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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 206 - (show annotations) (download)
Fri Aug 3 14:53:04 2007 UTC (7 years, 1 month ago) by ph10
File size: 134642 byte(s)
Fix loop for null-matching condition nested in an outer unlimited repeat.

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

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