimal and
1463 hexadecimal digits in the pattern. This is silly, because it handles
1464 only 0-9, a-f, and A-F, but the character types table is locale-
1465 specific, which means strange things might happen. A private
1466 table is now used for this - though it costs 256 bytes, a table is
1467 much faster than multiple explicit tests. Of course, the standard
1468 character types table is still used for matching digits in subject
1469 strings against \d.
1471 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1472 ending in _t are). So I've renamed it as ESC_tee.
1474 3. The first argument for regexec() in the POSIX wrapper should have been
1475 defined as "const".
1477 4. Changed pcretest to use malloc() for its buffers so that they can be
1478 Electric Fenced for debugging.
1480 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1481 to read one or more bytes before the start of the subject string. Often this
1482 had no effect on PCRE's behaviour, but in some circumstances it could
1483 provoke a segmentation fault.
1485 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1486 to try to read one or more bytes before the start of the subject string.
1488 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1489 UTF-8 support could misbehave in various ways if the subject string
1490 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1491 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1492 back over UTF-8 characters.)
1495 Version 4.2 14-Apr-03
1496 ---------------------
1498 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1500 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1501 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1502 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1503 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1504 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1506 Note: automatic setting of the BUILD variables is not yet working
1507 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1508 compile-time but not at link-time
1509 [LINK]: use for linking executables only
1510 make different versions for Windows and non-Windows
1511 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1512 libraries
1513 [LINK_FOR_BUILD]: new variable
1514 [OBJEXT]: use throughout
1515 [EXEEXT]: use throughout
1516 <winshared>: new target
1517 <wininstall>: new target
1518 <dftables.o>: use native compiler
1519 <dftables>: use native linker
1520 <install>: handle Windows platform correctly
1521 <clean>: ditto
1522 <check>: ditto
1523 copy DLL to top builddir before testing
1525 As part of these changes, -no-undefined was removed again. This was reported
1526 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1527 in any case.
1529 3. Some tidies to get rid of compiler warnings:
1531 . In the match_data structure, match_limit was an unsigned long int, whereas
1532 match_call_count was an int. I've made them both unsigned long ints.
1534 . In pcretest the fact that a const uschar * doesn't automatically cast to
1535 a void * provoked a warning.
1537 . Turning on some more compiler warnings threw up some "shadow" variables
1538 and a few more missing casts.
1540 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1541 option, a class that contained a single character with a value between 128
1542 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1544 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1545 option, a class that contained several characters, but with at least one
1546 whose value was between 128 and 255 caused PCRE to crash.
1549 Version 4.1 12-Mar-03
1550 ---------------------
1552 1. Compiling with gcc -pedantic found a couple of places where casts were
1553 needed, and a string in dftables.c that was longer than standard compilers are
1554 required to support.
1556 2. Compiling with Sun's compiler found a few more places where the code could
1557 be tidied up in order to avoid warnings.
1559 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1560 first of these names is deprecated in the latest Autoconf in favour of the name
1561 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1562 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1563 analogy I have changed it to CFLAGS_FOR_BUILD.
1565 4. Added -no-undefined to the linking command in the Makefile, because this is
1566 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1567 linking step for the pcreposix library.
1569 5. PCRE was failing to diagnose the case of two named groups with the same
1570 name.
1572 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1573 literal character that is needed in the subject for a match, and scans along to
1574 ensure that it is present before embarking on the full matching process. This
1575 saves time in cases of nested unlimited repeats that are never going to match.
1576 Problem: the scan can take a lot of time if the subject is very long (e.g.
1577 megabytes), thus penalizing straightforward matches. It is now done only if the
1578 amount of subject to be scanned is less than 1000 bytes.
1580 7. A lesser problem with the same optimization is that it was recording the
1581 first character of an anchored pattern as "needed", thus provoking a search
1582 right along the subject, even when the first match of the pattern was going to
1583 fail. The "needed" character is now not set for anchored patterns, unless it
1584 follows something in the pattern that is of non-fixed length. Thus, it still
1585 fulfils its original purpose of finding quick non-matches in cases of nested
1586 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1589 Version 4.0 17-Feb-03
1590 ---------------------
1592 1. If a comment in an extended regex that started immediately after a meta-item
1593 extended to the end of string, PCRE compiled incorrect data. This could lead to
1594 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1596 2. Moved to autoconf 2.53 and libtool 1.4.2.
1598 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1599 the special perltest8 script is no longer needed - all the tests can be run
1600 from a single perltest script.
1602 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1603 by \s. It has now been removed in PCRE. This means it isn't recognized as
1604 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1605 class [:space:] *does* include VT, thereby creating a mess.
1607 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1608 space and tab.
1610 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1611 its new features into the main test script, reducing the number of scripts.
1613 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1614 were backward compatible, and made the (?i) apply to the whole pattern, as if
1615 /i were given. Now it behaves more logically, and applies the option setting
1616 only to what follows. PCRE has been changed to follow suit. However, if it
1617 finds options settings right at the start of the pattern, it extracts them into
1618 the global options, as before. Thus, they show up in the info data.
1620 8. Added support for the \Q...\E escape sequence. Characters in between are
1621 treated as literals. This is slightly different from Perl in that $ and @ are
1622 also handled as literals inside the quotes. In Perl, they will cause variable
1623 interpolation. Note the following examples:
1625 Pattern PCRE matches Perl matches
1627 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1628 \Qabc\$xyz\E abc\$xyz abc\$xyz
1629 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1631 For compatibility with Perl, \Q...\E sequences are recognized inside character
1632 classes as well as outside them.
1634 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1635 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1636 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1637 signed/unsigned warnings.
1639 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1640 option for pcretest, so I've replaced it by a simple function that does just
1641 that job.
1643 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1644 "pcregrep -".
1646 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1647 Java package. This provides some syntactic sugar for simple cases of what my
1648 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1649 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1650 item, you can use this simplified notation. Note that only makes sense with
1651 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1652 greediness, whatever the setting of the PCRE_UNGREEDY option.
1654 13. A change of greediness default within a pattern was not taking effect at
1655 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1656 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1657 was abstracted outside.
1659 14. PCRE now supports the \G assertion. It is true when the current matching
1660 position is at the start point of the match. This differs from \A when the
1661 starting offset is non-zero. Used with the /g option of pcretest (or similar
1662 code), it works in the same way as it does for Perl's /g option. If all
1663 alternatives of a regex begin with \G, the expression is anchored to the start
1664 match position, and the "anchored" flag is set in the compiled expression.
1666 15. Some bugs concerning the handling of certain option changes within patterns
1667 have been fixed. These applied to options other than (?ims). For example,
1668 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1669 been the other way round. Some of this was related to change 7 above.
1671 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1672 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1673 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1674 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1676 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1677 mode. Unlike ".", it always matches newline, whatever the setting of
1678 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1679 assertions. Perl allows it, but it doesn't (in general) work because it can't
1680 calculate the length of the lookbehind. At least, that's the case for Perl
1681 5.8.0 - I've been told they are going to document that it doesn't work in
1682 future.
1684 18. Added an error diagnosis for escapes that PCRE does not support: these are
1685 \L, \l, \N, \P, \p, \U, \u, and \X.
1687 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1688 reading past the end of the pattern in cases such as /[abcd/.
1690 20. PCRE was getting more memory than necessary for patterns with classes that
1691 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1693 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1694 compiling PCRE for use with Virtual Pascal.
1696 22. Small fix to the Makefile to make it work properly if the build is done
1697 outside the source tree.
1699 23. Added a new extension: a condition to go with recursion. If a conditional
1700 subpattern starts with (?(R) the "true" branch is used if recursion has
1701 happened, whereas the "false" branch is used only at the top level.
1703 24. When there was a very long string of literal characters (over 255 bytes
1704 without UTF support, over 250 bytes with UTF support), the computation of how
1705 much memory was required could be incorrect, leading to segfaults or other
1706 strange effects.
1708 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1709 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1710 there was a subsequent back reference to those brackets. This meant that, for
1711 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1712 possible to check for precisely this case. All we can do is abandon the
1713 optimization if .* occurs inside capturing brackets when there are any back
1714 references whatsoever. (See below for a better fix that came later.)
1716 26. The handling of the optimization for finding the first character of a
1717 non-anchored pattern, and for finding a character that is required later in the
1718 match were failing in some cases. This didn't break the matching; it just
1719 failed to optimize when it could. The way this is done has been re-implemented.
1721 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1723 28. Added a new feature that provides some of the functionality that Perl
1724 provides with (?{...}). The facility is termed a "callout". The way it is done
1725 in PCRE is for the caller to provide an optional function, by setting
1726 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1727 global variable. By default it is unset, which disables all calling out. To get
1728 the function called, the regex must include (?C) at appropriate points. This
1729 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1730 This provides a means of identifying different callout points. When PCRE
1731 reaches such a point in the regex, if pcre_callout has been set, the external
1732 function is called. It is provided with data in a structure called
1733 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1734 matching continues; if it returns a non-zero value, the match at the current
1735 point fails. However, backtracking will occur if possible. [This was changed
1736 later and other features added - see item 49 below.]
1738 29. pcretest is upgraded to test the callout functionality. It provides a
1739 callout function that displays information. By default, it shows the start of
1740 the match and the current position in the text. There are some new data escapes
1741 to vary what happens:
1743 \C+ in addition, show current contents of captured substrings
1744 \C- do not supply a callout function
1745 \C!n return 1 when callout number n is reached
1746 \C!n!m return 1 when callout number n is reached for the mth time
1748 30. If pcregrep was called with the -l option and just a single file name, it
1749 output "<stdin>" if a match was found, instead of the file name.
1751 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1752 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1753 pcre_exec(). This saves a malloc/free per call. The default value of
1754 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1755 when configuring.
1757 32. The default maximum size of a compiled pattern is 64K. There have been a
1758 few cases of people hitting this limit. The code now uses macros to handle the
1759 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1760 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1761 configuring. Tests 2 and 5 work only with 2-byte links because they output
1762 debugging information about compiled patterns.
1764 33. Internal code re-arrangements:
1766 (a) Moved the debugging function for printing out a compiled regex into
1767 its own source file (printint.c) and used #include to pull it into
1768 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1769 separate copies.
1771 (b) Defined the list of op-code names for debugging as a macro in
1772 internal.h so that it is next to the definition of the opcodes.
1774 (c) Defined a table of op-code lengths for simpler skipping along compiled
1775 code. This is again a macro in internal.h so that it is next to the
1776 definition of the opcodes.
1778 34. Added support for recursive calls to individual subpatterns, along the
1779 lines of Robin Houston's patch (but implemented somewhat differently).
1781 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1782 allow it to read and process whole directories in Win32. This code was
1783 contributed by Lionel Fourquaux; it has not been tested by me.
1785 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1786 used to name a group. Names consist of alphanumerics and underscores, and must
1787 be unique. Back references use the syntax (?P=name) and recursive calls use
1788 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1789 numbers. The function pcre_fullinfo() can be used after compilation to extract
1790 a name/number map. There are three relevant calls:
1792 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1793 PCRE_INFO_NAMECOUNT yields the number of entries
1794 PCRE_INFO_NAMETABLE yields a pointer to the map.
1796 The map is a vector of fixed-size entries. The size of each entry depends on
1797 the length of the longest name used. The first two bytes of each entry are the
1798 group number, most significant byte first. There follows the corresponding
1799 name, zero terminated. The names are in alphabetical order.
1801 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1802 case instead of 255. Making it the same both with and without UTF-8 support
1803 means that the same test output works with both.
1805 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1806 calling malloc() with a zero argument.
1808 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1809 optimization. I've improved things by keeping a bitmap of backreferences with
1810 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1811 fact referenced, the optimization can be applied. It is unlikely that a
1812 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1813 the match to follow \n) will appear inside brackets with a number greater than
1814 31, but if it does, any back reference > 31 suppresses the optimization.
1816 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1817 of disabling numbered capturing parentheses. Any opening parenthesis that is
1818 not followed by ? behaves as if it were followed by ?: but named parentheses
1819 can still be used for capturing (and they will acquire numbers in the usual
1820 way).
1822 41. Redesigned the return codes from the match() function into yes/no/error so
1823 that errors can be passed back from deep inside the nested calls. A malloc
1824 failure while inside a recursive subpattern call now causes the
1825 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1827 42. It is now possible to set a limit on the number of times the match()
1828 function is called in a call to pcre_exec(). This facility makes it possible to
1829 limit the amount of recursion and backtracking, though not in a directly
1830 obvious way, because the match() function is used in a number of different
1831 circumstances. The count starts from zero for each position in the subject
1832 string (for non-anchored patterns). The default limit is, for compatibility, a
1833 large number, namely 10 000 000. You can change this in two ways:
1835 (a) When configuring PCRE before making, you can use --with-match-limit=n
1836 to set a default value for the compiled library.
1838 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1839 a different value is set. See 45 below.
1841 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1843 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1844 of things that can be changed at compile time. The first argument specifies
1845 what is wanted and the second points to where the information is to be placed.
1846 The current list of available information is:
1850 The output is an integer that is set to one if UTF-8 support is available;
1851 otherwise it is set to zero.
1855 The output is an integer that it set to the value of the code that is used for
1856 newline. It is either LF (10) or CR (13).
1860 The output is an integer that contains the number of bytes used for internal
1861 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1865 The output is an integer that contains the threshold above which the POSIX
1866 interface uses malloc() for output vectors. See item 31 above.
1870 The output is an unsigned integer that contains the default limit of the number
1871 of match() calls in a pcre_exec() execution. See 42 above.
1873 44. pcretest has been upgraded by the addition of the -C option. This causes it
1874 to extract all the available output from the new pcre_config() function, and to
1875 output it. The program then exits immediately.
1877 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1878 order to support additional features. One way would have been to define
1879 pcre_exec2() (for example) with extra arguments, but this would not have been
1880 extensible, and would also have required all calls to the original function to
1881 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1882 is used for passing in "extra" data from pcre_study().
1884 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1885 contains the following fields:
1887 flags a bitmap indicating which of the following fields are set
1888 study_data opaque data from pcre_study()
1889 match_limit a way of specifying a limit on match() calls for a specific
1890 call to pcre_exec()
1891 callout_data data for callouts (see 49 below)
1893 The flag bits are also defined in pcre.h, and are
1899 The pcre_study() function now returns one of these new pcre_extra blocks, with
1900 the actual study data pointed to by the study_data field, and the
1901 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1902 before. That is, this change is entirely upwards-compatible and requires no
1903 change to existing code.
1905 If you want to pass in additional data to pcre_exec(), you can either place it
1906 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1907 block.
1909 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1910 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1911 times with different match limits, until it finds the minimum value needed for
1912 pcre_exec() to complete. The value is then output. This can be instructive; for
1913 most simple matches the number is quite small, but for pathological cases it
1914 gets very large very quickly.
1916 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1917 returns the size of the data block pointed to by the study_data field in a
1918 pcre_extra block, that is, the value that was passed as the argument to
1919 pcre_malloc() when PCRE was getting memory in which to place the information
1920 created by pcre_study(). The fourth argument should point to a size_t variable.
1921 pcretest has been extended so that this information is shown after a successful
1922 pcre_study() call when information about the compiled regex is being displayed.
1924 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1925 because what follows is always an absolute path. (Later: it turns out that this
1926 is more than cosmetic for MinGW, because it doesn't like empty path
1927 components.)
1929 49. Some changes have been made to the callout feature (see 28 above):
1931 (i) A callout function now has three choices for what it returns:
1933 0 => success, carry on matching
1934 > 0 => failure at this point, but backtrack if possible
1935 < 0 => serious error, return this value from pcre_exec()
1937 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1938 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1939 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1940 use by callout functions. It will never be used by PCRE itself.
1942 (ii) The pcre_extra structure (see 45 above) has a void * field called
1943 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1944 pcre_callout_block structure has a field of the same name. The contents of
1945 the field passed in the pcre_extra structure are passed to the callout
1946 function in the corresponding field in the callout block. This makes it
1947 easier to use the same callout-containing regex from multiple threads. For
1948 testing, the pcretest program has a new data escape
1950 \C*n pass the number n (may be negative) as callout_data
1952 If the callout function in pcretest receives a non-zero value as
1953 callout_data, it returns that value.
1955 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1956 there were some redundant $(CFLAGS) in commands that are now specified as
1957 $(LINK), which already includes $(CFLAGS).
1959 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1960 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1961 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1962 one-byte characters throughout. Note that case-insensitive matching applies
1963 only to characters whose values are less than 256. PCRE doesn't support the
1964 notion of cases for higher-valued characters.
1966 (i) A character class whose characters are all within 0-255 is handled as
1967 a bit map, and the map is inverted for negative classes. Previously, a
1968 character > 255 always failed to match such a class; however it should
1969 match if the class was a negative one (e.g. [^ab]). This has been fixed.
1971 (ii) A negated character class with a single character < 255 is coded as
1972 "not this character" (OP_NOT). This wasn't working properly when the test
1973 character was multibyte, either singly or repeated.
1975 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1976 mode, for example: \x{100}{2,3}.
1978 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1979 singly or repeated) now correctly test multibyte characters. However,
1980 PCRE doesn't recognize any characters with values greater than 255 as
1981 digits, spaces, or word characters. Such characters always match \D, \S,
1982 and \W, and never match \d, \s, or \w.
1984 (v) Classes may now contain characters and character ranges with values
1985 greater than 255. For example: [ab\x{100}-\x{400}].
1987 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1988 PCRE in UTF-8 mode.
1990 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1991 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1992 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1993 value.)
1995 53. The single man page has become too large. I have therefore split it up into
1996 a number of separate man pages. These also give rise to individual HTML pages;
1997 these are now put in a separate directory, and there is an index.html page that
1998 lists them all. Some hyperlinking between the pages has been installed.
2000 54. Added convenience functions for handling named capturing parentheses.
2002 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2003 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2004 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2005 are faulted.
2007 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2008 calling configure. These values are used when compiling the dftables.c program
2009 which is run to generate the source of the default character tables. They
2010 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2011 you will need to set these values.
2013 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2016 Version 3.9 02-Jan-02
2017 ---------------------
2019 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2021 2. If --disable-static was given, the building process failed when trying to
2022 build pcretest and pcregrep. (For some reason it was using libtool to compile
2023 them, which is not right, as they aren't part of the library.)
2026 Version 3.8 18-Dec-01
2027 ---------------------
2029 1. The experimental UTF-8 code was completely screwed up. It was packing the
2030 bytes in the wrong order. How dumb can you get?
2033 Version 3.7 29-Oct-01
2034 ---------------------
2036 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2037 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2038 this didn't happen under Solaris 8, where I normally test things.
2040 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2041 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2042 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2043 this fix an hour or so after the initial 3.7 release.)
2046 Version 3.6 23-Oct-01
2047 ---------------------
2049 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2050 offsets passed as NULL with zero offset count.
2052 2. The config.guess and config.sub files had not been updated when I moved to
2053 the latest autoconf.
2056 Version 3.5 15-Aug-01
2057 ---------------------
2059 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2060 had been forgotten.
2062 2. By using declared but undefined structures, we can avoid using "void"
2063 definitions in pcre.h while keeping the internal definitions of the structures
2064 private.
2066 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2067 user point of view, this means that both static and shared libraries are built
2068 by default, but this can be individually controlled. More of the work of
2069 handling this static/shared cases is now inside libtool instead of PCRE's make
2070 file.
2072 4. The pcretest utility is now installed along with pcregrep because it is
2073 useful for users (to test regexs) and by doing this, it automatically gets
2074 relinked by libtool. The documentation has been turned into a man page, so
2075 there are now .1, .txt, and .html versions in /doc.
2077 5. Upgrades to pcregrep:
2078 (i) Added long-form option names like gnu grep.
2079 (ii) Added --help to list all options with an explanatory phrase.
2080 (iii) Added -r, --recursive to recurse into sub-directories.
2081 (iv) Added -f, --file to read patterns from a file.
2083 6. pcre_exec() was referring to its "code" argument before testing that
2084 argument for NULL (and giving an error if it was NULL).
2086 7. Upgraded Makefile.in to allow for compiling in a different directory from
2087 the source directory.
2089 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2090 options bits, the pointer it was passed was to an int instead of to an unsigned
2091 long int. This mattered only on 64-bit systems.
2093 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2094 generated) instead of pcre.in, which it its source. Also made the same change
2095 in several of the .c files.
2097 10. A new release of gcc defines printf() as a macro, which broke pcretest
2098 because it had an ifdef in the middle of a string argument for printf(). Fixed
2099 by using separate calls to printf().
2101 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2102 script, to force use of CR or LF instead of \n in the source. On non-Unix
2103 systems, the value can be set in config.h.
2105 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2106 absolute limit. Changed the text of the error message to make this clear, and
2107 likewise updated the man page.
2109 13. The limit of 99 on the number of capturing subpatterns has been removed.
2110 The new limit is 65535, which I hope will not be a "real" limit.
2113 Version 3.4 22-Aug-00
2114 ---------------------
2116 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2118 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2121 Version 3.3 01-Aug-00
2122 ---------------------
2124 1. If an octal character was given, but the value was greater than \377, it
2125 was not getting masked to the least significant bits, as documented. This could
2126 lead to crashes in some systems.
2128 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2129 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2131 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2132 These just pass their arguments on to (pcre_free)(), but they are provided
2133 because some uses of PCRE bind it to non-C systems that can call its functions,
2134 but cannot call free() or pcre_free() directly.
2136 4. Add "make test" as a synonym for "make check". Corrected some comments in
2137 the Makefile.
2139 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2140 Makefile.
2142 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2143 command called pgrep for grepping around the active processes.
2145 7. Added the beginnings of support for UTF-8 character strings.
2147 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2148 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2149 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2150 out for the ar command.)
2153 Version 3.2 12-May-00
2154 ---------------------
2156 This is purely a bug fixing release.
2158 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2159 of ZA. This was just one example of several cases that could provoke this bug,
2160 which was introduced by change 9 of version 2.00. The code for breaking
2161 infinite loops after an iteration that matches an empty string was't working
2162 correctly.
2164 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2165 when matched against abbab (for example). After matching an empty string, it
2166 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2167 caused it to match further down the string than it should.
2169 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2170 was there because it doesn't seem to be needed, and it causes trouble on some
2171 systems, as it is not a Standard C header. It has been removed.
2173 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2174 were reported on the Macintosh. The changes were from
2176 while ((c = *(++ptr)) != 0 && c != '\n');
2177 to
2178 while ((c = *(++ptr)) != 0 && c != '\n') ;
2180 Totally extraordinary, but if that's what it takes...
2182 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2183 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2184 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2185 assumes the way PCRE uses memmove() (always moving upwards).
2187 6. PCRE is being used in one environment where strchr() is not available. There
2188 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2189 faster code anyway.
2192 Version 3.1 09-Feb-00
2193 ---------------------
2195 The only change in this release is the fixing of some bugs in Makefile.in for
2196 the "install" target:
2198 (1) It was failing to install pcreposix.h.
2200 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2203 Version 3.0 01-Feb-00
2204 ---------------------
2206 1. Add support for the /+ modifier to perltest (to output $` like it does in
2207 pcretest).
2209 2. Add support for the /g modifier to perltest.
2211 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2212 matches null strings.
2214 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2215 pattern. Perl treats empty patterns specially - it reuses the most recent
2216 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2217 effect.
2219 5. The POSIX interface was broken in that it was just handing over the POSIX
2220 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2221 required a bigger vector, with some working space on the end. This means that
2222 the POSIX wrapper now has to get and free some memory, and copy the results.
2224 6. Added some simple autoconf support, placing the test data and the
2225 documentation in separate directories, re-organizing some of the
2226 information files, and making it build pcre-config (a GNU standard). Also added
2227 libtool support for building PCRE as a shared library, which is now the
2228 default.
2230 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2231 09 are not valid octal constants. Single digits will be used for minor values
2232 less than 10.
2234 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2235 existing programs that set these in the POSIX interface can use PCRE without
2236 modification.
2238 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2239 return all that pcre_info() returns, plus additional data. The pcre_info()
2240 function is retained for compatibility, but is considered to be obsolete.
2242 10. Added experimental recursion feature (?R) to handle one common case that
2243 Perl 5.6 will be able to do with (?p{...}).
2245 11. Added support for POSIX character classes like [:alpha:], which Perl is
2246 adopting.
2249 Version 2.08 31-Aug-99
2250 ----------------------
2252 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2253 trying to match at the startoffset position, but instead was moving forward to
2254 the next newline as if a previous match had failed.
2256 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2257 and could get into a loop if a null string was matched other than at the start
2258 of the subject.
2260 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2261 be distinguished at compile time, and for completeness also added PCRE_DATE.
2263 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2264 in GnuWin32 environments.
2267 Version 2.07 29-Jul-99
2268 ----------------------
2270 1. The documentation is now supplied in plain text form and HTML as well as in
2271 the form of man page sources.
2273 2. C++ compilers don't like assigning (void *) values to other pointer types.
2274 In particular this affects malloc(). Although there is no problem in Standard
2275 C, I've put in casts to keep C++ compilers happy.
2277 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2278 should be (const char *).
2280 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2281 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2282 However, I haven't made this a standard facility. The documentation doesn't
2283 mention it, and the Makefile doesn't support it.
2285 5. The Makefile now contains an "install" target, with editable destinations at
2286 the top of the file. The pcretest program is not installed.
2288 6. pgrep -V now gives the PCRE version number and date.
2290 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2291 causing the entire string to be ignored, instead of just the last character.
2293 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2294 non-matching string, it can take a very, very long time, even for strings of
2295 quite modest length, because of the nested recursion. PCRE now does better in
2296 some of these cases. It does this by remembering the last required literal
2297 character in the pattern, and pre-searching the subject to ensure it is present
2298 before running the real match. In other words, it applies a heuristic to detect
2299 some types of certain failure quickly, and in the above example, if presented
2300 with a string that has no trailing " it gives "no match" very quickly.
2302 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2303 other alternatives are tried instead.
2306 Version 2.06 09-Jun-99
2307 ----------------------
2309 1. Change pcretest's output for amount of store used to show just the code
2310 space, because the remainder (the data block) varies in size between 32-bit and
2311 64-bit systems.
2313 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2314 start matching at. This allows lookbehinds to work when searching for multiple
2315 occurrences in a string.
2317 3. Added additional options to pcretest for testing multiple occurrences:
2319 /+ outputs the rest of the string that follows a match
2320 /g loops for multiple occurrences, using the new startoffset argument
2321 /G loops for multiple occurrences by passing an incremented pointer
2323 4. PCRE wasn't doing the "first character" optimization for patterns starting
2324 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2325 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2326 the letter 'x'. On long subject strings, this gives a significant speed-up.
2329 Version 2.05 21-Apr-99
2330 ----------------------
2332 1. Changed the type of magic_number from int to long int so that it works
2333 properly on 16-bit systems.
2335 2. Fixed a bug which caused patterns starting with .* not to work correctly
2336 when the subject string contained newline characters. PCRE was assuming
2337 anchoring for such patterns in all cases, which is not correct because .* will
2338 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2339 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2340 must be retried after every newline in the subject.
2343 Version 2.04 18-Feb-99
2344 ----------------------
2346 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2347 computation of the store needed to hold the pattern was incorrect (too large).
2348 If such patterns were nested a few deep, this could multiply and become a real
2349 problem.
2351 2. Added /M option to pcretest to show the memory requirement of a specific
2352 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2354 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2355 compiled in such a way that the backtracking after subsequent failure was
2356 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2357 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2360 Version 2.03 02-Feb-99
2361 ----------------------
2363 1. Fixed typo and small mistake in man page.
2365 2. Added 4th condition (GPL supersedes if conflict) and created separate
2366 LICENCE file containing the conditions.
2368 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2369 Perl, that is the internal \ allows the delimiter to be included in the
2370 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2371 the final delimiter, add \ to the end of the pattern (to test the error).
2373 4. Added the convenience functions for extracting substrings after a successful
2374 match. Updated pcretest to make it able to test these functions.
2377 Version 2.02 14-Jan-99
2378 ----------------------
2380 1. Initialized the working variables associated with each extraction so that
2381 their saving and restoring doesn't refer to uninitialized store.
2383 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2384 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2385 fix the problem.
2387 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2388 calls, and wasn't printing the correct value for compiling calls. Increased the
2389 default value of LOOPREPEAT, and the number of significant figures in the
2390 times.
2392 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2394 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2395 a building problem on Windows NT with a FAT file system.
2398 Version 2.01 21-Oct-98
2399 ----------------------
2401 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2402 to character tables built by pcre_maketables() in the current locale. If NULL
2403 is passed, the default tables are used.
2406 Version 2.00 24-Sep-98
2407 ----------------------
2409 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2410 it any more.
2412 2. Allow quantification of (?>) groups, and make it work correctly.
2414 3. The first character computation wasn't working for (?>) groups.
2416 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2417 end of the subject) and add 5.005's \z, which really does match only at the
2418 very end of the subject.
2420 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2422 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2423 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2424 localized options. All options to pcre_study() were also removed.
2426 7. Add other new features from 5.005:
2428 $(?<= positive lookbehind
2429 $(?<! negative lookbehind
2430 (?imsx-imsx) added the unsetting capability
2431 such a setting is global if at outer level; local otherwise
2432 (?imsx-imsx:) non-capturing groups with option setting
2433 (?(cond)re|re) conditional pattern matching
2435 A backreference to itself in a repeated group matches the previous
2436 captured string.
2438 8. General tidying up of studying (both automatic and via "study")
2439 consequential on the addition of new assertions.
2441 9. As in 5.005, unlimited repeated groups that could match an empty substring
2442 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2443 runtime if any iteration does actually match an empty substring.
2445 10. Include the RunTest script in the distribution.
2447 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2448 discrepancies, some of which were old and were also with respect to 5.004. They
2449 have now been fixed.
2452 Version 1.09 28-Apr-98
2453 ----------------------
2455 1. A negated single character class followed by a quantifier with a minimum
2456 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2457 program crashes, or just wrong answers. This did not apply to negated classes
2458 containing more than one character, or to minima other than one.
2461 Version 1.08 27-Mar-98
2462 ----------------------
2464 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2466 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2467 latter must appear before anything that relies on it in the pattern.
2470 Version 1.07 16-Feb-98
2471 ----------------------
2473 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2474 repeat of a potentially empty string).
2477 Version 1.06 23-Jan-98
2478 ----------------------
2480 1. Added Markus Oberhumer's little patches for C++.
2482 2. Literal strings longer than 255 characters were broken.
2485 Version 1.05 23-Dec-97
2486 ----------------------
2488 1. Negated character classes containing more than one character were failing if
2489 PCRE_CASELESS was set at run time.
2492 Version 1.04 19-Dec-97
2493 ----------------------
2495 1. Corrected the man page, where some "const" qualifiers had been omitted.
2497 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2498 input syntax.
2500 3. Fixed memory leak which occurred when a regex with back references was
2501 matched with an offsets vector that wasn't big enough. The temporary memory
2502 that is used in this case wasn't being freed if the match failed.
2504 4. Tidied pcretest to ensure it frees memory that it gets.
2506 5. Temporary memory was being obtained in the case where the passed offsets
2507 vector was exactly big enough.
2509 6. Corrected definition of offsetof() from change 5 below.
2511 7. I had screwed up change 6 below and broken the rules for the use of
2512 setjmp(). Now fixed.
2515 Version 1.03 18-Dec-97
2516 ----------------------
2518 1. A erroneous regex with a missing opening parenthesis was correctly
2519 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2520 on some systems.
2522 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2523 it was reported that one broken compiler failed on the former because "code" is
2524 also an independent variable.
2526 3. The erroneous regex a[]b caused an array overrun reference.
2528 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2529 fail on data ending with that character. (It was going on too far, and checking
2530 the next character, typically a binary zero.) This was specific to the
2531 optimized code for single-character negative classes.
2533 5. Added a contributed patch from the TIN world which does the following:
2535 + Add an undef for memmove, in case the the system defines a macro for it.
2537 + Add a definition of offsetof(), in case there isn't one. (I don't know
2538 the reason behind this - offsetof() is part of the ANSI standard - but
2539 it does no harm).
2541 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2542 most of the places where whitespace preceded '#'. I have given up and
2543 allowed the remaining 2 cases to be at the margin.
2545 + Rename some variables in pcre to eliminate shadowing. This seems very
2546 pedantic, but does no harm, of course.
2548 6. Moved the call to setjmp() into its own function, to get rid of warnings
2549 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2551 7. Constructs such as \d{8,} were compiling into the equivalent of
2552 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2553 outcome, but in this particular case used more store than had been allocated,
2554 which caused the bug to be discovered because it threw up an internal error.
2556 8. The debugging code in both pcre and pcretest for outputting the compiled
2557 form of a regex was going wrong in the case of back references followed by
2558 curly-bracketed repeats.
2561 Version 1.02 12-Dec-97
2562 ----------------------
2564 1. Typos in pcre.3 and comments in the source fixed.
2566 2. Applied a contributed patch to get rid of places where it used to remove
2567 'const' from variables, and fixed some signed/unsigned and uninitialized
2568 variable warnings.
2570 3. Added the "runtest" target to Makefile.
2572 4. Set default compiler flag to -O2 rather than just -O.
2575 Version 1.01 19-Nov-97
2576 ----------------------
2578 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2579 like /([ab]*)*/, that is, for classes with more than one character in them.
2581 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2582 as /((?>a*))*/ (a PCRE_EXTRA facility).
2585 Version 1.00 18-Nov-97
2586 ----------------------
2588 1. Added compile-time macros to support systems such as SunOS4 which don't have
2589 memmove() or strerror() but have other things that can be used instead.
2591 2. Arranged that "make clean" removes the executables.
2594 Version 0.99 27-Oct-97
2595 ----------------------
2597 1. Fixed bug in code for optimizing classes with only one character. It was
2598 initializing a 32-byte map regardless, which could cause it to run off the end
2599 of the memory it had got.
2601 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2604 Version 0.98 22-Oct-97
2605 ----------------------
2607 1. Fixed bug in code for handling temporary memory usage when there are more
2608 back references than supplied space in the ovector. This could cause segfaults.
2611 Version 0.97 21-Oct-97
2612 ----------------------
2614 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2616 2. Optimized negated single characters not to use a bit map.
2618 3. Brought error texts together as macro definitions; clarified some of them;
2619 fixed one that was wrong - it said "range out of order" when it meant "invalid
2620 escape sequence".
2622 4. Changed some char * arguments to const char *.
2624 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2626 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2627 pcretest.
2630 Version 0.96 16-Oct-97
2631 ----------------------
2633 1. Added a simple "pgrep" utility to the distribution.
2635 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2636 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2637 where "ddd" means "one or more decimal digits".
2639 3. Fixed serious bug. If a pattern had a back reference, but the call to
2640 pcre_exec() didn't supply a large enough ovector to record the related
2641 identifying subpattern, the match always failed. PCRE now remembers the number
2642 of the largest back reference, and gets some temporary memory in which to save
2643 the offsets during matching if necessary, in order to ensure that
2644 backreferences always work.
2646 4. Increased the compatibility with Perl in a number of ways:
2648 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2649 to request this handling. The option can be set at compile or exec time.
2651 (b) $ matches before a terminating newline by default; an option
2652 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2653 mode). The option can be set at compile or exec time.
2655 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2656 the same as Perl's. If the decimal number it represents is less than 10
2657 or there aren't that many previous left capturing parentheses, an octal
2658 escape is read. Inside a character class, it's always an octal escape,
2659 even if it is a single digit.
2661 (d) An escaped but undefined alphabetic character is taken as a literal,
2662 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2663 escapes.
2665 (e) {0} is now permitted. (The previous item is removed from the compiled
2666 pattern).
2668 5. Changed all the names of code files so that the basic parts are no longer
2669 than 10 characters, and abolished the teeny "globals.c" file.
2671 6. Changed the handling of character classes; they are now done with a 32-byte
2672 bit map always.
2674 7. Added the -d and /D options to pcretest to make it possible to look at the
2675 internals of compilation without having to recompile pcre.
2678 Version 0.95 23-Sep-97
2679 ----------------------
2681 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2682 \x20 at the start of a run of normal characters. These were being treated as
2683 real characters, instead of the source characters being re-checked.
2686 Version 0.94 18-Sep-97
2687 ----------------------
2689 1. The functions are now thread-safe, with the caveat that the global variables
2690 containing pointers to malloc() and free() or alternative functions are the
2691 same for all threads.
2693 2. Get pcre_study() to generate a bitmap of initial characters for non-
2694 anchored patterns when this is possible, and use it if passed to pcre_exec().
2697 Version 0.93 15-Sep-97
2698 ----------------------
2700 1. /(b)|(:+)/ was computing an incorrect first character.
2702 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2703 but not actually doing anything yet.
2705 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2706 as Perl does (e.g. [-az] or [az-]).
2708 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2709 all possible positions.
2711 5. Split up into different modules to avoid including unneeded functions in a
2712 compiled binary. However, compile and exec are still in one module. The "study"
2713 function is split off.
2715 6. The character tables are now in a separate module whose source is generated
2716 by an auxiliary program - but can then be edited by hand if required. There are
2717 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2718 toupper() in the code.
2720 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2721 make them global. Abolish the function for setting them, as the caller can now
2722 set them directly.
2725 Version 0.92 11-Sep-97
2726 ----------------------
2728 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2729 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2731 2. Caseless matching was not working in character classes if the characters in
2732 the pattern were in upper case.
2734 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2736 4. Make PCRE_ANCHORED public and accept as a compile option.
2738 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2739 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2740 pass them.
2742 6. Give an error if bad option bits passed at compile or run time.
2744 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2745 pcretest to cause it to pass that flag.
2747 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2748 options, and the first character, if set.
2750 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2753 Version 0.91 10-Sep-97
2754 ----------------------
2756 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2757 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2759 2. PCRE was looping on encountering an indefinitely repeated back reference to
2760 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2761 Perl does - treats the match as successful.
2763 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5