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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 167 - (show annotations) (download)
Wed May 9 15:53:54 2007 UTC (6 years, 11 months ago) by ph10
File size: 128235 byte(s)
Add (?(-n) and (?(+n) relative conditions.

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

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