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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 158 - (show annotations) (download)
Tue May 1 10:30:14 2007 UTC (7 years, 5 months ago) by ph10
File size: 127175 byte(s)
More test updates.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12