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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 210 - (show annotations) (download)
Wed Aug 8 14:24:50 2007 UTC (7 years, 1 month ago) by ph10
File size: 134972 byte(s)
Add Perl 5.10's backtracking verbs.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.3 05-Jul-07
5 ---------------------
6
7 1. In the rejigging of the build system that eventually resulted in 7.1, the
8 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9 brackets there is not right, since it causes compilers to look for an
10 installed pcre.h, not the version that is in the source that is being
11 compiled (which of course may be different). I have changed it back to:
12
13 #include "pcre.h"
14
15 I have a vague recollection that the change was concerned with compiling in
16 different directories, but in the new build system, that is taken care of
17 by the VPATH setting the Makefile.
18
19 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20 when the subject happened to end in the byte 0x85 (e.g. if the last
21 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22 characters but of course it shouldn't be taken as a newline when it is part
23 of another character. The bug was that, for an unlimited repeat of . in
24 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25 characters when looking for a newline.
26
27 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28
29 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30 in debug output.
31
32 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33 long printing in the pcrecpp unittest when running under MinGW.
34
35 6. ESC_K was left out of the EBCDIC table.
36
37 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39 limit also applies to "virtual nesting" when a pattern is recursive, and in
40 this case 1000 isn't so big. I have been able to remove this limit at the
41 expense of backing off one optimization in certain circumstances. Normally,
42 when pcre_exec() would call its internal match() function recursively and
43 immediately return the result unconditionally, it uses a "tail recursion"
44 feature to save stack. However, when a subpattern that can match an empty
45 string has an unlimited repetition quantifier, it no longer makes this
46 optimization. That gives it a stack frame in which to save the data for
47 checking that an empty string has been matched. Previously this was taken
48 from the 1000-entry workspace that had been reserved. So now there is no
49 explicit limit, but more stack is used.
50
51 8. Applied Daniel's patches to solve problems with the import/export magic
52 syntax that is required for Windows, and which was going wrong for the
53 pcreposix and pcrecpp parts of the library. These were overlooked when this
54 problem was solved for the main library.
55
56 9. There were some crude static tests to avoid integer overflow when computing
57 the size of patterns that contain repeated groups with explicit upper
58 limits. As the maximum quantifier is 65535, the maximum group length was
59 set at 30,000 so that the product of these two numbers did not overflow a
60 32-bit integer. However, it turns out that people want to use groups that
61 are longer than 30,000 bytes (though not repeat them that many times).
62 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63 made it possible to implement the integer overflow checks in a much more
64 dynamic way, which I have now done. The artificial limitation on group
65 length has been removed - we now have only the limit on the total length of
66 the compiled pattern, which depends on the LINK_SIZE setting.
67
68 10. Fixed a bug in the documentation for get/copy named substring when
69 duplicate names are permitted. If none of the named substrings are set, the
70 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71 empty string.
72
73 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75 because the ] is interpreted as the first data character and the
76 terminating ] is not found. PCRE has been made compatible with Perl in this
77 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78 cause memory overwriting.
79
80 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81 string has been matched (to stop an infinite loop). It was not recognizing
82 a conditional subpattern that could match an empty string if that
83 subpattern was within another subpattern. For example, it looped when
84 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
85 condition was not nested. This bug has been fixed.
86
87 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88 past the start of the subject in the presence of bytes with the top bit
89 set, for example "\x8aBCD".
90
91 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93
94 14. Optimized (?!) to (*FAIL).
95
96
97 Version 7.2 19-Jun-07
98 ---------------------
99
100 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
101 which is apparently normally available under Windows.
102
103 2. Re-jig the pcregrep tests with different newline settings in an attempt
104 to make them independent of the local environment's newline setting.
105
106 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
107
108 4. Some of the "internals" tests were previously cut out when the link size
109 was not 2, because the output contained actual offsets. The recent new
110 "Z" feature of pcretest means that these can be cut out, making the tests
111 usable with all link sizes.
112
113 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
114 stack recursion. This gives a massive performance boost under BSD, but just
115 a small improvement under Linux. However, it saves one field in the frame
116 in all cases.
117
118 6. Added more features from the forthcoming Perl 5.10:
119
120 (a) (?-n) (where n is a string of digits) is a relative subroutine or
121 recursion call. It refers to the nth most recently opened parentheses.
122
123 (b) (?+n) is also a relative subroutine call; it refers to the nth next
124 to be opened parentheses.
125
126 (c) Conditions that refer to capturing parentheses can be specified
127 relatively, for example, (?(-2)... or (?(+3)...
128
129 (d) \K resets the start of the current match so that everything before
130 is not part of it.
131
132 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
133
134 (f) \g{name} is another synonym - part of Perl 5.10's unification of
135 reference syntax.
136
137 (g) (?| introduces a group in which the numbering of parentheses in each
138 alternative starts with the same number.
139
140 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
141
142 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
143 PCRE_INFO_JCHANGED.
144
145 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
146 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
147 for detecting groups that can match an empty string.
148
149 9. A pattern with a very large number of alternatives (more than several
150 hundred) was running out of internal workspace during the pre-compile
151 phase, where pcre_compile() figures out how much memory will be needed. A
152 bit of new cunning has reduced the workspace needed for groups with
153 alternatives. The 1000-alternative test pattern now uses 12 bytes of
154 workspace instead of running out of the 4096 that are available.
155
156 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
157
158 11. Applied patch from Google to remove an optimization that didn't quite work.
159 The report of the bug said:
160
161 pcrecpp::RE("a*").FullMatch("aaa") matches, while
162 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
163 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
164
165 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
166 it matched the wrong number of bytes.
167
168
169 Version 7.1 24-Apr-07
170 ---------------------
171
172 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
173 that is more "standard", making use of automake and other Autotools. There
174 is some re-arrangement of the files and adjustment of comments consequent
175 on this.
176
177 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
178 for recursive directory scanning broke on some systems because the files
179 are not scanned in any specific order and on different systems the order
180 was different. A call to "sort" has been inserted into RunGrepTest for the
181 approprate test as a short-term fix. In the longer term there may be an
182 alternative.
183
184 3. I had an email from Eric Raymond about problems translating some of PCRE's
185 man pages to HTML (despite the fact that I distribute HTML pages, some
186 people do their own conversions for various reasons). The problems
187 concerned the use of low-level troff macros .br and .in. I have therefore
188 removed all such uses from the man pages (some were redundant, some could
189 be replaced by .nf/.fi pairs). The 132html script that I use to generate
190 HTML has been updated to handle .nf/.fi and to complain if it encounters
191 .br or .in.
192
193 4. Updated comments in configure.ac that get placed in config.h.in and also
194 arranged for config.h to be included in the distribution, with the name
195 config.h.generic, for the benefit of those who have to compile without
196 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
197
198 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
199 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
200 makevp.bat for the current PCRE, using the additional files
201 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
202
203 6. A Windows user reported a minor discrepancy with test 2, which turned out
204 to be caused by a trailing space on an input line that had got lost in his
205 copy. The trailing space was an accident, so I've just removed it.
206
207 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
208 that is needed.
209
210 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
211 as "const" (a) because they are and (b) because it helps the PHP
212 maintainers who have recently made a script to detect big data structures
213 in the php code that should be moved to the .rodata section. I remembered
214 to update Builducptable as well, so it won't revert if ucptable.h is ever
215 re-created.
216
217 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
218 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
219 order to be able to cut out the UTF-8 tables in the latter when UTF-8
220 support is not required. This saves 1.5-2K of code, which is important in
221 some applications.
222
223 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
224 so as not to refer to the tables, even though these functions will never be
225 called when UTF-8 support is disabled. Otherwise there are problems with a
226 shared library.
227
228 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
229
230 (a) It was defining its arguments as char * instead of void *.
231
232 (b) It was assuming that all moves were upwards in memory; this was true
233 a long time ago when I wrote it, but is no longer the case.
234
235 The emulated memove() is provided for those environments that have neither
236 memmove() nor bcopy(). I didn't think anyone used it these days, but that
237 is clearly not the case, as these two bugs were recently reported.
238
239 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
240 and Detrail to create the HTML documentation, the .txt form of the man
241 pages, and it removes trailing spaces from listed files. It also creates
242 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
243 case, it wraps all the #defines with #ifndefs. This script should be run
244 before "make dist".
245
246 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
247 with Unicode property support.
248
249 (a) For a maximizing quantifier, if the two different cases of the
250 character were of different lengths in their UTF-8 codings (there are
251 some cases like this - I found 11), and the matching function had to
252 back up over a mixture of the two cases, it incorrectly assumed they
253 were both the same length.
254
255 (b) When PCRE was configured to use the heap rather than the stack for
256 recursion during matching, it was not correctly preserving the data for
257 the other case of a UTF-8 character when checking ahead for a match
258 while processing a minimizing repeat. If the check also involved
259 matching a wide character, but failed, corruption could cause an
260 erroneous result when trying to check for a repeat of the original
261 character.
262
263 13. Some tidying changes to the testing mechanism:
264
265 (a) The RunTest script now detects the internal link size and whether there
266 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
267 values substituted by "configure". (The RunGrepTest script already did
268 this for UTF-8.) The configure.ac script no longer substitutes the
269 relevant variables.
270
271 (b) The debugging options /B and /D in pcretest show the compiled bytecode
272 with length and offset values. This means that the output is different
273 for different internal link sizes. Test 2 is skipped for link sizes
274 other than 2 because of this, bypassing the problem. Unfortunately,
275 there was also a test in test 3 (the locale tests) that used /B and
276 failed for link sizes other than 2. Rather than cut the whole test out,
277 I have added a new /Z option to pcretest that replaces the length and
278 offset values with spaces. This is now used to make test 3 independent
279 of link size. (Test 2 will be tidied up later.)
280
281 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
282 segmentation fault instead of returning the appropriate error message.
283
284 15. In multiline mode when the newline sequence was set to "any", the pattern
285 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
286 This doesn't seem right; it now treats the CRLF combination as the line
287 ending, and so does not match in that case. It's only a pattern such as ^$
288 that would hit this one: something like ^ABC$ would have failed after \r
289 and then tried again after \r\n.
290
291 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
292 in an attempt to make files that differ only in their line terminators
293 compare equal. This works on Linux.
294
295 17. Under certain error circumstances pcregrep might try to free random memory
296 as it exited. This is now fixed, thanks to valgrind.
297
298 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
299 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
300 was because its rules for how to advance for /g after matching an empty
301 string at the end of a line did not allow for this case. They now check for
302 it specially.
303
304 20. pcretest is supposed to handle patterns and data of any length, by
305 extending its buffers when necessary. It was getting this wrong when the
306 buffer for a data line had to be extended.
307
308 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
309 CRLF as a newline sequence.
310
311 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
312 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
313 I have nevertheless tidied it up.
314
315 23. Added some casts to kill warnings from HP-UX ia64 compiler.
316
317 24. Added a man page for pcre-config.
318
319
320 Version 7.0 19-Dec-06
321 ---------------------
322
323 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
324 moving to gcc 4.1.1.
325
326 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
327 sys/time.h, which is documented as needed for this function. It doesn't
328 seem to matter on Linux, but it showed up on some releases of OS X.
329
330 3. It seems that there are systems where bytes whose values are greater than
331 127 match isprint() in the "C" locale. The "C" locale should be the
332 default when a C program starts up. In most systems, only ASCII printing
333 characters match isprint(). This difference caused the output from pcretest
334 to vary, making some of the tests fail. I have changed pcretest so that:
335
336 (a) When it is outputting text in the compiled version of a pattern, bytes
337 other than 32-126 are always shown as hex escapes.
338
339 (b) When it is outputting text that is a matched part of a subject string,
340 it does the same, unless a different locale has been set for the match
341 (using the /L modifier). In this case, it uses isprint() to decide.
342
343 4. Fixed a major bug that caused incorrect computation of the amount of memory
344 required for a compiled pattern when options that changed within the
345 pattern affected the logic of the preliminary scan that determines the
346 length. The relevant options are -x, and -i in UTF-8 mode. The result was
347 that the computed length was too small. The symptoms of this bug were
348 either the PCRE error "internal error: code overflow" from pcre_compile(),
349 or a glibc crash with a message such as "pcretest: free(): invalid next
350 size (fast)". Examples of patterns that provoked this bug (shown in
351 pcretest format) are:
352
353 /(?-x: )/x
354 /(?x)(?-x: \s*#\s*)/
355 /((?i)[\x{c0}])/8
356 /(?i:[\x{c0}])/8
357
358 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
359 is now done differently.
360
361 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
362 wrapper classes; (b) implement a new function in the C++ scanner that is
363 more efficient than the old way of doing things because it avoids levels of
364 recursion in the regex matching; (c) add a paragraph to the documentation
365 for the FullMatch() function.
366
367 6. The escape sequence \n was being treated as whatever was defined as
368 "newline". Not only was this contrary to the documentation, which states
369 that \n is character 10 (hex 0A), but it also went horribly wrong when
370 "newline" was defined as CRLF. This has been fixed.
371
372 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
373 was being set to -1 for the "end of line" case (supposedly a value that no
374 character can have). Though this value is never used (the check for end of
375 line is "zero bytes in current character"), it caused compiler complaints.
376 I've changed it to 0xffffffff.
377
378 8. In pcre_version.c, the version string was being built by a sequence of
379 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
380 string (as it is for production releases) called a macro with an empty
381 argument. The C standard says the result of this is undefined. The gcc
382 compiler treats it as an empty string (which was what was wanted) but it is
383 reported that Visual C gives an error. The source has been hacked around to
384 avoid this problem.
385
386 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
387 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
388 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
389 of them did).
390
391 10. Originally, pcretest opened its input and output without "b"; then I was
392 told that "b" was needed in some environments, so it was added for release
393 5.0 to both the input and output. (It makes no difference on Unix-like
394 systems.) Later I was told that it is wrong for the input on Windows. I've
395 now abstracted the modes into two macros, to make it easier to fiddle with
396 them, and removed "b" from the input mode under Windows.
397
398 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
399
400 12. Added -help and --help to pcretest as an official way of being reminded
401 of the options.
402
403 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
404 and pcrecpp.cc because they annoy compilers at high warning levels.
405
406 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
407
408 15. Fixed an occurrence of == in configure.ac that should have been = (shell
409 scripts are not C programs :-) and which was not noticed because it works
410 on Linux.
411
412 16. pcretest is supposed to handle any length of pattern and data line (as one
413 line or as a continued sequence of lines) by extending its input buffer if
414 necessary. This feature was broken for very long pattern lines, leading to
415 a string of junk being passed to pcre_compile() if the pattern was longer
416 than about 50K.
417
418 17. I have done a major re-factoring of the way pcre_compile() computes the
419 amount of memory needed for a compiled pattern. Previously, there was code
420 that made a preliminary scan of the pattern in order to do this. That was
421 OK when PCRE was new, but as the facilities have expanded, it has become
422 harder and harder to keep it in step with the real compile phase, and there
423 have been a number of bugs (see for example, 4 above). I have now found a
424 cunning way of running the real compile function in a "fake" mode that
425 enables it to compute how much memory it would need, while actually only
426 ever using a few hundred bytes of working memory and without too many
427 tests of the mode. This should make future maintenance and development
428 easier. A side effect of this work is that the limit of 200 on the nesting
429 depth of parentheses has been removed (though this was never a serious
430 limitation, I suspect). However, there is a downside: pcre_compile() now
431 runs more slowly than before (30% or more, depending on the pattern). I
432 hope this isn't a big issue. There is no effect on runtime performance.
433
434 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
435 newline (only possible for the last line of a file) and it was a
436 pattern that set a locale (followed by /Lsomething), pcretest crashed.
437
438 19. Added additional timing features to pcretest. (1) The -tm option now times
439 matching only, not compiling. (2) Both -t and -tm can be followed, as a
440 separate command line item, by a number that specifies the number of
441 repeats to use when timing. The default is 50000; this gives better
442 precision, but takes uncomfortably long for very large patterns.
443
444 20. Extended pcre_study() to be more clever in cases where a branch of a
445 subpattern has no definite first character. For example, (a*|b*)[cd] would
446 previously give no result from pcre_study(). Now it recognizes that the
447 first character must be a, b, c, or d.
448
449 21. There was an incorrect error "recursive call could loop indefinitely" if
450 a subpattern (or the entire pattern) that was being tested for matching an
451 empty string contained only one non-empty item after a nested subpattern.
452 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
453 incorrectly, because the \d was being skipped in the check.
454
455 22. The pcretest program now has a new pattern option /B and a command line
456 option -b, which is equivalent to adding /B to every pattern. This causes
457 it to show the compiled bytecode, without the additional information that
458 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
459 is the same as /B/I).
460
461 23. A new optimization is now able automatically to treat some sequences such
462 as a*b as a*+b. More specifically, if something simple (such as a character
463 or a simple class like \d) has an unlimited quantifier, and is followed by
464 something that cannot possibly match the quantified thing, the quantifier
465 is automatically "possessified".
466
467 24. A recursive reference to a subpattern whose number was greater than 39
468 went wrong under certain circumstances in UTF-8 mode. This bug could also
469 have affected the operation of pcre_study().
470
471 25. Realized that a little bit of performance could be had by replacing
472 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
473
474 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
475
476 27. Possessive quantifiers such as a++ were previously implemented by turning
477 them into atomic groups such as ($>a+). Now they have their own opcodes,
478 which improves performance. This includes the automatically created ones
479 from 23 above.
480
481 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
482 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
483 the first matched character to be a colon. This applied both to named and
484 numbered groups.
485
486 29. The ucpinternal.h header file was missing its idempotency #ifdef.
487
488 30. I was sent a "project" file called libpcre.a.dev which I understand makes
489 building PCRE on Windows easier, so I have included it in the distribution.
490
491 31. There is now a check in pcretest against a ridiculously large number being
492 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
493 loop, the loop is abandoned.
494
495 32. Forward references to subpatterns in conditions such as (?(2)...) where
496 subpattern 2 is defined later cause pcre_compile() to search forwards in
497 the pattern for the relevant set of parentheses. This search went wrong
498 when there were unescaped parentheses in a character class, parentheses
499 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
500
501 33. "Subroutine" calls and backreferences were previously restricted to
502 referencing subpatterns earlier in the regex. This restriction has now
503 been removed.
504
505 34. Added a number of extra features that are going to be in Perl 5.10. On the
506 whole, these are just syntactic alternatives for features that PCRE had
507 previously implemented using the Python syntax or my own invention. The
508 other formats are all retained for compatibility.
509
510 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
511 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
512 also .NET compatible.
513
514 (b) A recursion or subroutine call to a named group can now be defined as
515 (?&name) as well as (?P>name).
516
517 (c) A backreference to a named group can now be defined as \k<name> or
518 \k'name' as well as (?P=name). The new forms, as well as being in Perl
519 5.10, are also .NET compatible.
520
521 (d) A conditional reference to a named group can now use the syntax
522 (?(<name>) or (?('name') as well as (?(name).
523
524 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
525 groups (named and numbered) that are never evaluated inline, but can be
526 called as "subroutines" from elsewhere. In effect, the DEFINE condition
527 is always false. There may be only one alternative in such a group.
528
529 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
530 as the simple (?(R). The condition is true only if the most recent
531 recursion is that of the given number or name. It does not search out
532 through the entire recursion stack.
533
534 (g) The escape \gN or \g{N} has been added, where N is a positive or
535 negative number, specifying an absolute or relative reference.
536
537 35. Tidied to get rid of some further signed/unsigned compiler warnings and
538 some "unreachable code" warnings.
539
540 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
541 things, this adds five new scripts.
542
543 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
544 There were also incompatibilities regarding the handling of \Q..\E inside
545 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
546 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
547
548 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
549 matches an empty string, and forcibly breaks the loop. There were bugs in
550 this code in non-simple cases. For a pattern such as ^(a()*)* matched
551 against aaaa the result was just "a" rather than "aaaa", for example. Two
552 separate and independent bugs (that affected different cases) have been
553 fixed.
554
555 39. Refactored the code to abolish the use of different opcodes for small
556 capturing bracket numbers. This is a tidy that I avoided doing when I
557 removed the limit on the number of capturing brackets for 3.5 back in 2001.
558 The new approach is not only tidier, it makes it possible to reduce the
559 memory needed to fix the previous bug (38).
560
561 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
562 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
563 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
564 mode.
565
566 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
567 report.
568
569 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
570 copy construction and assignment in the C++ wrapper.
571
572 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
573 couple of bugs that could have given wrong results in the "--newline=crlf"
574 case.
575
576 44. Added a number of casts and did some reorganization of signed/unsigned int
577 variables following suggestions from Dair Grant. Also renamed the variable
578 "this" as "item" because it is a C++ keyword.
579
580 45. Arranged for dftables to add
581
582 #include "pcre_internal.h"
583
584 to pcre_chartables.c because without it, gcc 4.x may remove the array
585 definition from the final binary if PCRE is built into a static library and
586 dead code stripping is activated.
587
588 46. For an unanchored pattern, if a match attempt fails at the start of a
589 newline sequence, and the newline setting is CRLF or ANY, and the next two
590 characters are CRLF, advance by two characters instead of one.
591
592
593 Version 6.7 04-Jul-06
594 ---------------------
595
596 1. In order to handle tests when input lines are enormously long, pcretest has
597 been re-factored so that it automatically extends its buffers when
598 necessary. The code is crude, but this _is_ just a test program. The
599 default size has been increased from 32K to 50K.
600
601 2. The code in pcre_study() was using the value of the re argument before
602 testing it for NULL. (Of course, in any sensible call of the function, it
603 won't be NULL.)
604
605 3. The memmove() emulation function in pcre_internal.h, which is used on
606 systems that lack both memmove() and bcopy() - that is, hardly ever -
607 was missing a "static" storage class specifier.
608
609 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
610 containing an extended class (one that cannot be represented by a bitmap
611 because it contains high-valued characters or Unicode property items, e.g.
612 [\pZ]). Almost always one would set UTF-8 mode when processing such a
613 pattern, but PCRE should not loop if you do not (it no longer does).
614 [Detail: two cases were found: (a) a repeated subpattern containing an
615 extended class; (b) a recursive reference to a subpattern that followed a
616 previous extended class. It wasn't skipping over the extended class
617 correctly when UTF-8 mode was not set.]
618
619 5. A negated single-character class was not being recognized as fixed-length
620 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
621 compile error "lookbehind assertion is not fixed length".
622
623 6. The RunPerlTest auxiliary script was showing an unexpected difference
624 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
625 write a Perl script that can interpret lines of an input file either as
626 byte characters or as UTF-8, which is what "perltest" was being required to
627 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
628 can't do is switch easily at run time between having the "use utf8;" pragma
629 or not. In the end, I fudged it by using the RunPerlTest script to insert
630 "use utf8;" explicitly for the UTF-8 tests.
631
632 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
633 the end of the subject string, contrary to the documentation and to what
634 Perl does. This was true of both matching functions. Now it matches only at
635 the start of the subject and immediately after *internal* newlines.
636
637 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
638 a pointer to an int instead of a pointer to an unsigned long int. This
639 caused problems on 64-bit systems.
640
641 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
642 instance of the 'standard' template library not being so standard".
643
644 10. There was no check on the number of named subpatterns nor the maximum
645 length of a subpattern name. The product of these values is used to compute
646 the size of the memory block for a compiled pattern. By supplying a very
647 long subpattern name and a large number of named subpatterns, the size
648 computation could be caused to overflow. This is now prevented by limiting
649 the length of names to 32 characters, and the number of named subpatterns
650 to 10,000.
651
652 11. Subpatterns that are repeated with specific counts have to be replicated in
653 the compiled pattern. The size of memory for this was computed from the
654 length of the subpattern and the repeat count. The latter is limited to
655 65535, but there was no limit on the former, meaning that integer overflow
656 could in principle occur. The compiled length of a repeated subpattern is
657 now limited to 30,000 bytes in order to prevent this.
658
659 12. Added the optional facility to have named substrings with the same name.
660
661 13. Added the ability to use a named substring as a condition, using the
662 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
663 are numbers (not recommended). Forward references are permitted.
664
665 14. Added forward references in named backreferences (if you see what I mean).
666
667 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
668 pattern could run off the end of the subject. For example, the pattern
669 "(?s)(.{1,5})"8 did this with the subject "ab".
670
671 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
672 PCRE_CASELESS was set when matching characters that were quantified with ?
673 or *.
674
675 17. A character class other than a single negated character that had a minimum
676 but no maximum quantifier - for example [ab]{6,} - was not handled
677 correctly by pce_dfa_exec(). It would match only one character.
678
679 18. A valid (though odd) pattern that looked like a POSIX character
680 class but used an invalid character after [ (for example [[,abc,]]) caused
681 pcre_compile() to give the error "Failed: internal error: code overflow" or
682 in some cases to crash with a glibc free() error. This could even happen if
683 the pattern terminated after [[ but there just happened to be a sequence of
684 letters, a binary zero, and a closing ] in the memory that followed.
685
686 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
687 over the years. Originally (before any Unicode support), just the bottom 8
688 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
689 output from "man perlunicode" includes this:
690
691 The regular expression compiler produces polymorphic opcodes. That
692 is, the pattern adapts to the data and automatically switches to
693 the Unicode character scheme when presented with Unicode data--or
694 instead uses a traditional byte scheme when presented with byte
695 data.
696
697 Sadly, a wide octal escape does not cause a switch, and in a string with
698 no other multibyte characters, these octal escapes are treated as before.
699 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
700 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
701 Unicode string.
702
703 I have not perpetrated such confusion in PCRE. Up till now, it took just
704 the bottom 8 bits, as in old Perl. I have now made octal escapes with
705 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
706 translate to the appropriate multibyte character.
707
708 29. Applied some refactoring to reduce the number of warnings from Microsoft
709 and Borland compilers. This has included removing the fudge introduced
710 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
711 a warning about an unused variable.
712
713 21. PCRE has not included VT (character 0x0b) in the set of whitespace
714 characters since release 4.0, because Perl (from release 5.004) does not.
715 [Or at least, is documented not to: some releases seem to be in conflict
716 with the documentation.] However, when a pattern was studied with
717 pcre_study() and all its branches started with \s, PCRE still included VT
718 as a possible starting character. Of course, this did no harm; it just
719 caused an unnecessary match attempt.
720
721 22. Removed a now-redundant internal flag bit that recorded the fact that case
722 dependency changed within the pattern. This was once needed for "required
723 byte" processing, but is no longer used. This recovers a now-scarce options
724 bit. Also moved the least significant internal flag bit to the most-
725 significant bit of the word, which was not previously used (hangover from
726 the days when it was an int rather than a uint) to free up another bit for
727 the future.
728
729 23. Added support for CRLF line endings as well as CR and LF. As well as the
730 default being selectable at build time, it can now be changed at runtime
731 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
732 specify that it is scanning data with non-default line endings.
733
734 24. Changed the definition of CXXLINK to make it agree with the definition of
735 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
736
737 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
738 recursions. This makes a big different to stack usage for some patterns.
739
740 26. If a subpattern containing a named recursion or subroutine reference such
741 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
742 the space required for the compiled pattern went wrong and gave too small a
743 value. Depending on the environment, this could lead to "Failed: internal
744 error: code overflow at offset 49" or "glibc detected double free or
745 corruption" errors.
746
747 27. Applied patches from Google (a) to support the new newline modes and (b) to
748 advance over multibyte UTF-8 characters in GlobalReplace.
749
750 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
751 difference for some implementation of PCRE in some Windows version.
752
753 29. Added some extra testing facilities to pcretest:
754
755 \q<number> in a data line sets the "match limit" value
756 \Q<number> in a data line sets the "match recursion limt" value
757 -S <number> sets the stack size, where <number> is in megabytes
758
759 The -S option isn't available for Windows.
760
761
762 Version 6.6 06-Feb-06
763 ---------------------
764
765 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
766 in pcreposix.h. I have copied the definition from pcre.h.
767
768 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
769 because pcre.h is no longer a built file.
770
771 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
772 not normally included in the compiled code.
773
774
775 Version 6.5 01-Feb-06
776 ---------------------
777
778 1. When using the partial match feature with pcre_dfa_exec(), it was not
779 anchoring the second and subsequent partial matches at the new starting
780 point. This could lead to incorrect results. For example, with the pattern
781 /1234/, partially matching against "123" and then "a4" gave a match.
782
783 2. Changes to pcregrep:
784
785 (a) All non-match returns from pcre_exec() were being treated as failures
786 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
787 error message is output. Some extra information is given for the
788 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
789 probably the only errors that are likely to be caused by users (by
790 specifying a regex that has nested indefinite repeats, for instance).
791 If there are more than 20 of these errors, pcregrep is abandoned.
792
793 (b) A binary zero was treated as data while matching, but terminated the
794 output line if it was written out. This has been fixed: binary zeroes
795 are now no different to any other data bytes.
796
797 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
798 used to set a locale for matching. The --locale=xxxx long option has
799 been added (no short equivalent) to specify a locale explicitly on the
800 pcregrep command, overriding the environment variables.
801
802 (d) When -B was used with -n, some line numbers in the output were one less
803 than they should have been.
804
805 (e) Added the -o (--only-matching) option.
806
807 (f) If -A or -C was used with -c (count only), some lines of context were
808 accidentally printed for the final match.
809
810 (g) Added the -H (--with-filename) option.
811
812 (h) The combination of options -rh failed to suppress file names for files
813 that were found from directory arguments.
814
815 (i) Added the -D (--devices) and -d (--directories) options.
816
817 (j) Added the -F (--fixed-strings) option.
818
819 (k) Allow "-" to be used as a file name for -f as well as for a data file.
820
821 (l) Added the --colo(u)r option.
822
823 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
824 is not present by default.
825
826 3. A nasty bug was discovered in the handling of recursive patterns, that is,
827 items such as (?R) or (?1), when the recursion could match a number of
828 alternatives. If it matched one of the alternatives, but subsequently,
829 outside the recursion, there was a failure, the code tried to back up into
830 the recursion. However, because of the way PCRE is implemented, this is not
831 possible, and the result was an incorrect result from the match.
832
833 In order to prevent this happening, the specification of recursion has
834 been changed so that all such subpatterns are automatically treated as
835 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
836
837 4. I had overlooked the fact that, in some locales, there are characters for
838 which isalpha() is true but neither isupper() nor islower() are true. In
839 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
840 and ordfeminine) are like this. This affected the treatment of \w and \W
841 when they appeared in character classes, but not when they appeared outside
842 a character class. The bit map for "word" characters is now created
843 separately from the results of isalnum() instead of just taking it from the
844 upper, lower, and digit maps. (Plus the underscore character, of course.)
845
846 5. The above bug also affected the handling of POSIX character classes such as
847 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
848 permanent tables. Instead, the bit maps for such a class were previously
849 created as the appropriate unions of the upper, lower, and digit bitmaps.
850 Now they are created by subtraction from the [[:word:]] class, which has
851 its own bitmap.
852
853 6. The [[:blank:]] character class matches horizontal, but not vertical space.
854 It is created by subtracting the vertical space characters (\x09, \x0a,
855 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
856 subtraction was done in the overall bitmap for a character class, meaning
857 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
858 be recognized. This bug has been fixed.
859
860 7. Patches from the folks at Google:
861
862 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
863 real life, but is still worth protecting against".
864
865 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
866 regular expressions".
867
868 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
869 have it.
870
871 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
872 "configure" and the latter not, in order to fix a problem somebody had
873 with compiling the Arg class on HP-UX.
874
875 (e) Improve the error-handling of the C++ wrapper a little bit.
876
877 (f) New tests for checking recursion limiting.
878
879 8. The pcre_memmove() function, which is used only if the environment does not
880 have a standard memmove() function (and is therefore rarely compiled),
881 contained two bugs: (a) use of int instead of size_t, and (b) it was not
882 returning a result (though PCRE never actually uses the result).
883
884 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
885 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
886 returned instead of calling malloc() with an overflowing number that would
887 most likely cause subsequent chaos.
888
889 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
890
891 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
892 with this option is matched, the nmatch and pmatch options of regexec() are
893 ignored.
894
895 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
896 provided in case anyone wants to the the POSIX interface with UTF-8
897 strings.
898
899 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
900 C++ linking (needed for some HP-UX environments).
901
902 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
903 (unused parameter) and in the pcre_printint() function (omitted "default"
904 switch label when the default is to do nothing).
905
906 15. Added some code to make it possible, when PCRE is compiled as a C++
907 library, to replace subject pointers for pcre_exec() with a smart pointer
908 class, thus making it possible to process discontinuous strings.
909
910 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
911 much the same function. They were added by different people who were trying
912 to make PCRE easy to compile on non-Unix systems. It has been suggested
913 that PCRE_EXPORT be abolished now that there is more automatic apparatus
914 for compiling on Windows systems. I have therefore replaced it with
915 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
916 defaults to "extern" for C or "extern C" for C++, which works fine on
917 Unix-like systems. It is now possible to override the value of PCRE_DATA_
918 SCOPE with something explicit in config.h. In addition:
919
920 (a) pcreposix.h still had just "extern" instead of either of these macros;
921 I have replaced it with PCRE_DATA_SCOPE.
922
923 (b) Functions such as _pcre_xclass(), which are internal to the library,
924 but external in the C sense, all had PCRE_EXPORT in their definitions.
925 This is apparently wrong for the Windows case, so I have removed it.
926 (It makes no difference on Unix-like systems.)
927
928 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
929 of recursive calls to match(). This is different to MATCH_LIMIT because
930 that limits the total number of calls to match(), not all of which increase
931 the depth of recursion. Limiting the recursion depth limits the amount of
932 stack (or heap if NO_RECURSE is set) that is used. The default can be set
933 when PCRE is compiled, and changed at run time. A patch from Google adds
934 this functionality to the C++ interface.
935
936 18. Changes to the handling of Unicode character properties:
937
938 (a) Updated the table to Unicode 4.1.0.
939
940 (b) Recognize characters that are not in the table as "Cn" (undefined).
941
942 (c) I revised the way the table is implemented to a much improved format
943 which includes recognition of ranges. It now supports the ranges that
944 are defined in UnicodeData.txt, and it also amalgamates other
945 characters into ranges. This has reduced the number of entries in the
946 table from around 16,000 to around 3,000, thus reducing its size
947 considerably. I realized I did not need to use a tree structure after
948 all - a binary chop search is just as efficient. Having reduced the
949 number of entries, I extended their size from 6 bytes to 8 bytes to
950 allow for more data.
951
952 (d) Added support for Unicode script names via properties such as \p{Han}.
953
954 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
955 matching that character.
956
957 20. When matching a repeated Unicode property with a minimum greater than zero,
958 (for example \pL{2,}), PCRE could look past the end of the subject if it
959 reached it while seeking the minimum number of characters. This could
960 happen only if some of the characters were more than one byte long, because
961 there is a check for at least the minimum number of bytes.
962
963 21. Refactored the implementation of \p and \P so as to be more general, to
964 allow for more different types of property in future. This has changed the
965 compiled form incompatibly. Anybody with saved compiled patterns that use
966 \p or \P will have to recompile them.
967
968 22. Added "Any" and "L&" to the supported property types.
969
970 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
971 but give a compile time error if the value is greater than 0xff.
972
973 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
974 accidentally not being installed or uninstalled.
975
976 25. The pcre.h file was built from pcre.h.in, but the only changes that were
977 made were to insert the current release number. This seemed silly, because
978 it made things harder for people building PCRE on systems that don't run
979 "configure". I have turned pcre.h into a distributed file, no longer built
980 by "configure", with the version identification directly included. There is
981 no longer a pcre.h.in file.
982
983 However, this change necessitated a change to the pcre-config script as
984 well. It is built from pcre-config.in, and one of the substitutions was the
985 release number. I have updated configure.ac so that ./configure now finds
986 the release number by grepping pcre.h.
987
988 26. Added the ability to run the tests under valgrind.
989
990
991 Version 6.4 05-Sep-05
992 ---------------------
993
994 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
995 "--" to be printed when multiple files were scanned, even when none of the
996 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
997 consider it to be a bug, and have restored the previous behaviour.
998
999 2. A couple of code tidies to get rid of compiler warnings.
1000
1001 3. The pcretest program used to cheat by referring to symbols in the library
1002 whose names begin with _pcre_. These are internal symbols that are not
1003 really supposed to be visible externally, and in some environments it is
1004 possible to suppress them. The cheating is now confined to including
1005 certain files from the library's source, which is a bit cleaner.
1006
1007 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1008 file's purpose clearer.
1009
1010 5. Reorganized pcre_ucp_findchar().
1011
1012
1013 Version 6.3 15-Aug-05
1014 ---------------------
1015
1016 1. The file libpcre.pc.in did not have general read permission in the tarball.
1017
1018 2. There were some problems when building without C++ support:
1019
1020 (a) If C++ support was not built, "make install" and "make test" still
1021 tried to test it.
1022
1023 (b) There were problems when the value of CXX was explicitly set. Some
1024 changes have been made to try to fix these, and ...
1025
1026 (c) --disable-cpp can now be used to explicitly disable C++ support.
1027
1028 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1029 backslash in a target when C++ was disabled. This confuses some
1030 versions of "make", apparently. Using an intermediate variable solves
1031 this. (Same for CPP_LOBJ.)
1032
1033 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1034 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1035 necessary on certain architectures.
1036
1037 4. Added a setting of -export-symbols-regex to the link command to remove
1038 those symbols that are exported in the C sense, but actually are local
1039 within the library, and not documented. Their names all begin with
1040 "_pcre_". This is not a perfect job, because (a) we have to except some
1041 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1042 available (and never for static libraries). I have made a note to try to
1043 find a way round (a) in the future.
1044
1045
1046 Version 6.2 01-Aug-05
1047 ---------------------
1048
1049 1. There was no test for integer overflow of quantifier values. A construction
1050 such as {1111111111111111} would give undefined results. What is worse, if
1051 a minimum quantifier for a parenthesized subpattern overflowed and became
1052 negative, the calculation of the memory size went wrong. This could have
1053 led to memory overwriting.
1054
1055 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1056
1057 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1058 operating environments where this matters.
1059
1060 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1061 PCRE options from within the C++ wrapper.
1062
1063 5. Named capturing subpatterns were not being correctly counted when a pattern
1064 was compiled. This caused two problems: (a) If there were more than 100
1065 such subpatterns, the calculation of the memory needed for the whole
1066 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1067 back references of the form \12, where the number was greater than 9, were
1068 not recognized as back references, even though there were sufficient
1069 previous subpatterns.
1070
1071 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1072 versions of gcc, e.g. 2.95.4.
1073
1074
1075 Version 6.1 21-Jun-05
1076 ---------------------
1077
1078 1. There was one reference to the variable "posix" in pcretest.c that was not
1079 surrounded by "#if !defined NOPOSIX".
1080
1081 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1082 the cross-check on the old pcre_info() function, for the benefit of the
1083 cut-down version of PCRE that is currently imported into Exim.
1084
1085 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1086 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1087 patterns that start (?i) but I don't think that matters. The use of (?i) is
1088 just an example; this all applies to the other options as well.
1089
1090 4. Since libtool seems to echo the compile commands it is issuing, the output
1091 from "make" can be reduced a bit by putting "@" in front of each libtool
1092 compile command.
1093
1094 5. Patch from the folks at Google for configure.in to be a bit more thorough
1095 in checking for a suitable C++ installation before trying to compile the
1096 C++ stuff. This should fix a reported problem when a compiler was present,
1097 but no suitable headers.
1098
1099 6. The man pages all had just "PCRE" as their title. I have changed them to
1100 be the relevant file name. I have also arranged that these names are
1101 retained in the file doc/pcre.txt, which is a concatenation in text format
1102 of all the man pages except the little individual ones for each function.
1103
1104 7. The NON-UNIX-USE file had not been updated for the different set of source
1105 files that come with release 6. I also added a few comments about the C++
1106 wrapper.
1107
1108
1109 Version 6.0 07-Jun-05
1110 ---------------------
1111
1112 1. Some minor internal re-organization to help with my DFA experiments.
1113
1114 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1115 didn't matter for the library itself when fully configured, but did matter
1116 when compiling without UCP support, or within Exim, where the ucp files are
1117 not imported.
1118
1119 3. Refactoring of the library code to split up the various functions into
1120 different source modules. The addition of the new DFA matching code (see
1121 below) to a single monolithic source would have made it really too
1122 unwieldy, quite apart from causing all the code to be include in a
1123 statically linked application, when only some functions are used. This is
1124 relevant even without the DFA addition now that patterns can be compiled in
1125 one application and matched in another.
1126
1127 The downside of splitting up is that there have to be some external
1128 functions and data tables that are used internally in different modules of
1129 the library but which are not part of the API. These have all had their
1130 names changed to start with "_pcre_" so that they are unlikely to clash
1131 with other external names.
1132
1133 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1134 a different (DFA) algorithm. Although it is slower than the original
1135 function, it does have some advantages for certain types of matching
1136 problem.
1137
1138 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1139 including restarting after a partial match.
1140
1141 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1142 defined when compiling for Windows was sent to me. I have put it into the
1143 code, though I have no means of testing or verifying it.
1144
1145 7. Added the pcre_refcount() auxiliary function.
1146
1147 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1148 match before or at the first newline in the subject string. In pcretest,
1149 the /f option on a pattern can be used to set this.
1150
1151 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1152 would behave wrongly. This has been present in PCRE since release 4.0.
1153
1154 10. A number of changes to the pcregrep command:
1155
1156 (a) Refactored how -x works; insert ^(...)$ instead of setting
1157 PCRE_ANCHORED and checking the length, in preparation for adding
1158 something similar for -w.
1159
1160 (b) Added the -w (match as a word) option.
1161
1162 (c) Refactored the way lines are read and buffered so as to have more
1163 than one at a time available.
1164
1165 (d) Implemented a pcregrep test script.
1166
1167 (e) Added the -M (multiline match) option. This allows patterns to match
1168 over several lines of the subject. The buffering ensures that at least
1169 8K, or the rest of the document (whichever is the shorter) is available
1170 for matching (and similarly the previous 8K for lookbehind assertions).
1171
1172 (f) Changed the --help output so that it now says
1173
1174 -w, --word-regex(p)
1175
1176 instead of two lines, one with "regex" and the other with "regexp"
1177 because that confused at least one person since the short forms are the
1178 same. (This required a bit of code, as the output is generated
1179 automatically from a table. It wasn't just a text change.)
1180
1181 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1182 option but starts with a hyphen. Could be a pattern or a path name
1183 starting with a hyphen, for instance.
1184
1185 (h) "-" can be given as a file name to represent stdin.
1186
1187 (i) When file names are being printed, "(standard input)" is used for
1188 the standard input, for compatibility with GNU grep. Previously
1189 "<stdin>" was used.
1190
1191 (j) The option --label=xxx can be used to supply a name to be used for
1192 stdin when file names are being printed. There is no short form.
1193
1194 (k) Re-factored the options decoding logic because we are going to add
1195 two more options that take data. Such options can now be given in four
1196 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1197
1198 (l) Added the -A, -B, and -C options for requesting that lines of context
1199 around matches be printed.
1200
1201 (m) Added the -L option to print the names of files that do not contain
1202 any matching lines, that is, the complement of -l.
1203
1204 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1205 continue to scan other files.
1206
1207 (o) The -s option was incorrectly implemented. For compatibility with other
1208 greps, it now suppresses the error message for a non-existent or non-
1209 accessible file (but not the return code). There is a new option called
1210 -q that suppresses the output of matching lines, which was what -s was
1211 previously doing.
1212
1213 (p) Added --include and --exclude options to specify files for inclusion
1214 and exclusion when recursing.
1215
1216 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1217 Hopefully, it now does.
1218
1219 12. Missing cast in pcre_study().
1220
1221 13. Added an "uninstall" target to the makefile.
1222
1223 14. Replaced "extern" in the function prototypes in Makefile.in with
1224 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1225 world, but is set differently for Windows.
1226
1227 15. Added a second compiling function called pcre_compile2(). The only
1228 difference is that it has an extra argument, which is a pointer to an
1229 integer error code. When there is a compile-time failure, this is set
1230 non-zero, in addition to the error test pointer being set to point to an
1231 error message. The new argument may be NULL if no error number is required
1232 (but then you may as well call pcre_compile(), which is now just a
1233 wrapper). This facility is provided because some applications need a
1234 numeric error indication, but it has also enabled me to tidy up the way
1235 compile-time errors are handled in the POSIX wrapper.
1236
1237 16. Added VPATH=.libs to the makefile; this should help when building with one
1238 prefix path and installing with another. (Or so I'm told by someone who
1239 knows more about this stuff than I do.)
1240
1241 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1242 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1243 match everything, including newlines. This is not POSIX-compatible, but
1244 somebody wanted the feature. From pcretest it can be activated by using
1245 both the P and the s flags.
1246
1247 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1248
1249 19. libpcre.pc was being incorrectly installed as executable.
1250
1251 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1252 it now also looks for '\r' so that it will work unmodified on Windows.
1253
1254 21. Added Google's contributed C++ wrapper to the distribution.
1255
1256 22. Added some untidy missing memory free() calls in pcretest, to keep
1257 Electric Fence happy when testing.
1258
1259
1260
1261 Version 5.0 13-Sep-04
1262 ---------------------
1263
1264 1. Internal change: literal characters are no longer packed up into items
1265 containing multiple characters in a single byte-string. Each character
1266 is now matched using a separate opcode. However, there may be more than one
1267 byte in the character in UTF-8 mode.
1268
1269 2. The pcre_callout_block structure has two new fields: pattern_position and
1270 next_item_length. These contain the offset in the pattern to the next match
1271 item, and its length, respectively.
1272
1273 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1274 insertion of callouts before each pattern item. Added the /C option to
1275 pcretest to make use of this.
1276
1277 4. On the advice of a Windows user, the lines
1278
1279 #if defined(_WIN32) || defined(WIN32)
1280 _setmode( _fileno( stdout ), 0x8000 );
1281 #endif /* defined(_WIN32) || defined(WIN32) */
1282
1283 have been added to the source of pcretest. This apparently does useful
1284 magic in relation to line terminators.
1285
1286 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1287 for the benefit of those environments where the "b" makes a difference.
1288
1289 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1290 to know about it. I have put a hack into configure.in that adds in code
1291 to set GCC=yes if CC=icc. This seems to end up at a point in the
1292 generated configure script that is early enough to affect the setting of
1293 compiler options, which is what is needed, but I have no means of testing
1294 whether it really works. (The user who reported this had patched the
1295 generated configure script, which of course I cannot do.)
1296
1297 LATER: After change 22 below (new libtool files), the configure script
1298 seems to know about icc (and also ecc). Therefore, I have commented out
1299 this hack in configure.in.
1300
1301 7. Added support for pkg-config (2 patches were sent in).
1302
1303 8. Negated POSIX character classes that used a combination of internal tables
1304 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1305 [[:^ascii]]. Typically, they would match almost any characters. The other
1306 POSIX classes were not broken in this way.
1307
1308 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1309 to find the match, as PCRE was deluded into thinking that the match had to
1310 start at the start point or following a newline. The same bug applied to
1311 patterns with negative forward assertions or any backward assertions
1312 preceding ".*" at the start, unless the pattern required a fixed first
1313 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1314
1315 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1316 starting at the last subject character, bytes beyond the end of the subject
1317 string were read.
1318
1319 11. Renamed the variable "class" as "classbits" to make life easier for C++
1320 users. (Previously there was a macro definition, but it apparently wasn't
1321 enough.)
1322
1323 12. Added the new field "tables" to the extra data so that tables can be passed
1324 in at exec time, or the internal tables can be re-selected. This allows
1325 a compiled regex to be saved and re-used at a later time by a different
1326 program that might have everything at different addresses.
1327
1328 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1329 -R library as well as a -L library.
1330
1331 14. The debugging options of pcretest (-d on the command line or D on a
1332 pattern) showed incorrect output for anything following an extended class
1333 that contained multibyte characters and which was followed by a quantifier.
1334
1335 15. Added optional support for general category Unicode character properties
1336 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1337 support. It adds about 90K to the size of the library. The meanings of the
1338 inbuilt class escapes such as \d and \s have NOT been changed.
1339
1340 16. Updated pcredemo.c to include calls to free() to release the memory for the
1341 compiled pattern.
1342
1343 17. The generated file chartables.c was being created in the source directory
1344 instead of in the building directory. This caused the build to fail if the
1345 source directory was different from the building directory, and was
1346 read-only.
1347
1348 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1349 file. No doubt somebody will tell me if they don't make sense... Also added
1350 Dan Mooney's comments about building on OpenVMS.
1351
1352 19. Added support for partial matching via the PCRE_PARTIAL option for
1353 pcre_exec() and the \P data escape in pcretest.
1354
1355 20. Extended pcretest with 3 new pattern features:
1356
1357 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1358 write the compiled pattern to the file whose name is "rest-of-line".
1359 This is a straight binary dump of the data, with the saved pointer to
1360 the character tables forced to be NULL. The study data, if any, is
1361 written too. After writing, pcretest reads a new pattern.
1362
1363 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1364 compiled pattern from the given file. There must not be any
1365 occurrences of "<" in the file name (pretty unlikely); if there are,
1366 pcretest will instead treat the initial "<" as a pattern delimiter.
1367 After reading in the pattern, pcretest goes on to read data lines as
1368 usual.
1369
1370 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1371 and 16-bit fields in a compiled pattern, to simulate a pattern that
1372 was compiled on a host of opposite endianness.
1373
1374 21. The pcre-exec() function can now cope with patterns that were compiled on
1375 hosts of opposite endianness, with this restriction:
1376
1377 As for any compiled expression that is saved and used later, the tables
1378 pointer field cannot be preserved; the extra_data field in the arguments
1379 to pcre_exec() should be used to pass in a tables address if a value
1380 other than the default internal tables were used at compile time.
1381
1382 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1383 now diagnosed as an error. Previously, most of the time, a negative number
1384 would have been treated as zero, but if in addition "ovector" was passed as
1385 NULL, a crash could occur.
1386
1387 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1388 new versions from the libtool 1.5 distribution (the last one is a copy of
1389 a file called libtool.m4). This seems to have fixed the need to patch
1390 "configure" to support Darwin 1.3 (which I used to do). However, I still
1391 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1392 workstation).
1393
1394 24. Changed the PCRE licence to be the more standard "BSD" licence.
1395
1396
1397 Version 4.5 01-Dec-03
1398 ---------------------
1399
1400 1. There has been some re-arrangement of the code for the match() function so
1401 that it can be compiled in a version that does not call itself recursively.
1402 Instead, it keeps those local variables that need separate instances for
1403 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1404 needs to "recurse". Keeping track of where control must go is done by means
1405 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1406 hide most of the details from the main code, and operates only if
1407 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1408 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1409 operating.
1410
1411 To make it easier for callers to provide specially tailored get/free
1412 functions for this usage, two new functions, pcre_stack_malloc, and
1413 pcre_stack_free, are used. They are always called in strict stacking order,
1414 and the size of block requested is always the same.
1415
1416 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1417 PCRE has been compiled to use the stack or the heap for recursion. The
1418 -C option of pcretest uses this to show which version is compiled.
1419
1420 A new data escape \S, is added to pcretest; it causes the amounts of store
1421 obtained and freed by both kinds of malloc/free at match time to be added
1422 to the output.
1423
1424 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1425 what's available on my current Linux desktop machine.
1426
1427 3. When matching a UTF-8 string, the test for a valid string at the start has
1428 been extended. If start_offset is not zero, PCRE now checks that it points
1429 to a byte that is the start of a UTF-8 character. If not, it returns
1430 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1431 this is necessary because there may be backward assertions in the pattern.
1432 When matching the same subject several times, it may save resources to use
1433 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1434
1435 4. The code for checking the validity of UTF-8 strings has been tightened so
1436 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1437 containing "overlong sequences".
1438
1439 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1440 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1441 should have been "&", but it just so happened that all the cases this let
1442 through by mistake were picked up later in the function.
1443
1444 6. I had used a variable called "isblank" - this is a C99 function, causing
1445 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1446
1447 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1448 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1449 so the version is shown; (c) stop "make test" echoing "./RunTest".
1450
1451 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1452
1453 9. The prototype for memmove() for systems that don't have it was using
1454 size_t, but the inclusion of the header that defines size_t was later. I've
1455 moved the #includes for the C headers earlier to avoid this.
1456
1457 10. Added some adjustments to the code to make it easier to compiler on certain
1458 special systems:
1459
1460 (a) Some "const" qualifiers were missing.
1461 (b) Added the macro EXPORT before all exported functions; by default this
1462 is defined to be empty.
1463 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1464 that it reads its output file name as an argument instead of writing
1465 to the standard output and assuming this can be redirected.
1466
1467 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1468 class containing characters with values greater than 255, PCRE compilation
1469 went into a loop.
1470
1471 12. A recursive reference to a subpattern that was within another subpattern
1472 that had a minimum quantifier of zero caused PCRE to crash. For example,
1473 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1474 recursion. If the recursively-called subpattern itself had a zero repeat,
1475 that was OK.
1476
1477 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1478 buffer into which it was copied (for escape processing) was still set at
1479 1024, so long lines caused crashes.
1480
1481 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1482 "internal error: code overflow...". This applied to any character class
1483 that was followed by a possessive quantifier.
1484
1485 15. Modified the Makefile to add libpcre.la as a prerequisite for
1486 libpcreposix.la because I was told this is needed for a parallel build to
1487 work.
1488
1489 16. If a pattern that contained .* following optional items at the start was
1490 studied, the wrong optimizing data was generated, leading to matching
1491 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1492 matching string must start with a or b or c. The correct conclusion for
1493 this pattern is that a match can start with any character.
1494
1495
1496 Version 4.4 13-Aug-03
1497 ---------------------
1498
1499 1. In UTF-8 mode, a character class containing characters with values between
1500 127 and 255 was not handled correctly if the compiled pattern was studied.
1501 In fixing this, I have also improved the studying algorithm for such
1502 classes (slightly).
1503
1504 2. Three internal functions had redundant arguments passed to them. Removal
1505 might give a very teeny performance improvement.
1506
1507 3. Documentation bug: the value of the capture_top field in a callout is *one
1508 more than* the number of the hightest numbered captured substring.
1509
1510 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1511 in incorrectly linking with a previously installed version. They now link
1512 explicitly with libpcre.la.
1513
1514 5. configure.in no longer needs to recognize Cygwin specially.
1515
1516 6. A problem in pcre.in for Windows platforms is fixed.
1517
1518 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1519 pcretest, it used to include the size of the study block as part of its
1520 output. Unfortunately, the structure contains a field that has a different
1521 size on different hardware architectures. This meant that the tests that
1522 showed this size failed. As the block is currently always of a fixed size,
1523 this information isn't actually particularly useful in pcretest output, so
1524 I have just removed it.
1525
1526 8. Three pre-processor statements accidentally did not start in column 1.
1527 Sadly, there are *still* compilers around that complain, even though
1528 standard C has not required this for well over a decade. Sigh.
1529
1530 9. In pcretest, the code for checking callouts passed small integers in the
1531 callout_data field, which is a void * field. However, some picky compilers
1532 complained about the casts involved for this on 64-bit systems. Now
1533 pcretest passes the address of the small integer instead, which should get
1534 rid of the warnings.
1535
1536 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1537 both compile and run time, and gives an error if an invalid UTF-8 sequence
1538 is found. There is a option for disabling this check in cases where the
1539 string is known to be correct and/or the maximum performance is wanted.
1540
1541 11. In response to a bug report, I changed one line in Makefile.in from
1542
1543 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1544 to
1545 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1546
1547 to look similar to other lines, but I have no way of telling whether this
1548 is the right thing to do, as I do not use Windows. No doubt I'll get told
1549 if it's wrong...
1550
1551
1552 Version 4.3 21-May-03
1553 ---------------------
1554
1555 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1556 Makefile.
1557
1558 2. Some refactoring to improve the quality of the code:
1559
1560 (i) The utf8_table... variables are now declared "const".
1561
1562 (ii) The code for \cx, which used the "case flipping" table to upper case
1563 lower case letters, now just substracts 32. This is ASCII-specific,
1564 but the whole concept of \cx is ASCII-specific, so it seems
1565 reasonable.
1566
1567 (iii) PCRE was using its character types table to recognize decimal and
1568 hexadecimal digits in the pattern. This is silly, because it handles
1569 only 0-9, a-f, and A-F, but the character types table is locale-
1570 specific, which means strange things might happen. A private
1571 table is now used for this - though it costs 256 bytes, a table is
1572 much faster than multiple explicit tests. Of course, the standard
1573 character types table is still used for matching digits in subject
1574 strings against \d.
1575
1576 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1577 ending in _t are). So I've renamed it as ESC_tee.
1578
1579 3. The first argument for regexec() in the POSIX wrapper should have been
1580 defined as "const".
1581
1582 4. Changed pcretest to use malloc() for its buffers so that they can be
1583 Electric Fenced for debugging.
1584
1585 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1586 to read one or more bytes before the start of the subject string. Often this
1587 had no effect on PCRE's behaviour, but in some circumstances it could
1588 provoke a segmentation fault.
1589
1590 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1591 to try to read one or more bytes before the start of the subject string.
1592
1593 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1594 UTF-8 support could misbehave in various ways if the subject string
1595 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1596 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1597 back over UTF-8 characters.)
1598
1599
1600 Version 4.2 14-Apr-03
1601 ---------------------
1602
1603 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1604
1605 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1606 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1607 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1608 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1609 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1610 and BUILD_EXEEXT
1611 Note: automatic setting of the BUILD variables is not yet working
1612 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1613 compile-time but not at link-time
1614 [LINK]: use for linking executables only
1615 make different versions for Windows and non-Windows
1616 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1617 libraries
1618 [LINK_FOR_BUILD]: new variable
1619 [OBJEXT]: use throughout
1620 [EXEEXT]: use throughout
1621 <winshared>: new target
1622 <wininstall>: new target
1623 <dftables.o>: use native compiler
1624 <dftables>: use native linker
1625 <install>: handle Windows platform correctly
1626 <clean>: ditto
1627 <check>: ditto
1628 copy DLL to top builddir before testing
1629
1630 As part of these changes, -no-undefined was removed again. This was reported
1631 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1632 in any case.
1633
1634 3. Some tidies to get rid of compiler warnings:
1635
1636 . In the match_data structure, match_limit was an unsigned long int, whereas
1637 match_call_count was an int. I've made them both unsigned long ints.
1638
1639 . In pcretest the fact that a const uschar * doesn't automatically cast to
1640 a void * provoked a warning.
1641
1642 . Turning on some more compiler warnings threw up some "shadow" variables
1643 and a few more missing casts.
1644
1645 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1646 option, a class that contained a single character with a value between 128
1647 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1648
1649 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1650 option, a class that contained several characters, but with at least one
1651 whose value was between 128 and 255 caused PCRE to crash.
1652
1653
1654 Version 4.1 12-Mar-03
1655 ---------------------
1656
1657 1. Compiling with gcc -pedantic found a couple of places where casts were
1658 needed, and a string in dftables.c that was longer than standard compilers are
1659 required to support.
1660
1661 2. Compiling with Sun's compiler found a few more places where the code could
1662 be tidied up in order to avoid warnings.
1663
1664 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1665 first of these names is deprecated in the latest Autoconf in favour of the name
1666 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1667 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1668 analogy I have changed it to CFLAGS_FOR_BUILD.
1669
1670 4. Added -no-undefined to the linking command in the Makefile, because this is
1671 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1672 linking step for the pcreposix library.
1673
1674 5. PCRE was failing to diagnose the case of two named groups with the same
1675 name.
1676
1677 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1678 literal character that is needed in the subject for a match, and scans along to
1679 ensure that it is present before embarking on the full matching process. This
1680 saves time in cases of nested unlimited repeats that are never going to match.
1681 Problem: the scan can take a lot of time if the subject is very long (e.g.
1682 megabytes), thus penalizing straightforward matches. It is now done only if the
1683 amount of subject to be scanned is less than 1000 bytes.
1684
1685 7. A lesser problem with the same optimization is that it was recording the
1686 first character of an anchored pattern as "needed", thus provoking a search
1687 right along the subject, even when the first match of the pattern was going to
1688 fail. The "needed" character is now not set for anchored patterns, unless it
1689 follows something in the pattern that is of non-fixed length. Thus, it still
1690 fulfils its original purpose of finding quick non-matches in cases of nested
1691 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1692
1693
1694 Version 4.0 17-Feb-03
1695 ---------------------
1696
1697 1. If a comment in an extended regex that started immediately after a meta-item
1698 extended to the end of string, PCRE compiled incorrect data. This could lead to
1699 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1700
1701 2. Moved to autoconf 2.53 and libtool 1.4.2.
1702
1703 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1704 the special perltest8 script is no longer needed - all the tests can be run
1705 from a single perltest script.
1706
1707 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1708 by \s. It has now been removed in PCRE. This means it isn't recognized as
1709 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1710 class [:space:] *does* include VT, thereby creating a mess.
1711
1712 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1713 space and tab.
1714
1715 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1716 its new features into the main test script, reducing the number of scripts.
1717
1718 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1719 were backward compatible, and made the (?i) apply to the whole pattern, as if
1720 /i were given. Now it behaves more logically, and applies the option setting
1721 only to what follows. PCRE has been changed to follow suit. However, if it
1722 finds options settings right at the start of the pattern, it extracts them into
1723 the global options, as before. Thus, they show up in the info data.
1724
1725 8. Added support for the \Q...\E escape sequence. Characters in between are
1726 treated as literals. This is slightly different from Perl in that $ and @ are
1727 also handled as literals inside the quotes. In Perl, they will cause variable
1728 interpolation. Note the following examples:
1729
1730 Pattern PCRE matches Perl matches
1731
1732 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1733 \Qabc\$xyz\E abc\$xyz abc\$xyz
1734 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1735
1736 For compatibility with Perl, \Q...\E sequences are recognized inside character
1737 classes as well as outside them.
1738
1739 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1740 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1741 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1742 signed/unsigned warnings.
1743
1744 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1745 option for pcretest, so I've replaced it by a simple function that does just
1746 that job.
1747
1748 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1749 "pcregrep -".
1750
1751 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1752 Java package. This provides some syntactic sugar for simple cases of what my
1753 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1754 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1755 item, you can use this simplified notation. Note that only makes sense with
1756 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1757 greediness, whatever the setting of the PCRE_UNGREEDY option.
1758
1759 13. A change of greediness default within a pattern was not taking effect at
1760 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1761 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1762 was abstracted outside.
1763
1764 14. PCRE now supports the \G assertion. It is true when the current matching
1765 position is at the start point of the match. This differs from \A when the
1766 starting offset is non-zero. Used with the /g option of pcretest (or similar
1767 code), it works in the same way as it does for Perl's /g option. If all
1768 alternatives of a regex begin with \G, the expression is anchored to the start
1769 match position, and the "anchored" flag is set in the compiled expression.
1770
1771 15. Some bugs concerning the handling of certain option changes within patterns
1772 have been fixed. These applied to options other than (?ims). For example,
1773 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1774 been the other way round. Some of this was related to change 7 above.
1775
1776 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1777 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1778 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1779 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1780
1781 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1782 mode. Unlike ".", it always matches newline, whatever the setting of
1783 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1784 assertions. Perl allows it, but it doesn't (in general) work because it can't
1785 calculate the length of the lookbehind. At least, that's the case for Perl
1786 5.8.0 - I've been told they are going to document that it doesn't work in
1787 future.
1788
1789 18. Added an error diagnosis for escapes that PCRE does not support: these are
1790 \L, \l, \N, \P, \p, \U, \u, and \X.
1791
1792 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1793 reading past the end of the pattern in cases such as /[abcd/.
1794
1795 20. PCRE was getting more memory than necessary for patterns with classes that
1796 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1797
1798 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1799 compiling PCRE for use with Virtual Pascal.
1800
1801 22. Small fix to the Makefile to make it work properly if the build is done
1802 outside the source tree.
1803
1804 23. Added a new extension: a condition to go with recursion. If a conditional
1805 subpattern starts with (?(R) the "true" branch is used if recursion has
1806 happened, whereas the "false" branch is used only at the top level.
1807
1808 24. When there was a very long string of literal characters (over 255 bytes
1809 without UTF support, over 250 bytes with UTF support), the computation of how
1810 much memory was required could be incorrect, leading to segfaults or other
1811 strange effects.
1812
1813 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1814 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1815 there was a subsequent back reference to those brackets. This meant that, for
1816 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1817 possible to check for precisely this case. All we can do is abandon the
1818 optimization if .* occurs inside capturing brackets when there are any back
1819 references whatsoever. (See below for a better fix that came later.)
1820
1821 26. The handling of the optimization for finding the first character of a
1822 non-anchored pattern, and for finding a character that is required later in the
1823 match were failing in some cases. This didn't break the matching; it just
1824 failed to optimize when it could. The way this is done has been re-implemented.
1825
1826 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1827
1828 28. Added a new feature that provides some of the functionality that Perl
1829 provides with (?{...}). The facility is termed a "callout". The way it is done
1830 in PCRE is for the caller to provide an optional function, by setting
1831 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1832 global variable. By default it is unset, which disables all calling out. To get
1833 the function called, the regex must include (?C) at appropriate points. This
1834 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1835 This provides a means of identifying different callout points. When PCRE
1836 reaches such a point in the regex, if pcre_callout has been set, the external
1837 function is called. It is provided with data in a structure called
1838 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1839 matching continues; if it returns a non-zero value, the match at the current
1840 point fails. However, backtracking will occur if possible. [This was changed
1841 later and other features added - see item 49 below.]
1842
1843 29. pcretest is upgraded to test the callout functionality. It provides a
1844 callout function that displays information. By default, it shows the start of
1845 the match and the current position in the text. There are some new data escapes
1846 to vary what happens:
1847
1848 \C+ in addition, show current contents of captured substrings
1849 \C- do not supply a callout function
1850 \C!n return 1 when callout number n is reached
1851 \C!n!m return 1 when callout number n is reached for the mth time
1852
1853 30. If pcregrep was called with the -l option and just a single file name, it
1854 output "<stdin>" if a match was found, instead of the file name.
1855
1856 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1857 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1858 pcre_exec(). This saves a malloc/free per call. The default value of
1859 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1860 when configuring.
1861
1862 32. The default maximum size of a compiled pattern is 64K. There have been a
1863 few cases of people hitting this limit. The code now uses macros to handle the
1864 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1865 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1866 configuring. Tests 2 and 5 work only with 2-byte links because they output
1867 debugging information about compiled patterns.
1868
1869 33. Internal code re-arrangements:
1870
1871 (a) Moved the debugging function for printing out a compiled regex into
1872 its own source file (printint.c) and used #include to pull it into
1873 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1874 separate copies.
1875
1876 (b) Defined the list of op-code names for debugging as a macro in
1877 internal.h so that it is next to the definition of the opcodes.
1878
1879 (c) Defined a table of op-code lengths for simpler skipping along compiled
1880 code. This is again a macro in internal.h so that it is next to the
1881 definition of the opcodes.
1882
1883 34. Added support for recursive calls to individual subpatterns, along the
1884 lines of Robin Houston's patch (but implemented somewhat differently).
1885
1886 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1887 allow it to read and process whole directories in Win32. This code was
1888 contributed by Lionel Fourquaux; it has not been tested by me.
1889
1890 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1891 used to name a group. Names consist of alphanumerics and underscores, and must
1892 be unique. Back references use the syntax (?P=name) and recursive calls use
1893 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1894 numbers. The function pcre_fullinfo() can be used after compilation to extract
1895 a name/number map. There are three relevant calls:
1896
1897 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1898 PCRE_INFO_NAMECOUNT yields the number of entries
1899 PCRE_INFO_NAMETABLE yields a pointer to the map.
1900
1901 The map is a vector of fixed-size entries. The size of each entry depends on
1902 the length of the longest name used. The first two bytes of each entry are the
1903 group number, most significant byte first. There follows the corresponding
1904 name, zero terminated. The names are in alphabetical order.
1905
1906 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1907 case instead of 255. Making it the same both with and without UTF-8 support
1908 means that the same test output works with both.
1909
1910 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1911 calling malloc() with a zero argument.
1912
1913 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1914 optimization. I've improved things by keeping a bitmap of backreferences with
1915 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1916 fact referenced, the optimization can be applied. It is unlikely that a
1917 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1918 the match to follow \n) will appear inside brackets with a number greater than
1919 31, but if it does, any back reference > 31 suppresses the optimization.
1920
1921 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1922 of disabling numbered capturing parentheses. Any opening parenthesis that is
1923 not followed by ? behaves as if it were followed by ?: but named parentheses
1924 can still be used for capturing (and they will acquire numbers in the usual
1925 way).
1926
1927 41. Redesigned the return codes from the match() function into yes/no/error so
1928 that errors can be passed back from deep inside the nested calls. A malloc
1929 failure while inside a recursive subpattern call now causes the
1930 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1931
1932 42. It is now possible to set a limit on the number of times the match()
1933 function is called in a call to pcre_exec(). This facility makes it possible to
1934 limit the amount of recursion and backtracking, though not in a directly
1935 obvious way, because the match() function is used in a number of different
1936 circumstances. The count starts from zero for each position in the subject
1937 string (for non-anchored patterns). The default limit is, for compatibility, a
1938 large number, namely 10 000 000. You can change this in two ways:
1939
1940 (a) When configuring PCRE before making, you can use --with-match-limit=n
1941 to set a default value for the compiled library.
1942
1943 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1944 a different value is set. See 45 below.
1945
1946 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1947
1948 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1949 of things that can be changed at compile time. The first argument specifies
1950 what is wanted and the second points to where the information is to be placed.
1951 The current list of available information is:
1952
1953 PCRE_CONFIG_UTF8
1954
1955 The output is an integer that is set to one if UTF-8 support is available;
1956 otherwise it is set to zero.
1957
1958 PCRE_CONFIG_NEWLINE
1959
1960 The output is an integer that it set to the value of the code that is used for
1961 newline. It is either LF (10) or CR (13).
1962
1963 PCRE_CONFIG_LINK_SIZE
1964
1965 The output is an integer that contains the number of bytes used for internal
1966 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1967
1968 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1969
1970 The output is an integer that contains the threshold above which the POSIX
1971 interface uses malloc() for output vectors. See item 31 above.
1972
1973 PCRE_CONFIG_MATCH_LIMIT
1974
1975 The output is an unsigned integer that contains the default limit of the number
1976 of match() calls in a pcre_exec() execution. See 42 above.
1977
1978 44. pcretest has been upgraded by the addition of the -C option. This causes it
1979 to extract all the available output from the new pcre_config() function, and to
1980 output it. The program then exits immediately.
1981
1982 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1983 order to support additional features. One way would have been to define
1984 pcre_exec2() (for example) with extra arguments, but this would not have been
1985 extensible, and would also have required all calls to the original function to
1986 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1987 is used for passing in "extra" data from pcre_study().
1988
1989 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1990 contains the following fields:
1991
1992 flags a bitmap indicating which of the following fields are set
1993 study_data opaque data from pcre_study()
1994 match_limit a way of specifying a limit on match() calls for a specific
1995 call to pcre_exec()
1996 callout_data data for callouts (see 49 below)
1997
1998 The flag bits are also defined in pcre.h, and are
1999
2000 PCRE_EXTRA_STUDY_DATA
2001 PCRE_EXTRA_MATCH_LIMIT
2002 PCRE_EXTRA_CALLOUT_DATA
2003
2004 The pcre_study() function now returns one of these new pcre_extra blocks, with
2005 the actual study data pointed to by the study_data field, and the
2006 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2007 before. That is, this change is entirely upwards-compatible and requires no
2008 change to existing code.
2009
2010 If you want to pass in additional data to pcre_exec(), you can either place it
2011 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2012 block.
2013
2014 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2015 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2016 times with different match limits, until it finds the minimum value needed for
2017 pcre_exec() to complete. The value is then output. This can be instructive; for
2018 most simple matches the number is quite small, but for pathological cases it
2019 gets very large very quickly.
2020
2021 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2022 returns the size of the data block pointed to by the study_data field in a
2023 pcre_extra block, that is, the value that was passed as the argument to
2024 pcre_malloc() when PCRE was getting memory in which to place the information
2025 created by pcre_study(). The fourth argument should point to a size_t variable.
2026 pcretest has been extended so that this information is shown after a successful
2027 pcre_study() call when information about the compiled regex is being displayed.
2028
2029 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2030 because what follows is always an absolute path. (Later: it turns out that this
2031 is more than cosmetic for MinGW, because it doesn't like empty path
2032 components.)
2033
2034 49. Some changes have been made to the callout feature (see 28 above):
2035
2036 (i) A callout function now has three choices for what it returns:
2037
2038 0 => success, carry on matching
2039 > 0 => failure at this point, but backtrack if possible
2040 < 0 => serious error, return this value from pcre_exec()
2041
2042 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2043 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2044 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2045 use by callout functions. It will never be used by PCRE itself.
2046
2047 (ii) The pcre_extra structure (see 45 above) has a void * field called
2048 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2049 pcre_callout_block structure has a field of the same name. The contents of
2050 the field passed in the pcre_extra structure are passed to the callout
2051 function in the corresponding field in the callout block. This makes it
2052 easier to use the same callout-containing regex from multiple threads. For
2053 testing, the pcretest program has a new data escape
2054
2055 \C*n pass the number n (may be negative) as callout_data
2056
2057 If the callout function in pcretest receives a non-zero value as
2058 callout_data, it returns that value.
2059
2060 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2061 there were some redundant $(CFLAGS) in commands that are now specified as
2062 $(LINK), which already includes $(CFLAGS).
2063
2064 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2065 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2066 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2067 one-byte characters throughout. Note that case-insensitive matching applies
2068 only to characters whose values are less than 256. PCRE doesn't support the
2069 notion of cases for higher-valued characters.
2070
2071 (i) A character class whose characters are all within 0-255 is handled as
2072 a bit map, and the map is inverted for negative classes. Previously, a
2073 character > 255 always failed to match such a class; however it should
2074 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2075
2076 (ii) A negated character class with a single character < 255 is coded as
2077 "not this character" (OP_NOT). This wasn't working properly when the test
2078 character was multibyte, either singly or repeated.
2079
2080 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2081 mode, for example: \x{100}{2,3}.
2082
2083 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2084 singly or repeated) now correctly test multibyte characters. However,
2085 PCRE doesn't recognize any characters with values greater than 255 as
2086 digits, spaces, or word characters. Such characters always match \D, \S,
2087 and \W, and never match \d, \s, or \w.
2088
2089 (v) Classes may now contain characters and character ranges with values
2090 greater than 255. For example: [ab\x{100}-\x{400}].
2091
2092 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2093 PCRE in UTF-8 mode.
2094
2095 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2096 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2097 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2098 value.)
2099
2100 53. The single man page has become too large. I have therefore split it up into
2101 a number of separate man pages. These also give rise to individual HTML pages;
2102 these are now put in a separate directory, and there is an index.html page that
2103 lists them all. Some hyperlinking between the pages has been installed.
2104
2105 54. Added convenience functions for handling named capturing parentheses.
2106
2107 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2108 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2109 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2110 are faulted.
2111
2112 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2113 calling configure. These values are used when compiling the dftables.c program
2114 which is run to generate the source of the default character tables. They
2115 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2116 you will need to set these values.
2117
2118 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2119
2120
2121 Version 3.9 02-Jan-02
2122 ---------------------
2123
2124 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2125
2126 2. If --disable-static was given, the building process failed when trying to
2127 build pcretest and pcregrep. (For some reason it was using libtool to compile
2128 them, which is not right, as they aren't part of the library.)
2129
2130
2131 Version 3.8 18-Dec-01
2132 ---------------------
2133
2134 1. The experimental UTF-8 code was completely screwed up. It was packing the
2135 bytes in the wrong order. How dumb can you get?
2136
2137
2138 Version 3.7 29-Oct-01
2139 ---------------------
2140
2141 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2142 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2143 this didn't happen under Solaris 8, where I normally test things.
2144
2145 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2146 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2147 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2148 this fix an hour or so after the initial 3.7 release.)
2149
2150
2151 Version 3.6 23-Oct-01
2152 ---------------------
2153
2154 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2155 offsets passed as NULL with zero offset count.
2156
2157 2. The config.guess and config.sub files had not been updated when I moved to
2158 the latest autoconf.
2159
2160
2161 Version 3.5 15-Aug-01
2162 ---------------------
2163
2164 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2165 had been forgotten.
2166
2167 2. By using declared but undefined structures, we can avoid using "void"
2168 definitions in pcre.h while keeping the internal definitions of the structures
2169 private.
2170
2171 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2172 user point of view, this means that both static and shared libraries are built
2173 by default, but this can be individually controlled. More of the work of
2174 handling this static/shared cases is now inside libtool instead of PCRE's make
2175 file.
2176
2177 4. The pcretest utility is now installed along with pcregrep because it is
2178 useful for users (to test regexs) and by doing this, it automatically gets
2179 relinked by libtool. The documentation has been turned into a man page, so
2180 there are now .1, .txt, and .html versions in /doc.
2181
2182 5. Upgrades to pcregrep:
2183 (i) Added long-form option names like gnu grep.
2184 (ii) Added --help to list all options with an explanatory phrase.
2185 (iii) Added -r, --recursive to recurse into sub-directories.
2186 (iv) Added -f, --file to read patterns from a file.
2187
2188 6. pcre_exec() was referring to its "code" argument before testing that
2189 argument for NULL (and giving an error if it was NULL).
2190
2191 7. Upgraded Makefile.in to allow for compiling in a different directory from
2192 the source directory.
2193
2194 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2195 options bits, the pointer it was passed was to an int instead of to an unsigned
2196 long int. This mattered only on 64-bit systems.
2197
2198 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2199 generated) instead of pcre.in, which it its source. Also made the same change
2200 in several of the .c files.
2201
2202 10. A new release of gcc defines printf() as a macro, which broke pcretest
2203 because it had an ifdef in the middle of a string argument for printf(). Fixed
2204 by using separate calls to printf().
2205
2206 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2207 script, to force use of CR or LF instead of \n in the source. On non-Unix
2208 systems, the value can be set in config.h.
2209
2210 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2211 absolute limit. Changed the text of the error message to make this clear, and
2212 likewise updated the man page.
2213
2214 13. The limit of 99 on the number of capturing subpatterns has been removed.
2215 The new limit is 65535, which I hope will not be a "real" limit.
2216
2217
2218 Version 3.4 22-Aug-00
2219 ---------------------
2220
2221 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2222
2223 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2224
2225
2226 Version 3.3 01-Aug-00
2227 ---------------------
2228
2229 1. If an octal character was given, but the value was greater than \377, it
2230 was not getting masked to the least significant bits, as documented. This could
2231 lead to crashes in some systems.
2232
2233 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2234 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2235
2236 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2237 These just pass their arguments on to (pcre_free)(), but they are provided
2238 because some uses of PCRE bind it to non-C systems that can call its functions,
2239 but cannot call free() or pcre_free() directly.
2240
2241 4. Add "make test" as a synonym for "make check". Corrected some comments in
2242 the Makefile.
2243
2244 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2245 Makefile.
2246
2247 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2248 command called pgrep for grepping around the active processes.
2249
2250 7. Added the beginnings of support for UTF-8 character strings.
2251
2252 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2253 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2254 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2255 out for the ar command.)
2256
2257
2258 Version 3.2 12-May-00
2259 ---------------------
2260
2261 This is purely a bug fixing release.
2262
2263 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2264 of ZA. This was just one example of several cases that could provoke this bug,
2265 which was introduced by change 9 of version 2.00. The code for breaking
2266 infinite loops after an iteration that matches an empty string was't working
2267 correctly.
2268
2269 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2270 when matched against abbab (for example). After matching an empty string, it
2271 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2272 caused it to match further down the string than it should.
2273
2274 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2275 was there because it doesn't seem to be needed, and it causes trouble on some
2276 systems, as it is not a Standard C header. It has been removed.
2277
2278 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2279 were reported on the Macintosh. The changes were from
2280
2281 while ((c = *(++ptr)) != 0 && c != '\n');
2282 to
2283 while ((c = *(++ptr)) != 0 && c != '\n') ;
2284
2285 Totally extraordinary, but if that's what it takes...
2286
2287 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2288 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2289 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2290 assumes the way PCRE uses memmove() (always moving upwards).
2291
2292 6. PCRE is being used in one environment where strchr() is not available. There
2293 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2294 faster code anyway.
2295
2296
2297 Version 3.1 09-Feb-00
2298 ---------------------
2299
2300 The only change in this release is the fixing of some bugs in Makefile.in for
2301 the "install" target:
2302
2303 (1) It was failing to install pcreposix.h.
2304
2305 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2306
2307
2308 Version 3.0 01-Feb-00
2309 ---------------------
2310
2311 1. Add support for the /+ modifier to perltest (to output $` like it does in
2312 pcretest).
2313
2314 2. Add support for the /g modifier to perltest.
2315
2316 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2317 matches null strings.
2318
2319 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2320 pattern. Perl treats empty patterns specially - it reuses the most recent
2321 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2322 effect.
2323
2324 5. The POSIX interface was broken in that it was just handing over the POSIX
2325 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2326 required a bigger vector, with some working space on the end. This means that
2327 the POSIX wrapper now has to get and free some memory, and copy the results.
2328
2329 6. Added some simple autoconf support, placing the test data and the
2330 documentation in separate directories, re-organizing some of the
2331 information files, and making it build pcre-config (a GNU standard). Also added
2332 libtool support for building PCRE as a shared library, which is now the
2333 default.
2334
2335 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2336 09 are not valid octal constants. Single digits will be used for minor values
2337 less than 10.
2338
2339 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2340 existing programs that set these in the POSIX interface can use PCRE without
2341 modification.
2342
2343 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2344 return all that pcre_info() returns, plus additional data. The pcre_info()
2345 function is retained for compatibility, but is considered to be obsolete.
2346
2347 10. Added experimental recursion feature (?R) to handle one common case that
2348 Perl 5.6 will be able to do with (?p{...}).
2349
2350 11. Added support for POSIX character classes like [:alpha:], which Perl is
2351 adopting.
2352
2353
2354 Version 2.08 31-Aug-99
2355 ----------------------
2356
2357 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2358 trying to match at the startoffset position, but instead was moving forward to
2359 the next newline as if a previous match had failed.
2360
2361 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2362 and could get into a loop if a null string was matched other than at the start
2363 of the subject.
2364
2365 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2366 be distinguished at compile time, and for completeness also added PCRE_DATE.
2367
2368 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2369 in GnuWin32 environments.
2370
2371
2372 Version 2.07 29-Jul-99
2373 ----------------------
2374
2375 1. The documentation is now supplied in plain text form and HTML as well as in
2376 the form of man page sources.
2377
2378 2. C++ compilers don't like assigning (void *) values to other pointer types.
2379 In particular this affects malloc(). Although there is no problem in Standard
2380 C, I've put in casts to keep C++ compilers happy.
2381
2382 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2383 should be (const char *).
2384
2385 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2386 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2387 However, I haven't made this a standard facility. The documentation doesn't
2388 mention it, and the Makefile doesn't support it.
2389
2390 5. The Makefile now contains an "install" target, with editable destinations at
2391 the top of the file. The pcretest program is not installed.
2392
2393 6. pgrep -V now gives the PCRE version number and date.
2394
2395 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2396 causing the entire string to be ignored, instead of just the last character.
2397
2398 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2399 non-matching string, it can take a very, very long time, even for strings of
2400 quite modest length, because of the nested recursion. PCRE now does better in
2401 some of these cases. It does this by remembering the last required literal
2402 character in the pattern, and pre-searching the subject to ensure it is present
2403 before running the real match. In other words, it applies a heuristic to detect
2404 some types of certain failure quickly, and in the above example, if presented
2405 with a string that has no trailing " it gives "no match" very quickly.
2406
2407 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2408 other alternatives are tried instead.
2409
2410
2411 Version 2.06 09-Jun-99
2412 ----------------------
2413
2414 1. Change pcretest's output for amount of store used to show just the code
2415 space, because the remainder (the data block) varies in size between 32-bit and
2416 64-bit systems.
2417
2418 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2419 start matching at. This allows lookbehinds to work when searching for multiple
2420 occurrences in a string.
2421
2422 3. Added additional options to pcretest for testing multiple occurrences:
2423
2424 /+ outputs the rest of the string that follows a match
2425 /g loops for multiple occurrences, using the new startoffset argument
2426 /G loops for multiple occurrences by passing an incremented pointer
2427
2428 4. PCRE wasn't doing the "first character" optimization for patterns starting
2429 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2430 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2431 the letter 'x'. On long subject strings, this gives a significant speed-up.
2432
2433
2434 Version 2.05 21-Apr-99
2435 ----------------------
2436
2437 1. Changed the type of magic_number from int to long int so that it works
2438 properly on 16-bit systems.
2439
2440 2. Fixed a bug which caused patterns starting with .* not to work correctly
2441 when the subject string contained newline characters. PCRE was assuming
2442 anchoring for such patterns in all cases, which is not correct because .* will
2443 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2444 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2445 must be retried after every newline in the subject.
2446
2447
2448 Version 2.04 18-Feb-99
2449 ----------------------
2450
2451 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2452 computation of the store needed to hold the pattern was incorrect (too large).
2453 If such patterns were nested a few deep, this could multiply and become a real
2454 problem.
2455
2456 2. Added /M option to pcretest to show the memory requirement of a specific
2457 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2458
2459 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2460 compiled in such a way that the backtracking after subsequent failure was
2461 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2462 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2463
2464
2465 Version 2.03 02-Feb-99
2466 ----------------------
2467
2468 1. Fixed typo and small mistake in man page.
2469
2470 2. Added 4th condition (GPL supersedes if conflict) and created separate
2471 LICENCE file containing the conditions.
2472
2473 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2474 Perl, that is the internal \ allows the delimiter to be included in the
2475 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2476 the final delimiter, add \ to the end of the pattern (to test the error).
2477
2478 4. Added the convenience functions for extracting substrings after a successful
2479 match. Updated pcretest to make it able to test these functions.
2480
2481
2482 Version 2.02 14-Jan-99
2483 ----------------------
2484
2485 1. Initialized the working variables associated with each extraction so that
2486 their saving and restoring doesn't refer to uninitialized store.
2487
2488 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2489 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2490 fix the problem.
2491
2492 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2493 calls, and wasn't printing the correct value for compiling calls. Increased the
2494 default value of LOOPREPEAT, and the number of significant figures in the
2495 times.
2496
2497 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2498
2499 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2500 a building problem on Windows NT with a FAT file system.
2501
2502
2503 Version 2.01 21-Oct-98
2504 ----------------------
2505
2506 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2507 to character tables built by pcre_maketables() in the current locale. If NULL
2508 is passed, the default tables are used.
2509
2510
2511 Version 2.00 24-Sep-98
2512 ----------------------
2513
2514 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2515 it any more.
2516
2517 2. Allow quantification of (?>) groups, and make it work correctly.
2518
2519 3. The first character computation wasn't working for (?>) groups.
2520
2521 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2522 end of the subject) and add 5.005's \z, which really does match only at the
2523 very end of the subject.
2524
2525 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2526
2527 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2528 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2529 localized options. All options to pcre_study() were also removed.
2530
2531 7. Add other new features from 5.005:
2532
2533 $(?<= positive lookbehind
2534 $(?<! negative lookbehind
2535 (?imsx-imsx) added the unsetting capability
2536 such a setting is global if at outer level; local otherwise
2537 (?imsx-imsx:) non-capturing groups with option setting
2538 (?(cond)re|re) conditional pattern matching
2539
2540 A backreference to itself in a repeated group matches the previous
2541 captured string.
2542
2543 8. General tidying up of studying (both automatic and via "study")
2544 consequential on the addition of new assertions.
2545
2546 9. As in 5.005, unlimited repeated groups that could match an empty substring
2547 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2548 runtime if any iteration does actually match an empty substring.
2549
2550 10. Include the RunTest script in the distribution.
2551
2552 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2553 discrepancies, some of which were old and were also with respect to 5.004. They
2554 have now been fixed.
2555
2556
2557 Version 1.09 28-Apr-98
2558 ----------------------
2559
2560 1. A negated single character class followed by a quantifier with a minimum
2561 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2562 program crashes, or just wrong answers. This did not apply to negated classes
2563 containing more than one character, or to minima other than one.
2564
2565
2566 Version 1.08 27-Mar-98
2567 ----------------------
2568
2569 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2570
2571 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2572 latter must appear before anything that relies on it in the pattern.
2573
2574
2575 Version 1.07 16-Feb-98
2576 ----------------------
2577
2578 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2579 repeat of a potentially empty string).
2580
2581
2582 Version 1.06 23-Jan-98
2583 ----------------------
2584
2585 1. Added Markus Oberhumer's little patches for C++.
2586
2587 2. Literal strings longer than 255 characters were broken.
2588
2589
2590 Version 1.05 23-Dec-97
2591 ----------------------
2592
2593 1. Negated character classes containing more than one character were failing if
2594 PCRE_CASELESS was set at run time.
2595
2596
2597 Version 1.04 19-Dec-97
2598 ----------------------
2599
2600 1. Corrected the man page, where some "const" qualifiers had been omitted.
2601
2602 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2603 input syntax.
2604
2605 3. Fixed memory leak which occurred when a regex with back references was
2606 matched with an offsets vector that wasn't big enough. The temporary memory
2607 that is used in this case wasn't being freed if the match failed.
2608
2609 4. Tidied pcretest to ensure it frees memory that it gets.
2610
2611 5. Temporary memory was being obtained in the case where the passed offsets
2612 vector was exactly big enough.
2613
2614 6. Corrected definition of offsetof() from change 5 below.
2615
2616 7. I had screwed up change 6 below and broken the rules for the use of
2617 setjmp(). Now fixed.
2618
2619
2620 Version 1.03 18-Dec-97
2621 ----------------------
2622
2623 1. A erroneous regex with a missing opening parenthesis was correctly
2624 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2625 on some systems.
2626
2627 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2628 it was reported that one broken compiler failed on the former because "code" is
2629 also an independent variable.
2630
2631 3. The erroneous regex a[]b caused an array overrun reference.
2632
2633 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2634 fail on data ending with that character. (It was going on too far, and checking
2635 the next character, typically a binary zero.) This was specific to the
2636 optimized code for single-character negative classes.
2637
2638 5. Added a contributed patch from the TIN world which does the following:
2639
2640 + Add an undef for memmove, in case the the system defines a macro for it.
2641
2642 + Add a definition of offsetof(), in case there isn't one. (I don't know
2643 the reason behind this - offsetof() is part of the ANSI standard - but
2644 it does no harm).
2645
2646 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2647 most of the places where whitespace preceded '#'. I have given up and
2648 allowed the remaining 2 cases to be at the margin.
2649
2650 + Rename some variables in pcre to eliminate shadowing. This seems very
2651 pedantic, but does no harm, of course.
2652
2653 6. Moved the call to setjmp() into its own function, to get rid of warnings
2654 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2655
2656 7. Constructs such as \d{8,} were compiling into the equivalent of
2657 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2658 outcome, but in this particular case used more store than had been allocated,
2659 which caused the bug to be discovered because it threw up an internal error.
2660
2661 8. The debugging code in both pcre and pcretest for outputting the compiled
2662 form of a regex was going wrong in the case of back references followed by
2663 curly-bracketed repeats.
2664
2665
2666 Version 1.02 12-Dec-97
2667 ----------------------
2668
2669 1. Typos in pcre.3 and comments in the source fixed.
2670
2671 2. Applied a contributed patch to get rid of places where it used to remove
2672 'const' from variables, and fixed some signed/unsigned and uninitialized
2673 variable warnings.
2674
2675 3. Added the "runtest" target to Makefile.
2676
2677 4. Set default compiler flag to -O2 rather than just -O.
2678
2679
2680 Version 1.01 19-Nov-97
2681 ----------------------
2682
2683 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2684 like /([ab]*)*/, that is, for classes with more than one character in them.
2685
2686 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2687 as /((?>a*))*/ (a PCRE_EXTRA facility).
2688
2689
2690 Version 1.00 18-Nov-97
2691 ----------------------
2692
2693 1. Added compile-time macros to support systems such as SunOS4 which don't have
2694 memmove() or strerror() but have other things that can be used instead.
2695
2696 2. Arranged that "make clean" removes the executables.
2697
2698
2699 Version 0.99 27-Oct-97
2700 ----------------------
2701
2702 1. Fixed bug in code for optimizing classes with only one character. It was
2703 initializing a 32-byte map regardless, which could cause it to run off the end
2704 of the memory it had got.
2705
2706 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2707
2708
2709 Version 0.98 22-Oct-97
2710 ----------------------
2711
2712 1. Fixed bug in code for handling temporary memory usage when there are more
2713 back references than supplied space in the ovector. This could cause segfaults.
2714
2715
2716 Version 0.97 21-Oct-97
2717 ----------------------
2718
2719 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2720
2721 2. Optimized negated single characters not to use a bit map.
2722
2723 3. Brought error texts together as macro definitions; clarified some of them;
2724 fixed one that was wrong - it said "range out of order" when it meant "invalid
2725 escape sequence".
2726
2727 4. Changed some char * arguments to const char *.
2728
2729 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2730
2731 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2732 pcretest.
2733
2734
2735 Version 0.96 16-Oct-97
2736 ----------------------
2737
2738 1. Added a simple "pgrep" utility to the distribution.
2739
2740 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2741 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2742 where "ddd" means "one or more decimal digits".
2743
2744 3. Fixed serious bug. If a pattern had a back reference, but the call to
2745 pcre_exec() didn't supply a large enough ovector to record the related
2746 identifying subpattern, the match always failed. PCRE now remembers the number
2747 of the largest back reference, and gets some temporary memory in which to save
2748 the offsets during matching if necessary, in order to ensure that
2749 backreferences always work.
2750
2751 4. Increased the compatibility with Perl in a number of ways:
2752
2753 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2754 to request this handling. The option can be set at compile or exec time.
2755
2756 (b) $ matches before a terminating newline by default; an option
2757 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2758 mode). The option can be set at compile or exec time.
2759
2760 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2761 the same as Perl's. If the decimal number it represents is less than 10
2762 or there aren't that many previous left capturing parentheses, an octal
2763 escape is read. Inside a character class, it's always an octal escape,
2764 even if it is a single digit.
2765
2766 (d) An escaped but undefined alphabetic character is taken as a literal,
2767 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2768 escapes.
2769
2770 (e) {0} is now permitted. (The previous item is removed from the compiled
2771 pattern).
2772
2773 5. Changed all the names of code files so that the basic parts are no longer
2774 than 10 characters, and abolished the teeny "globals.c" file.
2775
2776 6. Changed the handling of character classes; they are now done with a 32-byte
2777 bit map always.
2778
2779 7. Added the -d and /D options to pcretest to make it possible to look at the
2780 internals of compilation without having to recompile pcre.
2781
2782
2783 Version 0.95 23-Sep-97
2784 ----------------------
2785
2786 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2787 \x20 at the start of a run of normal characters. These were being treated as
2788 real characters, instead of the source characters being re-checked.
2789
2790
2791 Version 0.94 18-Sep-97
2792 ----------------------
2793
2794 1. The functions are now thread-safe, with the caveat that the global variables
2795 containing pointers to malloc() and free() or alternative functions are the
2796 same for all threads.
2797
2798 2. Get pcre_study() to generate a bitmap of initial characters for non-
2799 anchored patterns when this is possible, and use it if passed to pcre_exec().
2800
2801
2802 Version 0.93 15-Sep-97
2803 ----------------------
2804
2805 1. /(b)|(:+)/ was computing an incorrect first character.
2806
2807 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2808 but not actually doing anything yet.
2809
2810 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2811 as Perl does (e.g. [-az] or [az-]).
2812
2813 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2814 all possible positions.
2815
2816 5. Split up into different modules to avoid including unneeded functions in a
2817 compiled binary. However, compile and exec are still in one module. The "study"
2818 function is split off.
2819
2820 6. The character tables are now in a separate module whose source is generated
2821 by an auxiliary program - but can then be edited by hand if required. There are
2822 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2823 toupper() in the code.
2824
2825 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2826 make them global. Abolish the function for setting them, as the caller can now
2827 set them directly.
2828
2829
2830 Version 0.92 11-Sep-97
2831 ----------------------
2832
2833 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2834 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2835
2836 2. Caseless matching was not working in character classes if the characters in
2837 the pattern were in upper case.
2838
2839 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2840
2841 4. Make PCRE_ANCHORED public and accept as a compile option.
2842
2843 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2844 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2845 pass them.
2846
2847 6. Give an error if bad option bits passed at compile or run time.
2848
2849 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2850 pcretest to cause it to pass that flag.
2851
2852 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2853 options, and the first character, if set.
2854
2855 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2856
2857
2858 Version 0.91 10-Sep-97
2859 ----------------------
2860
2861 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2862 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2863
2864 2. PCRE was looping on encountering an indefinitely repeated back reference to
2865 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2866 Perl does - treats the match as successful.
2867
2868 ****

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