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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 217 - (show annotations) (download)
Wed Aug 15 15:21:04 2007 UTC (7 years ago) by ph10
File size: 136239 byte(s)
Crash in debug output when printing the name of a Unicode property.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.3 09-Aug-07
5 ---------------------
6
7 1. In the rejigging of the build system that eventually resulted in 7.1, the
8 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9 brackets there is not right, since it causes compilers to look for an
10 installed pcre.h, not the version that is in the source that is being
11 compiled (which of course may be different). I have changed it back to:
12
13 #include "pcre.h"
14
15 I have a vague recollection that the change was concerned with compiling in
16 different directories, but in the new build system, that is taken care of
17 by the VPATH setting the Makefile.
18
19 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20 when the subject happened to end in the byte 0x85 (e.g. if the last
21 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22 characters but of course it shouldn't be taken as a newline when it is part
23 of another character. The bug was that, for an unlimited repeat of . in
24 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25 characters when looking for a newline.
26
27 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28
29 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30 in debug output.
31
32 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33 long printing in the pcrecpp unittest when running under MinGW.
34
35 6. ESC_K was left out of the EBCDIC table.
36
37 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39 limit also applies to "virtual nesting" when a pattern is recursive, and in
40 this case 1000 isn't so big. I have been able to remove this limit at the
41 expense of backing off one optimization in certain circumstances. Normally,
42 when pcre_exec() would call its internal match() function recursively and
43 immediately return the result unconditionally, it uses a "tail recursion"
44 feature to save stack. However, when a subpattern that can match an empty
45 string has an unlimited repetition quantifier, it no longer makes this
46 optimization. That gives it a stack frame in which to save the data for
47 checking that an empty string has been matched. Previously this was taken
48 from the 1000-entry workspace that had been reserved. So now there is no
49 explicit limit, but more stack is used.
50
51 8. Applied Daniel's patches to solve problems with the import/export magic
52 syntax that is required for Windows, and which was going wrong for the
53 pcreposix and pcrecpp parts of the library. These were overlooked when this
54 problem was solved for the main library.
55
56 9. There were some crude static tests to avoid integer overflow when computing
57 the size of patterns that contain repeated groups with explicit upper
58 limits. As the maximum quantifier is 65535, the maximum group length was
59 set at 30,000 so that the product of these two numbers did not overflow a
60 32-bit integer. However, it turns out that people want to use groups that
61 are longer than 30,000 bytes (though not repeat them that many times).
62 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63 made it possible to implement the integer overflow checks in a much more
64 dynamic way, which I have now done. The artificial limitation on group
65 length has been removed - we now have only the limit on the total length of
66 the compiled pattern, which depends on the LINK_SIZE setting.
67
68 10. Fixed a bug in the documentation for get/copy named substring when
69 duplicate names are permitted. If none of the named substrings are set, the
70 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71 empty string.
72
73 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75 because the ] is interpreted as the first data character and the
76 terminating ] is not found. PCRE has been made compatible with Perl in this
77 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78 cause memory overwriting.
79
80 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81 string has been matched (to stop an infinite loop). It was not recognizing
82 a conditional subpattern that could match an empty string if that
83 subpattern was within another subpattern. For example, it looped when
84 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
85 condition was not nested. This bug has been fixed.
86
87 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88 past the start of the subject in the presence of bytes with the top bit
89 set, for example "\x8aBCD".
90
91 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93
94 14. Optimized (?!) to (*FAIL).
95
96 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
97 This restricts code points to be within the range 0 to 0x10FFFF, excluding
98 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
99 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
100 does: it's just the validity check that is more restrictive.
101
102 16. Inserted checks for integer overflows during escape sequence (backslash)
103 processing, and also fixed erroneous offset values for syntax errors during
104 backslash processing.
105
106 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
107 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
108
109 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
110 caused an overrun.
111
112 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
113 something other than just ASCII characters) inside a group that had an
114 unlimited repeat caused a loop at compile time (while checking to see
115 whether the group could match an empty string).
116
117 20. Debugging a pattern containing \p or \P could cause a crash. For example,
118 [\P{Any}] did so. (Error in the code for printing property names.)
119
120
121 Version 7.2 19-Jun-07
122 ---------------------
123
124 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
125 which is apparently normally available under Windows.
126
127 2. Re-jig the pcregrep tests with different newline settings in an attempt
128 to make them independent of the local environment's newline setting.
129
130 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
131
132 4. Some of the "internals" tests were previously cut out when the link size
133 was not 2, because the output contained actual offsets. The recent new
134 "Z" feature of pcretest means that these can be cut out, making the tests
135 usable with all link sizes.
136
137 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
138 stack recursion. This gives a massive performance boost under BSD, but just
139 a small improvement under Linux. However, it saves one field in the frame
140 in all cases.
141
142 6. Added more features from the forthcoming Perl 5.10:
143
144 (a) (?-n) (where n is a string of digits) is a relative subroutine or
145 recursion call. It refers to the nth most recently opened parentheses.
146
147 (b) (?+n) is also a relative subroutine call; it refers to the nth next
148 to be opened parentheses.
149
150 (c) Conditions that refer to capturing parentheses can be specified
151 relatively, for example, (?(-2)... or (?(+3)...
152
153 (d) \K resets the start of the current match so that everything before
154 is not part of it.
155
156 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
157
158 (f) \g{name} is another synonym - part of Perl 5.10's unification of
159 reference syntax.
160
161 (g) (?| introduces a group in which the numbering of parentheses in each
162 alternative starts with the same number.
163
164 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
165
166 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
167 PCRE_INFO_JCHANGED.
168
169 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
170 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
171 for detecting groups that can match an empty string.
172
173 9. A pattern with a very large number of alternatives (more than several
174 hundred) was running out of internal workspace during the pre-compile
175 phase, where pcre_compile() figures out how much memory will be needed. A
176 bit of new cunning has reduced the workspace needed for groups with
177 alternatives. The 1000-alternative test pattern now uses 12 bytes of
178 workspace instead of running out of the 4096 that are available.
179
180 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
181
182 11. Applied patch from Google to remove an optimization that didn't quite work.
183 The report of the bug said:
184
185 pcrecpp::RE("a*").FullMatch("aaa") matches, while
186 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
187 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
188
189 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
190 it matched the wrong number of bytes.
191
192
193 Version 7.1 24-Apr-07
194 ---------------------
195
196 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
197 that is more "standard", making use of automake and other Autotools. There
198 is some re-arrangement of the files and adjustment of comments consequent
199 on this.
200
201 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
202 for recursive directory scanning broke on some systems because the files
203 are not scanned in any specific order and on different systems the order
204 was different. A call to "sort" has been inserted into RunGrepTest for the
205 approprate test as a short-term fix. In the longer term there may be an
206 alternative.
207
208 3. I had an email from Eric Raymond about problems translating some of PCRE's
209 man pages to HTML (despite the fact that I distribute HTML pages, some
210 people do their own conversions for various reasons). The problems
211 concerned the use of low-level troff macros .br and .in. I have therefore
212 removed all such uses from the man pages (some were redundant, some could
213 be replaced by .nf/.fi pairs). The 132html script that I use to generate
214 HTML has been updated to handle .nf/.fi and to complain if it encounters
215 .br or .in.
216
217 4. Updated comments in configure.ac that get placed in config.h.in and also
218 arranged for config.h to be included in the distribution, with the name
219 config.h.generic, for the benefit of those who have to compile without
220 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
221
222 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
223 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
224 makevp.bat for the current PCRE, using the additional files
225 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
226
227 6. A Windows user reported a minor discrepancy with test 2, which turned out
228 to be caused by a trailing space on an input line that had got lost in his
229 copy. The trailing space was an accident, so I've just removed it.
230
231 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
232 that is needed.
233
234 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
235 as "const" (a) because they are and (b) because it helps the PHP
236 maintainers who have recently made a script to detect big data structures
237 in the php code that should be moved to the .rodata section. I remembered
238 to update Builducptable as well, so it won't revert if ucptable.h is ever
239 re-created.
240
241 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
242 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
243 order to be able to cut out the UTF-8 tables in the latter when UTF-8
244 support is not required. This saves 1.5-2K of code, which is important in
245 some applications.
246
247 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
248 so as not to refer to the tables, even though these functions will never be
249 called when UTF-8 support is disabled. Otherwise there are problems with a
250 shared library.
251
252 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
253
254 (a) It was defining its arguments as char * instead of void *.
255
256 (b) It was assuming that all moves were upwards in memory; this was true
257 a long time ago when I wrote it, but is no longer the case.
258
259 The emulated memove() is provided for those environments that have neither
260 memmove() nor bcopy(). I didn't think anyone used it these days, but that
261 is clearly not the case, as these two bugs were recently reported.
262
263 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
264 and Detrail to create the HTML documentation, the .txt form of the man
265 pages, and it removes trailing spaces from listed files. It also creates
266 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
267 case, it wraps all the #defines with #ifndefs. This script should be run
268 before "make dist".
269
270 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
271 with Unicode property support.
272
273 (a) For a maximizing quantifier, if the two different cases of the
274 character were of different lengths in their UTF-8 codings (there are
275 some cases like this - I found 11), and the matching function had to
276 back up over a mixture of the two cases, it incorrectly assumed they
277 were both the same length.
278
279 (b) When PCRE was configured to use the heap rather than the stack for
280 recursion during matching, it was not correctly preserving the data for
281 the other case of a UTF-8 character when checking ahead for a match
282 while processing a minimizing repeat. If the check also involved
283 matching a wide character, but failed, corruption could cause an
284 erroneous result when trying to check for a repeat of the original
285 character.
286
287 13. Some tidying changes to the testing mechanism:
288
289 (a) The RunTest script now detects the internal link size and whether there
290 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
291 values substituted by "configure". (The RunGrepTest script already did
292 this for UTF-8.) The configure.ac script no longer substitutes the
293 relevant variables.
294
295 (b) The debugging options /B and /D in pcretest show the compiled bytecode
296 with length and offset values. This means that the output is different
297 for different internal link sizes. Test 2 is skipped for link sizes
298 other than 2 because of this, bypassing the problem. Unfortunately,
299 there was also a test in test 3 (the locale tests) that used /B and
300 failed for link sizes other than 2. Rather than cut the whole test out,
301 I have added a new /Z option to pcretest that replaces the length and
302 offset values with spaces. This is now used to make test 3 independent
303 of link size. (Test 2 will be tidied up later.)
304
305 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
306 segmentation fault instead of returning the appropriate error message.
307
308 15. In multiline mode when the newline sequence was set to "any", the pattern
309 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
310 This doesn't seem right; it now treats the CRLF combination as the line
311 ending, and so does not match in that case. It's only a pattern such as ^$
312 that would hit this one: something like ^ABC$ would have failed after \r
313 and then tried again after \r\n.
314
315 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
316 in an attempt to make files that differ only in their line terminators
317 compare equal. This works on Linux.
318
319 17. Under certain error circumstances pcregrep might try to free random memory
320 as it exited. This is now fixed, thanks to valgrind.
321
322 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
323 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
324 was because its rules for how to advance for /g after matching an empty
325 string at the end of a line did not allow for this case. They now check for
326 it specially.
327
328 20. pcretest is supposed to handle patterns and data of any length, by
329 extending its buffers when necessary. It was getting this wrong when the
330 buffer for a data line had to be extended.
331
332 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
333 CRLF as a newline sequence.
334
335 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
336 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
337 I have nevertheless tidied it up.
338
339 23. Added some casts to kill warnings from HP-UX ia64 compiler.
340
341 24. Added a man page for pcre-config.
342
343
344 Version 7.0 19-Dec-06
345 ---------------------
346
347 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
348 moving to gcc 4.1.1.
349
350 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
351 sys/time.h, which is documented as needed for this function. It doesn't
352 seem to matter on Linux, but it showed up on some releases of OS X.
353
354 3. It seems that there are systems where bytes whose values are greater than
355 127 match isprint() in the "C" locale. The "C" locale should be the
356 default when a C program starts up. In most systems, only ASCII printing
357 characters match isprint(). This difference caused the output from pcretest
358 to vary, making some of the tests fail. I have changed pcretest so that:
359
360 (a) When it is outputting text in the compiled version of a pattern, bytes
361 other than 32-126 are always shown as hex escapes.
362
363 (b) When it is outputting text that is a matched part of a subject string,
364 it does the same, unless a different locale has been set for the match
365 (using the /L modifier). In this case, it uses isprint() to decide.
366
367 4. Fixed a major bug that caused incorrect computation of the amount of memory
368 required for a compiled pattern when options that changed within the
369 pattern affected the logic of the preliminary scan that determines the
370 length. The relevant options are -x, and -i in UTF-8 mode. The result was
371 that the computed length was too small. The symptoms of this bug were
372 either the PCRE error "internal error: code overflow" from pcre_compile(),
373 or a glibc crash with a message such as "pcretest: free(): invalid next
374 size (fast)". Examples of patterns that provoked this bug (shown in
375 pcretest format) are:
376
377 /(?-x: )/x
378 /(?x)(?-x: \s*#\s*)/
379 /((?i)[\x{c0}])/8
380 /(?i:[\x{c0}])/8
381
382 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
383 is now done differently.
384
385 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
386 wrapper classes; (b) implement a new function in the C++ scanner that is
387 more efficient than the old way of doing things because it avoids levels of
388 recursion in the regex matching; (c) add a paragraph to the documentation
389 for the FullMatch() function.
390
391 6. The escape sequence \n was being treated as whatever was defined as
392 "newline". Not only was this contrary to the documentation, which states
393 that \n is character 10 (hex 0A), but it also went horribly wrong when
394 "newline" was defined as CRLF. This has been fixed.
395
396 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
397 was being set to -1 for the "end of line" case (supposedly a value that no
398 character can have). Though this value is never used (the check for end of
399 line is "zero bytes in current character"), it caused compiler complaints.
400 I've changed it to 0xffffffff.
401
402 8. In pcre_version.c, the version string was being built by a sequence of
403 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
404 string (as it is for production releases) called a macro with an empty
405 argument. The C standard says the result of this is undefined. The gcc
406 compiler treats it as an empty string (which was what was wanted) but it is
407 reported that Visual C gives an error. The source has been hacked around to
408 avoid this problem.
409
410 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
411 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
412 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
413 of them did).
414
415 10. Originally, pcretest opened its input and output without "b"; then I was
416 told that "b" was needed in some environments, so it was added for release
417 5.0 to both the input and output. (It makes no difference on Unix-like
418 systems.) Later I was told that it is wrong for the input on Windows. I've
419 now abstracted the modes into two macros, to make it easier to fiddle with
420 them, and removed "b" from the input mode under Windows.
421
422 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
423
424 12. Added -help and --help to pcretest as an official way of being reminded
425 of the options.
426
427 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
428 and pcrecpp.cc because they annoy compilers at high warning levels.
429
430 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
431
432 15. Fixed an occurrence of == in configure.ac that should have been = (shell
433 scripts are not C programs :-) and which was not noticed because it works
434 on Linux.
435
436 16. pcretest is supposed to handle any length of pattern and data line (as one
437 line or as a continued sequence of lines) by extending its input buffer if
438 necessary. This feature was broken for very long pattern lines, leading to
439 a string of junk being passed to pcre_compile() if the pattern was longer
440 than about 50K.
441
442 17. I have done a major re-factoring of the way pcre_compile() computes the
443 amount of memory needed for a compiled pattern. Previously, there was code
444 that made a preliminary scan of the pattern in order to do this. That was
445 OK when PCRE was new, but as the facilities have expanded, it has become
446 harder and harder to keep it in step with the real compile phase, and there
447 have been a number of bugs (see for example, 4 above). I have now found a
448 cunning way of running the real compile function in a "fake" mode that
449 enables it to compute how much memory it would need, while actually only
450 ever using a few hundred bytes of working memory and without too many
451 tests of the mode. This should make future maintenance and development
452 easier. A side effect of this work is that the limit of 200 on the nesting
453 depth of parentheses has been removed (though this was never a serious
454 limitation, I suspect). However, there is a downside: pcre_compile() now
455 runs more slowly than before (30% or more, depending on the pattern). I
456 hope this isn't a big issue. There is no effect on runtime performance.
457
458 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
459 newline (only possible for the last line of a file) and it was a
460 pattern that set a locale (followed by /Lsomething), pcretest crashed.
461
462 19. Added additional timing features to pcretest. (1) The -tm option now times
463 matching only, not compiling. (2) Both -t and -tm can be followed, as a
464 separate command line item, by a number that specifies the number of
465 repeats to use when timing. The default is 50000; this gives better
466 precision, but takes uncomfortably long for very large patterns.
467
468 20. Extended pcre_study() to be more clever in cases where a branch of a
469 subpattern has no definite first character. For example, (a*|b*)[cd] would
470 previously give no result from pcre_study(). Now it recognizes that the
471 first character must be a, b, c, or d.
472
473 21. There was an incorrect error "recursive call could loop indefinitely" if
474 a subpattern (or the entire pattern) that was being tested for matching an
475 empty string contained only one non-empty item after a nested subpattern.
476 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
477 incorrectly, because the \d was being skipped in the check.
478
479 22. The pcretest program now has a new pattern option /B and a command line
480 option -b, which is equivalent to adding /B to every pattern. This causes
481 it to show the compiled bytecode, without the additional information that
482 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
483 is the same as /B/I).
484
485 23. A new optimization is now able automatically to treat some sequences such
486 as a*b as a*+b. More specifically, if something simple (such as a character
487 or a simple class like \d) has an unlimited quantifier, and is followed by
488 something that cannot possibly match the quantified thing, the quantifier
489 is automatically "possessified".
490
491 24. A recursive reference to a subpattern whose number was greater than 39
492 went wrong under certain circumstances in UTF-8 mode. This bug could also
493 have affected the operation of pcre_study().
494
495 25. Realized that a little bit of performance could be had by replacing
496 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
497
498 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
499
500 27. Possessive quantifiers such as a++ were previously implemented by turning
501 them into atomic groups such as ($>a+). Now they have their own opcodes,
502 which improves performance. This includes the automatically created ones
503 from 23 above.
504
505 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
506 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
507 the first matched character to be a colon. This applied both to named and
508 numbered groups.
509
510 29. The ucpinternal.h header file was missing its idempotency #ifdef.
511
512 30. I was sent a "project" file called libpcre.a.dev which I understand makes
513 building PCRE on Windows easier, so I have included it in the distribution.
514
515 31. There is now a check in pcretest against a ridiculously large number being
516 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
517 loop, the loop is abandoned.
518
519 32. Forward references to subpatterns in conditions such as (?(2)...) where
520 subpattern 2 is defined later cause pcre_compile() to search forwards in
521 the pattern for the relevant set of parentheses. This search went wrong
522 when there were unescaped parentheses in a character class, parentheses
523 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
524
525 33. "Subroutine" calls and backreferences were previously restricted to
526 referencing subpatterns earlier in the regex. This restriction has now
527 been removed.
528
529 34. Added a number of extra features that are going to be in Perl 5.10. On the
530 whole, these are just syntactic alternatives for features that PCRE had
531 previously implemented using the Python syntax or my own invention. The
532 other formats are all retained for compatibility.
533
534 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
535 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
536 also .NET compatible.
537
538 (b) A recursion or subroutine call to a named group can now be defined as
539 (?&name) as well as (?P>name).
540
541 (c) A backreference to a named group can now be defined as \k<name> or
542 \k'name' as well as (?P=name). The new forms, as well as being in Perl
543 5.10, are also .NET compatible.
544
545 (d) A conditional reference to a named group can now use the syntax
546 (?(<name>) or (?('name') as well as (?(name).
547
548 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
549 groups (named and numbered) that are never evaluated inline, but can be
550 called as "subroutines" from elsewhere. In effect, the DEFINE condition
551 is always false. There may be only one alternative in such a group.
552
553 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
554 as the simple (?(R). The condition is true only if the most recent
555 recursion is that of the given number or name. It does not search out
556 through the entire recursion stack.
557
558 (g) The escape \gN or \g{N} has been added, where N is a positive or
559 negative number, specifying an absolute or relative reference.
560
561 35. Tidied to get rid of some further signed/unsigned compiler warnings and
562 some "unreachable code" warnings.
563
564 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
565 things, this adds five new scripts.
566
567 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
568 There were also incompatibilities regarding the handling of \Q..\E inside
569 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
570 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
571
572 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
573 matches an empty string, and forcibly breaks the loop. There were bugs in
574 this code in non-simple cases. For a pattern such as ^(a()*)* matched
575 against aaaa the result was just "a" rather than "aaaa", for example. Two
576 separate and independent bugs (that affected different cases) have been
577 fixed.
578
579 39. Refactored the code to abolish the use of different opcodes for small
580 capturing bracket numbers. This is a tidy that I avoided doing when I
581 removed the limit on the number of capturing brackets for 3.5 back in 2001.
582 The new approach is not only tidier, it makes it possible to reduce the
583 memory needed to fix the previous bug (38).
584
585 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
586 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
587 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
588 mode.
589
590 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
591 report.
592
593 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
594 copy construction and assignment in the C++ wrapper.
595
596 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
597 couple of bugs that could have given wrong results in the "--newline=crlf"
598 case.
599
600 44. Added a number of casts and did some reorganization of signed/unsigned int
601 variables following suggestions from Dair Grant. Also renamed the variable
602 "this" as "item" because it is a C++ keyword.
603
604 45. Arranged for dftables to add
605
606 #include "pcre_internal.h"
607
608 to pcre_chartables.c because without it, gcc 4.x may remove the array
609 definition from the final binary if PCRE is built into a static library and
610 dead code stripping is activated.
611
612 46. For an unanchored pattern, if a match attempt fails at the start of a
613 newline sequence, and the newline setting is CRLF or ANY, and the next two
614 characters are CRLF, advance by two characters instead of one.
615
616
617 Version 6.7 04-Jul-06
618 ---------------------
619
620 1. In order to handle tests when input lines are enormously long, pcretest has
621 been re-factored so that it automatically extends its buffers when
622 necessary. The code is crude, but this _is_ just a test program. The
623 default size has been increased from 32K to 50K.
624
625 2. The code in pcre_study() was using the value of the re argument before
626 testing it for NULL. (Of course, in any sensible call of the function, it
627 won't be NULL.)
628
629 3. The memmove() emulation function in pcre_internal.h, which is used on
630 systems that lack both memmove() and bcopy() - that is, hardly ever -
631 was missing a "static" storage class specifier.
632
633 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
634 containing an extended class (one that cannot be represented by a bitmap
635 because it contains high-valued characters or Unicode property items, e.g.
636 [\pZ]). Almost always one would set UTF-8 mode when processing such a
637 pattern, but PCRE should not loop if you do not (it no longer does).
638 [Detail: two cases were found: (a) a repeated subpattern containing an
639 extended class; (b) a recursive reference to a subpattern that followed a
640 previous extended class. It wasn't skipping over the extended class
641 correctly when UTF-8 mode was not set.]
642
643 5. A negated single-character class was not being recognized as fixed-length
644 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
645 compile error "lookbehind assertion is not fixed length".
646
647 6. The RunPerlTest auxiliary script was showing an unexpected difference
648 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
649 write a Perl script that can interpret lines of an input file either as
650 byte characters or as UTF-8, which is what "perltest" was being required to
651 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
652 can't do is switch easily at run time between having the "use utf8;" pragma
653 or not. In the end, I fudged it by using the RunPerlTest script to insert
654 "use utf8;" explicitly for the UTF-8 tests.
655
656 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
657 the end of the subject string, contrary to the documentation and to what
658 Perl does. This was true of both matching functions. Now it matches only at
659 the start of the subject and immediately after *internal* newlines.
660
661 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
662 a pointer to an int instead of a pointer to an unsigned long int. This
663 caused problems on 64-bit systems.
664
665 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
666 instance of the 'standard' template library not being so standard".
667
668 10. There was no check on the number of named subpatterns nor the maximum
669 length of a subpattern name. The product of these values is used to compute
670 the size of the memory block for a compiled pattern. By supplying a very
671 long subpattern name and a large number of named subpatterns, the size
672 computation could be caused to overflow. This is now prevented by limiting
673 the length of names to 32 characters, and the number of named subpatterns
674 to 10,000.
675
676 11. Subpatterns that are repeated with specific counts have to be replicated in
677 the compiled pattern. The size of memory for this was computed from the
678 length of the subpattern and the repeat count. The latter is limited to
679 65535, but there was no limit on the former, meaning that integer overflow
680 could in principle occur. The compiled length of a repeated subpattern is
681 now limited to 30,000 bytes in order to prevent this.
682
683 12. Added the optional facility to have named substrings with the same name.
684
685 13. Added the ability to use a named substring as a condition, using the
686 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
687 are numbers (not recommended). Forward references are permitted.
688
689 14. Added forward references in named backreferences (if you see what I mean).
690
691 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
692 pattern could run off the end of the subject. For example, the pattern
693 "(?s)(.{1,5})"8 did this with the subject "ab".
694
695 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
696 PCRE_CASELESS was set when matching characters that were quantified with ?
697 or *.
698
699 17. A character class other than a single negated character that had a minimum
700 but no maximum quantifier - for example [ab]{6,} - was not handled
701 correctly by pce_dfa_exec(). It would match only one character.
702
703 18. A valid (though odd) pattern that looked like a POSIX character
704 class but used an invalid character after [ (for example [[,abc,]]) caused
705 pcre_compile() to give the error "Failed: internal error: code overflow" or
706 in some cases to crash with a glibc free() error. This could even happen if
707 the pattern terminated after [[ but there just happened to be a sequence of
708 letters, a binary zero, and a closing ] in the memory that followed.
709
710 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
711 over the years. Originally (before any Unicode support), just the bottom 8
712 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
713 output from "man perlunicode" includes this:
714
715 The regular expression compiler produces polymorphic opcodes. That
716 is, the pattern adapts to the data and automatically switches to
717 the Unicode character scheme when presented with Unicode data--or
718 instead uses a traditional byte scheme when presented with byte
719 data.
720
721 Sadly, a wide octal escape does not cause a switch, and in a string with
722 no other multibyte characters, these octal escapes are treated as before.
723 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
724 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
725 Unicode string.
726
727 I have not perpetrated such confusion in PCRE. Up till now, it took just
728 the bottom 8 bits, as in old Perl. I have now made octal escapes with
729 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
730 translate to the appropriate multibyte character.
731
732 29. Applied some refactoring to reduce the number of warnings from Microsoft
733 and Borland compilers. This has included removing the fudge introduced
734 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
735 a warning about an unused variable.
736
737 21. PCRE has not included VT (character 0x0b) in the set of whitespace
738 characters since release 4.0, because Perl (from release 5.004) does not.
739 [Or at least, is documented not to: some releases seem to be in conflict
740 with the documentation.] However, when a pattern was studied with
741 pcre_study() and all its branches started with \s, PCRE still included VT
742 as a possible starting character. Of course, this did no harm; it just
743 caused an unnecessary match attempt.
744
745 22. Removed a now-redundant internal flag bit that recorded the fact that case
746 dependency changed within the pattern. This was once needed for "required
747 byte" processing, but is no longer used. This recovers a now-scarce options
748 bit. Also moved the least significant internal flag bit to the most-
749 significant bit of the word, which was not previously used (hangover from
750 the days when it was an int rather than a uint) to free up another bit for
751 the future.
752
753 23. Added support for CRLF line endings as well as CR and LF. As well as the
754 default being selectable at build time, it can now be changed at runtime
755 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
756 specify that it is scanning data with non-default line endings.
757
758 24. Changed the definition of CXXLINK to make it agree with the definition of
759 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
760
761 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
762 recursions. This makes a big different to stack usage for some patterns.
763
764 26. If a subpattern containing a named recursion or subroutine reference such
765 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
766 the space required for the compiled pattern went wrong and gave too small a
767 value. Depending on the environment, this could lead to "Failed: internal
768 error: code overflow at offset 49" or "glibc detected double free or
769 corruption" errors.
770
771 27. Applied patches from Google (a) to support the new newline modes and (b) to
772 advance over multibyte UTF-8 characters in GlobalReplace.
773
774 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
775 difference for some implementation of PCRE in some Windows version.
776
777 29. Added some extra testing facilities to pcretest:
778
779 \q<number> in a data line sets the "match limit" value
780 \Q<number> in a data line sets the "match recursion limt" value
781 -S <number> sets the stack size, where <number> is in megabytes
782
783 The -S option isn't available for Windows.
784
785
786 Version 6.6 06-Feb-06
787 ---------------------
788
789 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
790 in pcreposix.h. I have copied the definition from pcre.h.
791
792 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
793 because pcre.h is no longer a built file.
794
795 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
796 not normally included in the compiled code.
797
798
799 Version 6.5 01-Feb-06
800 ---------------------
801
802 1. When using the partial match feature with pcre_dfa_exec(), it was not
803 anchoring the second and subsequent partial matches at the new starting
804 point. This could lead to incorrect results. For example, with the pattern
805 /1234/, partially matching against "123" and then "a4" gave a match.
806
807 2. Changes to pcregrep:
808
809 (a) All non-match returns from pcre_exec() were being treated as failures
810 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
811 error message is output. Some extra information is given for the
812 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
813 probably the only errors that are likely to be caused by users (by
814 specifying a regex that has nested indefinite repeats, for instance).
815 If there are more than 20 of these errors, pcregrep is abandoned.
816
817 (b) A binary zero was treated as data while matching, but terminated the
818 output line if it was written out. This has been fixed: binary zeroes
819 are now no different to any other data bytes.
820
821 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
822 used to set a locale for matching. The --locale=xxxx long option has
823 been added (no short equivalent) to specify a locale explicitly on the
824 pcregrep command, overriding the environment variables.
825
826 (d) When -B was used with -n, some line numbers in the output were one less
827 than they should have been.
828
829 (e) Added the -o (--only-matching) option.
830
831 (f) If -A or -C was used with -c (count only), some lines of context were
832 accidentally printed for the final match.
833
834 (g) Added the -H (--with-filename) option.
835
836 (h) The combination of options -rh failed to suppress file names for files
837 that were found from directory arguments.
838
839 (i) Added the -D (--devices) and -d (--directories) options.
840
841 (j) Added the -F (--fixed-strings) option.
842
843 (k) Allow "-" to be used as a file name for -f as well as for a data file.
844
845 (l) Added the --colo(u)r option.
846
847 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
848 is not present by default.
849
850 3. A nasty bug was discovered in the handling of recursive patterns, that is,
851 items such as (?R) or (?1), when the recursion could match a number of
852 alternatives. If it matched one of the alternatives, but subsequently,
853 outside the recursion, there was a failure, the code tried to back up into
854 the recursion. However, because of the way PCRE is implemented, this is not
855 possible, and the result was an incorrect result from the match.
856
857 In order to prevent this happening, the specification of recursion has
858 been changed so that all such subpatterns are automatically treated as
859 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
860
861 4. I had overlooked the fact that, in some locales, there are characters for
862 which isalpha() is true but neither isupper() nor islower() are true. In
863 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
864 and ordfeminine) are like this. This affected the treatment of \w and \W
865 when they appeared in character classes, but not when they appeared outside
866 a character class. The bit map for "word" characters is now created
867 separately from the results of isalnum() instead of just taking it from the
868 upper, lower, and digit maps. (Plus the underscore character, of course.)
869
870 5. The above bug also affected the handling of POSIX character classes such as
871 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
872 permanent tables. Instead, the bit maps for such a class were previously
873 created as the appropriate unions of the upper, lower, and digit bitmaps.
874 Now they are created by subtraction from the [[:word:]] class, which has
875 its own bitmap.
876
877 6. The [[:blank:]] character class matches horizontal, but not vertical space.
878 It is created by subtracting the vertical space characters (\x09, \x0a,
879 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
880 subtraction was done in the overall bitmap for a character class, meaning
881 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
882 be recognized. This bug has been fixed.
883
884 7. Patches from the folks at Google:
885
886 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
887 real life, but is still worth protecting against".
888
889 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
890 regular expressions".
891
892 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
893 have it.
894
895 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
896 "configure" and the latter not, in order to fix a problem somebody had
897 with compiling the Arg class on HP-UX.
898
899 (e) Improve the error-handling of the C++ wrapper a little bit.
900
901 (f) New tests for checking recursion limiting.
902
903 8. The pcre_memmove() function, which is used only if the environment does not
904 have a standard memmove() function (and is therefore rarely compiled),
905 contained two bugs: (a) use of int instead of size_t, and (b) it was not
906 returning a result (though PCRE never actually uses the result).
907
908 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
909 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
910 returned instead of calling malloc() with an overflowing number that would
911 most likely cause subsequent chaos.
912
913 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
914
915 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
916 with this option is matched, the nmatch and pmatch options of regexec() are
917 ignored.
918
919 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
920 provided in case anyone wants to the the POSIX interface with UTF-8
921 strings.
922
923 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
924 C++ linking (needed for some HP-UX environments).
925
926 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
927 (unused parameter) and in the pcre_printint() function (omitted "default"
928 switch label when the default is to do nothing).
929
930 15. Added some code to make it possible, when PCRE is compiled as a C++
931 library, to replace subject pointers for pcre_exec() with a smart pointer
932 class, thus making it possible to process discontinuous strings.
933
934 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
935 much the same function. They were added by different people who were trying
936 to make PCRE easy to compile on non-Unix systems. It has been suggested
937 that PCRE_EXPORT be abolished now that there is more automatic apparatus
938 for compiling on Windows systems. I have therefore replaced it with
939 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
940 defaults to "extern" for C or "extern C" for C++, which works fine on
941 Unix-like systems. It is now possible to override the value of PCRE_DATA_
942 SCOPE with something explicit in config.h. In addition:
943
944 (a) pcreposix.h still had just "extern" instead of either of these macros;
945 I have replaced it with PCRE_DATA_SCOPE.
946
947 (b) Functions such as _pcre_xclass(), which are internal to the library,
948 but external in the C sense, all had PCRE_EXPORT in their definitions.
949 This is apparently wrong for the Windows case, so I have removed it.
950 (It makes no difference on Unix-like systems.)
951
952 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
953 of recursive calls to match(). This is different to MATCH_LIMIT because
954 that limits the total number of calls to match(), not all of which increase
955 the depth of recursion. Limiting the recursion depth limits the amount of
956 stack (or heap if NO_RECURSE is set) that is used. The default can be set
957 when PCRE is compiled, and changed at run time. A patch from Google adds
958 this functionality to the C++ interface.
959
960 18. Changes to the handling of Unicode character properties:
961
962 (a) Updated the table to Unicode 4.1.0.
963
964 (b) Recognize characters that are not in the table as "Cn" (undefined).
965
966 (c) I revised the way the table is implemented to a much improved format
967 which includes recognition of ranges. It now supports the ranges that
968 are defined in UnicodeData.txt, and it also amalgamates other
969 characters into ranges. This has reduced the number of entries in the
970 table from around 16,000 to around 3,000, thus reducing its size
971 considerably. I realized I did not need to use a tree structure after
972 all - a binary chop search is just as efficient. Having reduced the
973 number of entries, I extended their size from 6 bytes to 8 bytes to
974 allow for more data.
975
976 (d) Added support for Unicode script names via properties such as \p{Han}.
977
978 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
979 matching that character.
980
981 20. When matching a repeated Unicode property with a minimum greater than zero,
982 (for example \pL{2,}), PCRE could look past the end of the subject if it
983 reached it while seeking the minimum number of characters. This could
984 happen only if some of the characters were more than one byte long, because
985 there is a check for at least the minimum number of bytes.
986
987 21. Refactored the implementation of \p and \P so as to be more general, to
988 allow for more different types of property in future. This has changed the
989 compiled form incompatibly. Anybody with saved compiled patterns that use
990 \p or \P will have to recompile them.
991
992 22. Added "Any" and "L&" to the supported property types.
993
994 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
995 but give a compile time error if the value is greater than 0xff.
996
997 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
998 accidentally not being installed or uninstalled.
999
1000 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1001 made were to insert the current release number. This seemed silly, because
1002 it made things harder for people building PCRE on systems that don't run
1003 "configure". I have turned pcre.h into a distributed file, no longer built
1004 by "configure", with the version identification directly included. There is
1005 no longer a pcre.h.in file.
1006
1007 However, this change necessitated a change to the pcre-config script as
1008 well. It is built from pcre-config.in, and one of the substitutions was the
1009 release number. I have updated configure.ac so that ./configure now finds
1010 the release number by grepping pcre.h.
1011
1012 26. Added the ability to run the tests under valgrind.
1013
1014
1015 Version 6.4 05-Sep-05
1016 ---------------------
1017
1018 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1019 "--" to be printed when multiple files were scanned, even when none of the
1020 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1021 consider it to be a bug, and have restored the previous behaviour.
1022
1023 2. A couple of code tidies to get rid of compiler warnings.
1024
1025 3. The pcretest program used to cheat by referring to symbols in the library
1026 whose names begin with _pcre_. These are internal symbols that are not
1027 really supposed to be visible externally, and in some environments it is
1028 possible to suppress them. The cheating is now confined to including
1029 certain files from the library's source, which is a bit cleaner.
1030
1031 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1032 file's purpose clearer.
1033
1034 5. Reorganized pcre_ucp_findchar().
1035
1036
1037 Version 6.3 15-Aug-05
1038 ---------------------
1039
1040 1. The file libpcre.pc.in did not have general read permission in the tarball.
1041
1042 2. There were some problems when building without C++ support:
1043
1044 (a) If C++ support was not built, "make install" and "make test" still
1045 tried to test it.
1046
1047 (b) There were problems when the value of CXX was explicitly set. Some
1048 changes have been made to try to fix these, and ...
1049
1050 (c) --disable-cpp can now be used to explicitly disable C++ support.
1051
1052 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1053 backslash in a target when C++ was disabled. This confuses some
1054 versions of "make", apparently. Using an intermediate variable solves
1055 this. (Same for CPP_LOBJ.)
1056
1057 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1058 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1059 necessary on certain architectures.
1060
1061 4. Added a setting of -export-symbols-regex to the link command to remove
1062 those symbols that are exported in the C sense, but actually are local
1063 within the library, and not documented. Their names all begin with
1064 "_pcre_". This is not a perfect job, because (a) we have to except some
1065 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1066 available (and never for static libraries). I have made a note to try to
1067 find a way round (a) in the future.
1068
1069
1070 Version 6.2 01-Aug-05
1071 ---------------------
1072
1073 1. There was no test for integer overflow of quantifier values. A construction
1074 such as {1111111111111111} would give undefined results. What is worse, if
1075 a minimum quantifier for a parenthesized subpattern overflowed and became
1076 negative, the calculation of the memory size went wrong. This could have
1077 led to memory overwriting.
1078
1079 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1080
1081 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1082 operating environments where this matters.
1083
1084 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1085 PCRE options from within the C++ wrapper.
1086
1087 5. Named capturing subpatterns were not being correctly counted when a pattern
1088 was compiled. This caused two problems: (a) If there were more than 100
1089 such subpatterns, the calculation of the memory needed for the whole
1090 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1091 back references of the form \12, where the number was greater than 9, were
1092 not recognized as back references, even though there were sufficient
1093 previous subpatterns.
1094
1095 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1096 versions of gcc, e.g. 2.95.4.
1097
1098
1099 Version 6.1 21-Jun-05
1100 ---------------------
1101
1102 1. There was one reference to the variable "posix" in pcretest.c that was not
1103 surrounded by "#if !defined NOPOSIX".
1104
1105 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1106 the cross-check on the old pcre_info() function, for the benefit of the
1107 cut-down version of PCRE that is currently imported into Exim.
1108
1109 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1110 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1111 patterns that start (?i) but I don't think that matters. The use of (?i) is
1112 just an example; this all applies to the other options as well.
1113
1114 4. Since libtool seems to echo the compile commands it is issuing, the output
1115 from "make" can be reduced a bit by putting "@" in front of each libtool
1116 compile command.
1117
1118 5. Patch from the folks at Google for configure.in to be a bit more thorough
1119 in checking for a suitable C++ installation before trying to compile the
1120 C++ stuff. This should fix a reported problem when a compiler was present,
1121 but no suitable headers.
1122
1123 6. The man pages all had just "PCRE" as their title. I have changed them to
1124 be the relevant file name. I have also arranged that these names are
1125 retained in the file doc/pcre.txt, which is a concatenation in text format
1126 of all the man pages except the little individual ones for each function.
1127
1128 7. The NON-UNIX-USE file had not been updated for the different set of source
1129 files that come with release 6. I also added a few comments about the C++
1130 wrapper.
1131
1132
1133 Version 6.0 07-Jun-05
1134 ---------------------
1135
1136 1. Some minor internal re-organization to help with my DFA experiments.
1137
1138 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1139 didn't matter for the library itself when fully configured, but did matter
1140 when compiling without UCP support, or within Exim, where the ucp files are
1141 not imported.
1142
1143 3. Refactoring of the library code to split up the various functions into
1144 different source modules. The addition of the new DFA matching code (see
1145 below) to a single monolithic source would have made it really too
1146 unwieldy, quite apart from causing all the code to be include in a
1147 statically linked application, when only some functions are used. This is
1148 relevant even without the DFA addition now that patterns can be compiled in
1149 one application and matched in another.
1150
1151 The downside of splitting up is that there have to be some external
1152 functions and data tables that are used internally in different modules of
1153 the library but which are not part of the API. These have all had their
1154 names changed to start with "_pcre_" so that they are unlikely to clash
1155 with other external names.
1156
1157 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1158 a different (DFA) algorithm. Although it is slower than the original
1159 function, it does have some advantages for certain types of matching
1160 problem.
1161
1162 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1163 including restarting after a partial match.
1164
1165 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1166 defined when compiling for Windows was sent to me. I have put it into the
1167 code, though I have no means of testing or verifying it.
1168
1169 7. Added the pcre_refcount() auxiliary function.
1170
1171 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1172 match before or at the first newline in the subject string. In pcretest,
1173 the /f option on a pattern can be used to set this.
1174
1175 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1176 would behave wrongly. This has been present in PCRE since release 4.0.
1177
1178 10. A number of changes to the pcregrep command:
1179
1180 (a) Refactored how -x works; insert ^(...)$ instead of setting
1181 PCRE_ANCHORED and checking the length, in preparation for adding
1182 something similar for -w.
1183
1184 (b) Added the -w (match as a word) option.
1185
1186 (c) Refactored the way lines are read and buffered so as to have more
1187 than one at a time available.
1188
1189 (d) Implemented a pcregrep test script.
1190
1191 (e) Added the -M (multiline match) option. This allows patterns to match
1192 over several lines of the subject. The buffering ensures that at least
1193 8K, or the rest of the document (whichever is the shorter) is available
1194 for matching (and similarly the previous 8K for lookbehind assertions).
1195
1196 (f) Changed the --help output so that it now says
1197
1198 -w, --word-regex(p)
1199
1200 instead of two lines, one with "regex" and the other with "regexp"
1201 because that confused at least one person since the short forms are the
1202 same. (This required a bit of code, as the output is generated
1203 automatically from a table. It wasn't just a text change.)
1204
1205 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1206 option but starts with a hyphen. Could be a pattern or a path name
1207 starting with a hyphen, for instance.
1208
1209 (h) "-" can be given as a file name to represent stdin.
1210
1211 (i) When file names are being printed, "(standard input)" is used for
1212 the standard input, for compatibility with GNU grep. Previously
1213 "<stdin>" was used.
1214
1215 (j) The option --label=xxx can be used to supply a name to be used for
1216 stdin when file names are being printed. There is no short form.
1217
1218 (k) Re-factored the options decoding logic because we are going to add
1219 two more options that take data. Such options can now be given in four
1220 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1221
1222 (l) Added the -A, -B, and -C options for requesting that lines of context
1223 around matches be printed.
1224
1225 (m) Added the -L option to print the names of files that do not contain
1226 any matching lines, that is, the complement of -l.
1227
1228 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1229 continue to scan other files.
1230
1231 (o) The -s option was incorrectly implemented. For compatibility with other
1232 greps, it now suppresses the error message for a non-existent or non-
1233 accessible file (but not the return code). There is a new option called
1234 -q that suppresses the output of matching lines, which was what -s was
1235 previously doing.
1236
1237 (p) Added --include and --exclude options to specify files for inclusion
1238 and exclusion when recursing.
1239
1240 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1241 Hopefully, it now does.
1242
1243 12. Missing cast in pcre_study().
1244
1245 13. Added an "uninstall" target to the makefile.
1246
1247 14. Replaced "extern" in the function prototypes in Makefile.in with
1248 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1249 world, but is set differently for Windows.
1250
1251 15. Added a second compiling function called pcre_compile2(). The only
1252 difference is that it has an extra argument, which is a pointer to an
1253 integer error code. When there is a compile-time failure, this is set
1254 non-zero, in addition to the error test pointer being set to point to an
1255 error message. The new argument may be NULL if no error number is required
1256 (but then you may as well call pcre_compile(), which is now just a
1257 wrapper). This facility is provided because some applications need a
1258 numeric error indication, but it has also enabled me to tidy up the way
1259 compile-time errors are handled in the POSIX wrapper.
1260
1261 16. Added VPATH=.libs to the makefile; this should help when building with one
1262 prefix path and installing with another. (Or so I'm told by someone who
1263 knows more about this stuff than I do.)
1264
1265 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1266 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1267 match everything, including newlines. This is not POSIX-compatible, but
1268 somebody wanted the feature. From pcretest it can be activated by using
1269 both the P and the s flags.
1270
1271 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1272
1273 19. libpcre.pc was being incorrectly installed as executable.
1274
1275 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1276 it now also looks for '\r' so that it will work unmodified on Windows.
1277
1278 21. Added Google's contributed C++ wrapper to the distribution.
1279
1280 22. Added some untidy missing memory free() calls in pcretest, to keep
1281 Electric Fence happy when testing.
1282
1283
1284
1285 Version 5.0 13-Sep-04
1286 ---------------------
1287
1288 1. Internal change: literal characters are no longer packed up into items
1289 containing multiple characters in a single byte-string. Each character
1290 is now matched using a separate opcode. However, there may be more than one
1291 byte in the character in UTF-8 mode.
1292
1293 2. The pcre_callout_block structure has two new fields: pattern_position and
1294 next_item_length. These contain the offset in the pattern to the next match
1295 item, and its length, respectively.
1296
1297 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1298 insertion of callouts before each pattern item. Added the /C option to
1299 pcretest to make use of this.
1300
1301 4. On the advice of a Windows user, the lines
1302
1303 #if defined(_WIN32) || defined(WIN32)
1304 _setmode( _fileno( stdout ), 0x8000 );
1305 #endif /* defined(_WIN32) || defined(WIN32) */
1306
1307 have been added to the source of pcretest. This apparently does useful
1308 magic in relation to line terminators.
1309
1310 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1311 for the benefit of those environments where the "b" makes a difference.
1312
1313 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1314 to know about it. I have put a hack into configure.in that adds in code
1315 to set GCC=yes if CC=icc. This seems to end up at a point in the
1316 generated configure script that is early enough to affect the setting of
1317 compiler options, which is what is needed, but I have no means of testing
1318 whether it really works. (The user who reported this had patched the
1319 generated configure script, which of course I cannot do.)
1320
1321 LATER: After change 22 below (new libtool files), the configure script
1322 seems to know about icc (and also ecc). Therefore, I have commented out
1323 this hack in configure.in.
1324
1325 7. Added support for pkg-config (2 patches were sent in).
1326
1327 8. Negated POSIX character classes that used a combination of internal tables
1328 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1329 [[:^ascii]]. Typically, they would match almost any characters. The other
1330 POSIX classes were not broken in this way.
1331
1332 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1333 to find the match, as PCRE was deluded into thinking that the match had to
1334 start at the start point or following a newline. The same bug applied to
1335 patterns with negative forward assertions or any backward assertions
1336 preceding ".*" at the start, unless the pattern required a fixed first
1337 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1338
1339 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1340 starting at the last subject character, bytes beyond the end of the subject
1341 string were read.
1342
1343 11. Renamed the variable "class" as "classbits" to make life easier for C++
1344 users. (Previously there was a macro definition, but it apparently wasn't
1345 enough.)
1346
1347 12. Added the new field "tables" to the extra data so that tables can be passed
1348 in at exec time, or the internal tables can be re-selected. This allows
1349 a compiled regex to be saved and re-used at a later time by a different
1350 program that might have everything at different addresses.
1351
1352 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1353 -R library as well as a -L library.
1354
1355 14. The debugging options of pcretest (-d on the command line or D on a
1356 pattern) showed incorrect output for anything following an extended class
1357 that contained multibyte characters and which was followed by a quantifier.
1358
1359 15. Added optional support for general category Unicode character properties
1360 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1361 support. It adds about 90K to the size of the library. The meanings of the
1362 inbuilt class escapes such as \d and \s have NOT been changed.
1363
1364 16. Updated pcredemo.c to include calls to free() to release the memory for the
1365 compiled pattern.
1366
1367 17. The generated file chartables.c was being created in the source directory
1368 instead of in the building directory. This caused the build to fail if the
1369 source directory was different from the building directory, and was
1370 read-only.
1371
1372 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1373 file. No doubt somebody will tell me if they don't make sense... Also added
1374 Dan Mooney's comments about building on OpenVMS.
1375
1376 19. Added support for partial matching via the PCRE_PARTIAL option for
1377 pcre_exec() and the \P data escape in pcretest.
1378
1379 20. Extended pcretest with 3 new pattern features:
1380
1381 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1382 write the compiled pattern to the file whose name is "rest-of-line".
1383 This is a straight binary dump of the data, with the saved pointer to
1384 the character tables forced to be NULL. The study data, if any, is
1385 written too. After writing, pcretest reads a new pattern.
1386
1387 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1388 compiled pattern from the given file. There must not be any
1389 occurrences of "<" in the file name (pretty unlikely); if there are,
1390 pcretest will instead treat the initial "<" as a pattern delimiter.
1391 After reading in the pattern, pcretest goes on to read data lines as
1392 usual.
1393
1394 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1395 and 16-bit fields in a compiled pattern, to simulate a pattern that
1396 was compiled on a host of opposite endianness.
1397
1398 21. The pcre-exec() function can now cope with patterns that were compiled on
1399 hosts of opposite endianness, with this restriction:
1400
1401 As for any compiled expression that is saved and used later, the tables
1402 pointer field cannot be preserved; the extra_data field in the arguments
1403 to pcre_exec() should be used to pass in a tables address if a value
1404 other than the default internal tables were used at compile time.
1405
1406 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1407 now diagnosed as an error. Previously, most of the time, a negative number
1408 would have been treated as zero, but if in addition "ovector" was passed as
1409 NULL, a crash could occur.
1410
1411 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1412 new versions from the libtool 1.5 distribution (the last one is a copy of
1413 a file called libtool.m4). This seems to have fixed the need to patch
1414 "configure" to support Darwin 1.3 (which I used to do). However, I still
1415 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1416 workstation).
1417
1418 24. Changed the PCRE licence to be the more standard "BSD" licence.
1419
1420
1421 Version 4.5 01-Dec-03
1422 ---------------------
1423
1424 1. There has been some re-arrangement of the code for the match() function so
1425 that it can be compiled in a version that does not call itself recursively.
1426 Instead, it keeps those local variables that need separate instances for
1427 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1428 needs to "recurse". Keeping track of where control must go is done by means
1429 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1430 hide most of the details from the main code, and operates only if
1431 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1432 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1433 operating.
1434
1435 To make it easier for callers to provide specially tailored get/free
1436 functions for this usage, two new functions, pcre_stack_malloc, and
1437 pcre_stack_free, are used. They are always called in strict stacking order,
1438 and the size of block requested is always the same.
1439
1440 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1441 PCRE has been compiled to use the stack or the heap for recursion. The
1442 -C option of pcretest uses this to show which version is compiled.
1443
1444 A new data escape \S, is added to pcretest; it causes the amounts of store
1445 obtained and freed by both kinds of malloc/free at match time to be added
1446 to the output.
1447
1448 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1449 what's available on my current Linux desktop machine.
1450
1451 3. When matching a UTF-8 string, the test for a valid string at the start has
1452 been extended. If start_offset is not zero, PCRE now checks that it points
1453 to a byte that is the start of a UTF-8 character. If not, it returns
1454 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1455 this is necessary because there may be backward assertions in the pattern.
1456 When matching the same subject several times, it may save resources to use
1457 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1458
1459 4. The code for checking the validity of UTF-8 strings has been tightened so
1460 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1461 containing "overlong sequences".
1462
1463 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1464 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1465 should have been "&", but it just so happened that all the cases this let
1466 through by mistake were picked up later in the function.
1467
1468 6. I had used a variable called "isblank" - this is a C99 function, causing
1469 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1470
1471 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1472 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1473 so the version is shown; (c) stop "make test" echoing "./RunTest".
1474
1475 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1476
1477 9. The prototype for memmove() for systems that don't have it was using
1478 size_t, but the inclusion of the header that defines size_t was later. I've
1479 moved the #includes for the C headers earlier to avoid this.
1480
1481 10. Added some adjustments to the code to make it easier to compiler on certain
1482 special systems:
1483
1484 (a) Some "const" qualifiers were missing.
1485 (b) Added the macro EXPORT before all exported functions; by default this
1486 is defined to be empty.
1487 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1488 that it reads its output file name as an argument instead of writing
1489 to the standard output and assuming this can be redirected.
1490
1491 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1492 class containing characters with values greater than 255, PCRE compilation
1493 went into a loop.
1494
1495 12. A recursive reference to a subpattern that was within another subpattern
1496 that had a minimum quantifier of zero caused PCRE to crash. For example,
1497 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1498 recursion. If the recursively-called subpattern itself had a zero repeat,
1499 that was OK.
1500
1501 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1502 buffer into which it was copied (for escape processing) was still set at
1503 1024, so long lines caused crashes.
1504
1505 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1506 "internal error: code overflow...". This applied to any character class
1507 that was followed by a possessive quantifier.
1508
1509 15. Modified the Makefile to add libpcre.la as a prerequisite for
1510 libpcreposix.la because I was told this is needed for a parallel build to
1511 work.
1512
1513 16. If a pattern that contained .* following optional items at the start was
1514 studied, the wrong optimizing data was generated, leading to matching
1515 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1516 matching string must start with a or b or c. The correct conclusion for
1517 this pattern is that a match can start with any character.
1518
1519
1520 Version 4.4 13-Aug-03
1521 ---------------------
1522
1523 1. In UTF-8 mode, a character class containing characters with values between
1524 127 and 255 was not handled correctly if the compiled pattern was studied.
1525 In fixing this, I have also improved the studying algorithm for such
1526 classes (slightly).
1527
1528 2. Three internal functions had redundant arguments passed to them. Removal
1529 might give a very teeny performance improvement.
1530
1531 3. Documentation bug: the value of the capture_top field in a callout is *one
1532 more than* the number of the hightest numbered captured substring.
1533
1534 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1535 in incorrectly linking with a previously installed version. They now link
1536 explicitly with libpcre.la.
1537
1538 5. configure.in no longer needs to recognize Cygwin specially.
1539
1540 6. A problem in pcre.in for Windows platforms is fixed.
1541
1542 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1543 pcretest, it used to include the size of the study block as part of its
1544 output. Unfortunately, the structure contains a field that has a different
1545 size on different hardware architectures. This meant that the tests that
1546 showed this size failed. As the block is currently always of a fixed size,
1547 this information isn't actually particularly useful in pcretest output, so
1548 I have just removed it.
1549
1550 8. Three pre-processor statements accidentally did not start in column 1.
1551 Sadly, there are *still* compilers around that complain, even though
1552 standard C has not required this for well over a decade. Sigh.
1553
1554 9. In pcretest, the code for checking callouts passed small integers in the
1555 callout_data field, which is a void * field. However, some picky compilers
1556 complained about the casts involved for this on 64-bit systems. Now
1557 pcretest passes the address of the small integer instead, which should get
1558 rid of the warnings.
1559
1560 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1561 both compile and run time, and gives an error if an invalid UTF-8 sequence
1562 is found. There is a option for disabling this check in cases where the
1563 string is known to be correct and/or the maximum performance is wanted.
1564
1565 11. In response to a bug report, I changed one line in Makefile.in from
1566
1567 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1568 to
1569 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1570
1571 to look similar to other lines, but I have no way of telling whether this
1572 is the right thing to do, as I do not use Windows. No doubt I'll get told
1573 if it's wrong...
1574
1575
1576 Version 4.3 21-May-03
1577 ---------------------
1578
1579 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1580 Makefile.
1581
1582 2. Some refactoring to improve the quality of the code:
1583
1584 (i) The utf8_table... variables are now declared "const".
1585
1586 (ii) The code for \cx, which used the "case flipping" table to upper case
1587 lower case letters, now just substracts 32. This is ASCII-specific,
1588 but the whole concept of \cx is ASCII-specific, so it seems
1589 reasonable.
1590
1591 (iii) PCRE was using its character types table to recognize decimal and
1592 hexadecimal digits in the pattern. This is silly, because it handles
1593 only 0-9, a-f, and A-F, but the character types table is locale-
1594 specific, which means strange things might happen. A private
1595 table is now used for this - though it costs 256 bytes, a table is
1596 much faster than multiple explicit tests. Of course, the standard
1597 character types table is still used for matching digits in subject
1598 strings against \d.
1599
1600 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1601 ending in _t are). So I've renamed it as ESC_tee.
1602
1603 3. The first argument for regexec() in the POSIX wrapper should have been
1604 defined as "const".
1605
1606 4. Changed pcretest to use malloc() for its buffers so that they can be
1607 Electric Fenced for debugging.
1608
1609 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1610 to read one or more bytes before the start of the subject string. Often this
1611 had no effect on PCRE's behaviour, but in some circumstances it could
1612 provoke a segmentation fault.
1613
1614 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1615 to try to read one or more bytes before the start of the subject string.
1616
1617 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1618 UTF-8 support could misbehave in various ways if the subject string
1619 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1620 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1621 back over UTF-8 characters.)
1622
1623
1624 Version 4.2 14-Apr-03
1625 ---------------------
1626
1627 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1628
1629 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1630 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1631 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1632 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1633 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1634 and BUILD_EXEEXT
1635 Note: automatic setting of the BUILD variables is not yet working
1636 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1637 compile-time but not at link-time
1638 [LINK]: use for linking executables only
1639 make different versions for Windows and non-Windows
1640 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1641 libraries
1642 [LINK_FOR_BUILD]: new variable
1643 [OBJEXT]: use throughout
1644 [EXEEXT]: use throughout
1645 <winshared>: new target
1646 <wininstall>: new target
1647 <dftables.o>: use native compiler
1648 <dftables>: use native linker
1649 <install>: handle Windows platform correctly
1650 <clean>: ditto
1651 <check>: ditto
1652 copy DLL to top builddir before testing
1653
1654 As part of these changes, -no-undefined was removed again. This was reported
1655 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1656 in any case.
1657
1658 3. Some tidies to get rid of compiler warnings:
1659
1660 . In the match_data structure, match_limit was an unsigned long int, whereas
1661 match_call_count was an int. I've made them both unsigned long ints.
1662
1663 . In pcretest the fact that a const uschar * doesn't automatically cast to
1664 a void * provoked a warning.
1665
1666 . Turning on some more compiler warnings threw up some "shadow" variables
1667 and a few more missing casts.
1668
1669 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1670 option, a class that contained a single character with a value between 128
1671 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1672
1673 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1674 option, a class that contained several characters, but with at least one
1675 whose value was between 128 and 255 caused PCRE to crash.
1676
1677
1678 Version 4.1 12-Mar-03
1679 ---------------------
1680
1681 1. Compiling with gcc -pedantic found a couple of places where casts were
1682 needed, and a string in dftables.c that was longer than standard compilers are
1683 required to support.
1684
1685 2. Compiling with Sun's compiler found a few more places where the code could
1686 be tidied up in order to avoid warnings.
1687
1688 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1689 first of these names is deprecated in the latest Autoconf in favour of the name
1690 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1691 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1692 analogy I have changed it to CFLAGS_FOR_BUILD.
1693
1694 4. Added -no-undefined to the linking command in the Makefile, because this is
1695 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1696 linking step for the pcreposix library.
1697
1698 5. PCRE was failing to diagnose the case of two named groups with the same
1699 name.
1700
1701 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1702 literal character that is needed in the subject for a match, and scans along to
1703 ensure that it is present before embarking on the full matching process. This
1704 saves time in cases of nested unlimited repeats that are never going to match.
1705 Problem: the scan can take a lot of time if the subject is very long (e.g.
1706 megabytes), thus penalizing straightforward matches. It is now done only if the
1707 amount of subject to be scanned is less than 1000 bytes.
1708
1709 7. A lesser problem with the same optimization is that it was recording the
1710 first character of an anchored pattern as "needed", thus provoking a search
1711 right along the subject, even when the first match of the pattern was going to
1712 fail. The "needed" character is now not set for anchored patterns, unless it
1713 follows something in the pattern that is of non-fixed length. Thus, it still
1714 fulfils its original purpose of finding quick non-matches in cases of nested
1715 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1716
1717
1718 Version 4.0 17-Feb-03
1719 ---------------------
1720
1721 1. If a comment in an extended regex that started immediately after a meta-item
1722 extended to the end of string, PCRE compiled incorrect data. This could lead to
1723 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1724
1725 2. Moved to autoconf 2.53 and libtool 1.4.2.
1726
1727 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1728 the special perltest8 script is no longer needed - all the tests can be run
1729 from a single perltest script.
1730
1731 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1732 by \s. It has now been removed in PCRE. This means it isn't recognized as
1733 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1734 class [:space:] *does* include VT, thereby creating a mess.
1735
1736 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1737 space and tab.
1738
1739 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1740 its new features into the main test script, reducing the number of scripts.
1741
1742 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1743 were backward compatible, and made the (?i) apply to the whole pattern, as if
1744 /i were given. Now it behaves more logically, and applies the option setting
1745 only to what follows. PCRE has been changed to follow suit. However, if it
1746 finds options settings right at the start of the pattern, it extracts them into
1747 the global options, as before. Thus, they show up in the info data.
1748
1749 8. Added support for the \Q...\E escape sequence. Characters in between are
1750 treated as literals. This is slightly different from Perl in that $ and @ are
1751 also handled as literals inside the quotes. In Perl, they will cause variable
1752 interpolation. Note the following examples:
1753
1754 Pattern PCRE matches Perl matches
1755
1756 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1757 \Qabc\$xyz\E abc\$xyz abc\$xyz
1758 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1759
1760 For compatibility with Perl, \Q...\E sequences are recognized inside character
1761 classes as well as outside them.
1762
1763 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1764 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1765 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1766 signed/unsigned warnings.
1767
1768 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1769 option for pcretest, so I've replaced it by a simple function that does just
1770 that job.
1771
1772 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1773 "pcregrep -".
1774
1775 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1776 Java package. This provides some syntactic sugar for simple cases of what my
1777 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1778 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1779 item, you can use this simplified notation. Note that only makes sense with
1780 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1781 greediness, whatever the setting of the PCRE_UNGREEDY option.
1782
1783 13. A change of greediness default within a pattern was not taking effect at
1784 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1785 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1786 was abstracted outside.
1787
1788 14. PCRE now supports the \G assertion. It is true when the current matching
1789 position is at the start point of the match. This differs from \A when the
1790 starting offset is non-zero. Used with the /g option of pcretest (or similar
1791 code), it works in the same way as it does for Perl's /g option. If all
1792 alternatives of a regex begin with \G, the expression is anchored to the start
1793 match position, and the "anchored" flag is set in the compiled expression.
1794
1795 15. Some bugs concerning the handling of certain option changes within patterns
1796 have been fixed. These applied to options other than (?ims). For example,
1797 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1798 been the other way round. Some of this was related to change 7 above.
1799
1800 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1801 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1802 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1803 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1804
1805 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1806 mode. Unlike ".", it always matches newline, whatever the setting of
1807 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1808 assertions. Perl allows it, but it doesn't (in general) work because it can't
1809 calculate the length of the lookbehind. At least, that's the case for Perl
1810 5.8.0 - I've been told they are going to document that it doesn't work in
1811 future.
1812
1813 18. Added an error diagnosis for escapes that PCRE does not support: these are
1814 \L, \l, \N, \P, \p, \U, \u, and \X.
1815
1816 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1817 reading past the end of the pattern in cases such as /[abcd/.
1818
1819 20. PCRE was getting more memory than necessary for patterns with classes that
1820 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1821
1822 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1823 compiling PCRE for use with Virtual Pascal.
1824
1825 22. Small fix to the Makefile to make it work properly if the build is done
1826 outside the source tree.
1827
1828 23. Added a new extension: a condition to go with recursion. If a conditional
1829 subpattern starts with (?(R) the "true" branch is used if recursion has
1830 happened, whereas the "false" branch is used only at the top level.
1831
1832 24. When there was a very long string of literal characters (over 255 bytes
1833 without UTF support, over 250 bytes with UTF support), the computation of how
1834 much memory was required could be incorrect, leading to segfaults or other
1835 strange effects.
1836
1837 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1838 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1839 there was a subsequent back reference to those brackets. This meant that, for
1840 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1841 possible to check for precisely this case. All we can do is abandon the
1842 optimization if .* occurs inside capturing brackets when there are any back
1843 references whatsoever. (See below for a better fix that came later.)
1844
1845 26. The handling of the optimization for finding the first character of a
1846 non-anchored pattern, and for finding a character that is required later in the
1847 match were failing in some cases. This didn't break the matching; it just
1848 failed to optimize when it could. The way this is done has been re-implemented.
1849
1850 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1851
1852 28. Added a new feature that provides some of the functionality that Perl
1853 provides with (?{...}). The facility is termed a "callout". The way it is done
1854 in PCRE is for the caller to provide an optional function, by setting
1855 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1856 global variable. By default it is unset, which disables all calling out. To get
1857 the function called, the regex must include (?C) at appropriate points. This
1858 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1859 This provides a means of identifying different callout points. When PCRE
1860 reaches such a point in the regex, if pcre_callout has been set, the external
1861 function is called. It is provided with data in a structure called
1862 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1863 matching continues; if it returns a non-zero value, the match at the current
1864 point fails. However, backtracking will occur if possible. [This was changed
1865 later and other features added - see item 49 below.]
1866
1867 29. pcretest is upgraded to test the callout functionality. It provides a
1868 callout function that displays information. By default, it shows the start of
1869 the match and the current position in the text. There are some new data escapes
1870 to vary what happens:
1871
1872 \C+ in addition, show current contents of captured substrings
1873 \C- do not supply a callout function
1874 \C!n return 1 when callout number n is reached
1875 \C!n!m return 1 when callout number n is reached for the mth time
1876
1877 30. If pcregrep was called with the -l option and just a single file name, it
1878 output "<stdin>" if a match was found, instead of the file name.
1879
1880 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1881 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1882 pcre_exec(). This saves a malloc/free per call. The default value of
1883 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1884 when configuring.
1885
1886 32. The default maximum size of a compiled pattern is 64K. There have been a
1887 few cases of people hitting this limit. The code now uses macros to handle the
1888 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1889 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1890 configuring. Tests 2 and 5 work only with 2-byte links because they output
1891 debugging information about compiled patterns.
1892
1893 33. Internal code re-arrangements:
1894
1895 (a) Moved the debugging function for printing out a compiled regex into
1896 its own source file (printint.c) and used #include to pull it into
1897 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1898 separate copies.
1899
1900 (b) Defined the list of op-code names for debugging as a macro in
1901 internal.h so that it is next to the definition of the opcodes.
1902
1903 (c) Defined a table of op-code lengths for simpler skipping along compiled
1904 code. This is again a macro in internal.h so that it is next to the
1905 definition of the opcodes.
1906
1907 34. Added support for recursive calls to individual subpatterns, along the
1908 lines of Robin Houston's patch (but implemented somewhat differently).
1909
1910 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1911 allow it to read and process whole directories in Win32. This code was
1912 contributed by Lionel Fourquaux; it has not been tested by me.
1913
1914 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1915 used to name a group. Names consist of alphanumerics and underscores, and must
1916 be unique. Back references use the syntax (?P=name) and recursive calls use
1917 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1918 numbers. The function pcre_fullinfo() can be used after compilation to extract
1919 a name/number map. There are three relevant calls:
1920
1921 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1922 PCRE_INFO_NAMECOUNT yields the number of entries
1923 PCRE_INFO_NAMETABLE yields a pointer to the map.
1924
1925 The map is a vector of fixed-size entries. The size of each entry depends on
1926 the length of the longest name used. The first two bytes of each entry are the
1927 group number, most significant byte first. There follows the corresponding
1928 name, zero terminated. The names are in alphabetical order.
1929
1930 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1931 case instead of 255. Making it the same both with and without UTF-8 support
1932 means that the same test output works with both.
1933
1934 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1935 calling malloc() with a zero argument.
1936
1937 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1938 optimization. I've improved things by keeping a bitmap of backreferences with
1939 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1940 fact referenced, the optimization can be applied. It is unlikely that a
1941 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1942 the match to follow \n) will appear inside brackets with a number greater than
1943 31, but if it does, any back reference > 31 suppresses the optimization.
1944
1945 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1946 of disabling numbered capturing parentheses. Any opening parenthesis that is
1947 not followed by ? behaves as if it were followed by ?: but named parentheses
1948 can still be used for capturing (and they will acquire numbers in the usual
1949 way).
1950
1951 41. Redesigned the return codes from the match() function into yes/no/error so
1952 that errors can be passed back from deep inside the nested calls. A malloc
1953 failure while inside a recursive subpattern call now causes the
1954 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1955
1956 42. It is now possible to set a limit on the number of times the match()
1957 function is called in a call to pcre_exec(). This facility makes it possible to
1958 limit the amount of recursion and backtracking, though not in a directly
1959 obvious way, because the match() function is used in a number of different
1960 circumstances. The count starts from zero for each position in the subject
1961 string (for non-anchored patterns). The default limit is, for compatibility, a
1962 large number, namely 10 000 000. You can change this in two ways:
1963
1964 (a) When configuring PCRE before making, you can use --with-match-limit=n
1965 to set a default value for the compiled library.
1966
1967 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1968 a different value is set. See 45 below.
1969
1970 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1971
1972 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1973 of things that can be changed at compile time. The first argument specifies
1974 what is wanted and the second points to where the information is to be placed.
1975 The current list of available information is:
1976
1977 PCRE_CONFIG_UTF8
1978
1979 The output is an integer that is set to one if UTF-8 support is available;
1980 otherwise it is set to zero.
1981
1982 PCRE_CONFIG_NEWLINE
1983
1984 The output is an integer that it set to the value of the code that is used for
1985 newline. It is either LF (10) or CR (13).
1986
1987 PCRE_CONFIG_LINK_SIZE
1988
1989 The output is an integer that contains the number of bytes used for internal
1990 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1991
1992 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1993
1994 The output is an integer that contains the threshold above which the POSIX
1995 interface uses malloc() for output vectors. See item 31 above.
1996
1997 PCRE_CONFIG_MATCH_LIMIT
1998
1999 The output is an unsigned integer that contains the default limit of the number
2000 of match() calls in a pcre_exec() execution. See 42 above.
2001
2002 44. pcretest has been upgraded by the addition of the -C option. This causes it
2003 to extract all the available output from the new pcre_config() function, and to
2004 output it. The program then exits immediately.
2005
2006 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2007 order to support additional features. One way would have been to define
2008 pcre_exec2() (for example) with extra arguments, but this would not have been
2009 extensible, and would also have required all calls to the original function to
2010 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2011 is used for passing in "extra" data from pcre_study().
2012
2013 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2014 contains the following fields:
2015
2016 flags a bitmap indicating which of the following fields are set
2017 study_data opaque data from pcre_study()
2018 match_limit a way of specifying a limit on match() calls for a specific
2019 call to pcre_exec()
2020 callout_data data for callouts (see 49 below)
2021
2022 The flag bits are also defined in pcre.h, and are
2023
2024 PCRE_EXTRA_STUDY_DATA
2025 PCRE_EXTRA_MATCH_LIMIT
2026 PCRE_EXTRA_CALLOUT_DATA
2027
2028 The pcre_study() function now returns one of these new pcre_extra blocks, with
2029 the actual study data pointed to by the study_data field, and the
2030 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2031 before. That is, this change is entirely upwards-compatible and requires no
2032 change to existing code.
2033
2034 If you want to pass in additional data to pcre_exec(), you can either place it
2035 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2036 block.
2037
2038 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2039 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2040 times with different match limits, until it finds the minimum value needed for
2041 pcre_exec() to complete. The value is then output. This can be instructive; for
2042 most simple matches the number is quite small, but for pathological cases it
2043 gets very large very quickly.
2044
2045 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2046 returns the size of the data block pointed to by the study_data field in a
2047 pcre_extra block, that is, the value that was passed as the argument to
2048 pcre_malloc() when PCRE was getting memory in which to place the information
2049 created by pcre_study(). The fourth argument should point to a size_t variable.
2050 pcretest has been extended so that this information is shown after a successful
2051 pcre_study() call when information about the compiled regex is being displayed.
2052
2053 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2054 because what follows is always an absolute path. (Later: it turns out that this
2055 is more than cosmetic for MinGW, because it doesn't like empty path
2056 components.)
2057
2058 49. Some changes have been made to the callout feature (see 28 above):
2059
2060 (i) A callout function now has three choices for what it returns:
2061
2062 0 => success, carry on matching
2063 > 0 => failure at this point, but backtrack if possible
2064 < 0 => serious error, return this value from pcre_exec()
2065
2066 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2067 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2068 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2069 use by callout functions. It will never be used by PCRE itself.
2070
2071 (ii) The pcre_extra structure (see 45 above) has a void * field called
2072 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2073 pcre_callout_block structure has a field of the same name. The contents of
2074 the field passed in the pcre_extra structure are passed to the callout
2075 function in the corresponding field in the callout block. This makes it
2076 easier to use the same callout-containing regex from multiple threads. For
2077 testing, the pcretest program has a new data escape
2078
2079 \C*n pass the number n (may be negative) as callout_data
2080
2081 If the callout function in pcretest receives a non-zero value as
2082 callout_data, it returns that value.
2083
2084 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2085 there were some redundant $(CFLAGS) in commands that are now specified as
2086 $(LINK), which already includes $(CFLAGS).
2087
2088 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2089 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2090 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2091 one-byte characters throughout. Note that case-insensitive matching applies
2092 only to characters whose values are less than 256. PCRE doesn't support the
2093 notion of cases for higher-valued characters.
2094
2095 (i) A character class whose characters are all within 0-255 is handled as
2096 a bit map, and the map is inverted for negative classes. Previously, a
2097 character > 255 always failed to match such a class; however it should
2098 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2099
2100 (ii) A negated character class with a single character < 255 is coded as
2101 "not this character" (OP_NOT). This wasn't working properly when the test
2102 character was multibyte, either singly or repeated.
2103
2104 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2105 mode, for example: \x{100}{2,3}.
2106
2107 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2108 singly or repeated) now correctly test multibyte characters. However,
2109 PCRE doesn't recognize any characters with values greater than 255 as
2110 digits, spaces, or word characters. Such characters always match \D, \S,
2111 and \W, and never match \d, \s, or \w.
2112
2113 (v) Classes may now contain characters and character ranges with values
2114 greater than 255. For example: [ab\x{100}-\x{400}].
2115
2116 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2117 PCRE in UTF-8 mode.
2118
2119 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2120 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2121 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2122 value.)
2123
2124 53. The single man page has become too large. I have therefore split it up into
2125 a number of separate man pages. These also give rise to individual HTML pages;
2126 these are now put in a separate directory, and there is an index.html page that
2127 lists them all. Some hyperlinking between the pages has been installed.
2128
2129 54. Added convenience functions for handling named capturing parentheses.
2130
2131 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2132 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2133 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2134 are faulted.
2135
2136 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2137 calling configure. These values are used when compiling the dftables.c program
2138 which is run to generate the source of the default character tables. They
2139 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2140 you will need to set these values.
2141
2142 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2143
2144
2145 Version 3.9 02-Jan-02
2146 ---------------------
2147
2148 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2149
2150 2. If --disable-static was given, the building process failed when trying to
2151 build pcretest and pcregrep. (For some reason it was using libtool to compile
2152 them, which is not right, as they aren't part of the library.)
2153
2154
2155 Version 3.8 18-Dec-01
2156 ---------------------
2157
2158 1. The experimental UTF-8 code was completely screwed up. It was packing the
2159 bytes in the wrong order. How dumb can you get?
2160
2161
2162 Version 3.7 29-Oct-01
2163 ---------------------
2164
2165 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2166 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2167 this didn't happen under Solaris 8, where I normally test things.
2168
2169 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2170 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2171 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2172 this fix an hour or so after the initial 3.7 release.)
2173
2174
2175 Version 3.6 23-Oct-01
2176 ---------------------
2177
2178 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2179 offsets passed as NULL with zero offset count.
2180
2181 2. The config.guess and config.sub files had not been updated when I moved to
2182 the latest autoconf.
2183
2184
2185 Version 3.5 15-Aug-01
2186 ---------------------
2187
2188 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2189 had been forgotten.
2190
2191 2. By using declared but undefined structures, we can avoid using "void"
2192 definitions in pcre.h while keeping the internal definitions of the structures
2193 private.
2194
2195 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2196 user point of view, this means that both static and shared libraries are built
2197 by default, but this can be individually controlled. More of the work of
2198 handling this static/shared cases is now inside libtool instead of PCRE's make
2199 file.
2200
2201 4. The pcretest utility is now installed along with pcregrep because it is
2202 useful for users (to test regexs) and by doing this, it automatically gets
2203 relinked by libtool. The documentation has been turned into a man page, so
2204 there are now .1, .txt, and .html versions in /doc.
2205
2206 5. Upgrades to pcregrep:
2207 (i) Added long-form option names like gnu grep.
2208 (ii) Added --help to list all options with an explanatory phrase.
2209 (iii) Added -r, --recursive to recurse into sub-directories.
2210 (iv) Added -f, --file to read patterns from a file.
2211
2212 6. pcre_exec() was referring to its "code" argument before testing that
2213 argument for NULL (and giving an error if it was NULL).
2214
2215 7. Upgraded Makefile.in to allow for compiling in a different directory from
2216 the source directory.
2217
2218 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2219 options bits, the pointer it was passed was to an int instead of to an unsigned
2220 long int. This mattered only on 64-bit systems.
2221
2222 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2223 generated) instead of pcre.in, which it its source. Also made the same change
2224 in several of the .c files.
2225
2226 10. A new release of gcc defines printf() as a macro, which broke pcretest
2227 because it had an ifdef in the middle of a string argument for printf(). Fixed
2228 by using separate calls to printf().
2229
2230 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2231 script, to force use of CR or LF instead of \n in the source. On non-Unix
2232 systems, the value can be set in config.h.
2233
2234 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2235 absolute limit. Changed the text of the error message to make this clear, and
2236 likewise updated the man page.
2237
2238 13. The limit of 99 on the number of capturing subpatterns has been removed.
2239 The new limit is 65535, which I hope will not be a "real" limit.
2240
2241
2242 Version 3.4 22-Aug-00
2243 ---------------------
2244
2245 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2246
2247 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2248
2249
2250 Version 3.3 01-Aug-00
2251 ---------------------
2252
2253 1. If an octal character was given, but the value was greater than \377, it
2254 was not getting masked to the least significant bits, as documented. This could
2255 lead to crashes in some systems.
2256
2257 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2258 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2259
2260 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2261 These just pass their arguments on to (pcre_free)(), but they are provided
2262 because some uses of PCRE bind it to non-C systems that can call its functions,
2263 but cannot call free() or pcre_free() directly.
2264
2265 4. Add "make test" as a synonym for "make check". Corrected some comments in
2266 the Makefile.
2267
2268 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2269 Makefile.
2270
2271 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2272 command called pgrep for grepping around the active processes.
2273
2274 7. Added the beginnings of support for UTF-8 character strings.
2275
2276 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2277 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2278 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2279 out for the ar command.)
2280
2281
2282 Version 3.2 12-May-00
2283 ---------------------
2284
2285 This is purely a bug fixing release.
2286
2287 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2288 of ZA. This was just one example of several cases that could provoke this bug,
2289 which was introduced by change 9 of version 2.00. The code for breaking
2290 infinite loops after an iteration that matches an empty string was't working
2291 correctly.
2292
2293 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2294 when matched against abbab (for example). After matching an empty string, it
2295 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2296 caused it to match further down the string than it should.
2297
2298 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2299 was there because it doesn't seem to be needed, and it causes trouble on some
2300 systems, as it is not a Standard C header. It has been removed.
2301
2302 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2303 were reported on the Macintosh. The changes were from
2304
2305 while ((c = *(++ptr)) != 0 && c != '\n');
2306 to
2307 while ((c = *(++ptr)) != 0 && c != '\n') ;
2308
2309 Totally extraordinary, but if that's what it takes...
2310
2311 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2312 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2313 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2314 assumes the way PCRE uses memmove() (always moving upwards).
2315
2316 6. PCRE is being used in one environment where strchr() is not available. There
2317 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2318 faster code anyway.
2319
2320
2321 Version 3.1 09-Feb-00
2322 ---------------------
2323
2324 The only change in this release is the fixing of some bugs in Makefile.in for
2325 the "install" target:
2326
2327 (1) It was failing to install pcreposix.h.
2328
2329 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2330
2331
2332 Version 3.0 01-Feb-00
2333 ---------------------
2334
2335 1. Add support for the /+ modifier to perltest (to output $` like it does in
2336 pcretest).
2337
2338 2. Add support for the /g modifier to perltest.
2339
2340 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2341 matches null strings.
2342
2343 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2344 pattern. Perl treats empty patterns specially - it reuses the most recent
2345 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2346 effect.
2347
2348 5. The POSIX interface was broken in that it was just handing over the POSIX
2349 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2350 required a bigger vector, with some working space on the end. This means that
2351 the POSIX wrapper now has to get and free some memory, and copy the results.
2352
2353 6. Added some simple autoconf support, placing the test data and the
2354 documentation in separate directories, re-organizing some of the
2355 information files, and making it build pcre-config (a GNU standard). Also added
2356 libtool support for building PCRE as a shared library, which is now the
2357 default.
2358
2359 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2360 09 are not valid octal constants. Single digits will be used for minor values
2361 less than 10.
2362
2363 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2364 existing programs that set these in the POSIX interface can use PCRE without
2365 modification.
2366
2367 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2368 return all that pcre_info() returns, plus additional data. The pcre_info()
2369 function is retained for compatibility, but is considered to be obsolete.
2370
2371 10. Added experimental recursion feature (?R) to handle one common case that
2372 Perl 5.6 will be able to do with (?p{...}).
2373
2374 11. Added support for POSIX character classes like [:alpha:], which Perl is
2375 adopting.
2376
2377
2378 Version 2.08 31-Aug-99
2379 ----------------------
2380
2381 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2382 trying to match at the startoffset position, but instead was moving forward to
2383 the next newline as if a previous match had failed.
2384
2385 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2386 and could get into a loop if a null string was matched other than at the start
2387 of the subject.
2388
2389 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2390 be distinguished at compile time, and for completeness also added PCRE_DATE.
2391
2392 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2393 in GnuWin32 environments.
2394
2395
2396 Version 2.07 29-Jul-99
2397 ----------------------
2398
2399 1. The documentation is now supplied in plain text form and HTML as well as in
2400 the form of man page sources.
2401
2402 2. C++ compilers don't like assigning (void *) values to other pointer types.
2403 In particular this affects malloc(). Although there is no problem in Standard
2404 C, I've put in casts to keep C++ compilers happy.
2405
2406 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2407 should be (const char *).
2408
2409 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2410 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2411 However, I haven't made this a standard facility. The documentation doesn't
2412 mention it, and the Makefile doesn't support it.
2413
2414 5. The Makefile now contains an "install" target, with editable destinations at
2415 the top of the file. The pcretest program is not installed.
2416
2417 6. pgrep -V now gives the PCRE version number and date.
2418
2419 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2420 causing the entire string to be ignored, instead of just the last character.
2421
2422 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2423 non-matching string, it can take a very, very long time, even for strings of
2424 quite modest length, because of the nested recursion. PCRE now does better in
2425 some of these cases. It does this by remembering the last required literal
2426 character in the pattern, and pre-searching the subject to ensure it is present
2427 before running the real match. In other words, it applies a heuristic to detect
2428 some types of certain failure quickly, and in the above example, if presented
2429 with a string that has no trailing " it gives "no match" very quickly.
2430
2431 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2432 other alternatives are tried instead.
2433
2434
2435 Version 2.06 09-Jun-99
2436 ----------------------
2437
2438 1. Change pcretest's output for amount of store used to show just the code
2439 space, because the remainder (the data block) varies in size between 32-bit and
2440 64-bit systems.
2441
2442 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2443 start matching at. This allows lookbehinds to work when searching for multiple
2444 occurrences in a string.
2445
2446 3. Added additional options to pcretest for testing multiple occurrences:
2447
2448 /+ outputs the rest of the string that follows a match
2449 /g loops for multiple occurrences, using the new startoffset argument
2450 /G loops for multiple occurrences by passing an incremented pointer
2451
2452 4. PCRE wasn't doing the "first character" optimization for patterns starting
2453 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2454 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2455 the letter 'x'. On long subject strings, this gives a significant speed-up.
2456
2457
2458 Version 2.05 21-Apr-99
2459 ----------------------
2460
2461 1. Changed the type of magic_number from int to long int so that it works
2462 properly on 16-bit systems.
2463
2464 2. Fixed a bug which caused patterns starting with .* not to work correctly
2465 when the subject string contained newline characters. PCRE was assuming
2466 anchoring for such patterns in all cases, which is not correct because .* will
2467 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2468 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2469 must be retried after every newline in the subject.
2470
2471
2472 Version 2.04 18-Feb-99
2473 ----------------------
2474
2475 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2476 computation of the store needed to hold the pattern was incorrect (too large).
2477 If such patterns were nested a few deep, this could multiply and become a real
2478 problem.
2479
2480 2. Added /M option to pcretest to show the memory requirement of a specific
2481 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2482
2483 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2484 compiled in such a way that the backtracking after subsequent failure was
2485 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2486 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2487
2488
2489 Version 2.03 02-Feb-99
2490 ----------------------
2491
2492 1. Fixed typo and small mistake in man page.
2493
2494 2. Added 4th condition (GPL supersedes if conflict) and created separate
2495 LICENCE file containing the conditions.
2496
2497 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2498 Perl, that is the internal \ allows the delimiter to be included in the
2499 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2500 the final delimiter, add \ to the end of the pattern (to test the error).
2501
2502 4. Added the convenience functions for extracting substrings after a successful
2503 match. Updated pcretest to make it able to test these functions.
2504
2505
2506 Version 2.02 14-Jan-99
2507 ----------------------
2508
2509 1. Initialized the working variables associated with each extraction so that
2510 their saving and restoring doesn't refer to uninitialized store.
2511
2512 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2513 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2514 fix the problem.
2515
2516 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2517 calls, and wasn't printing the correct value for compiling calls. Increased the
2518 default value of LOOPREPEAT, and the number of significant figures in the
2519 times.
2520
2521 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2522
2523 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2524 a building problem on Windows NT with a FAT file system.
2525
2526
2527 Version 2.01 21-Oct-98
2528 ----------------------
2529
2530 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2531 to character tables built by pcre_maketables() in the current locale. If NULL
2532 is passed, the default tables are used.
2533
2534
2535 Version 2.00 24-Sep-98
2536 ----------------------
2537
2538 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2539 it any more.
2540
2541 2. Allow quantification of (?>) groups, and make it work correctly.
2542
2543 3. The first character computation wasn't working for (?>) groups.
2544
2545 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2546 end of the subject) and add 5.005's \z, which really does match only at the
2547 very end of the subject.
2548
2549 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2550
2551 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2552 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2553 localized options. All options to pcre_study() were also removed.
2554
2555 7. Add other new features from 5.005:
2556
2557 $(?<= positive lookbehind
2558 $(?<! negative lookbehind
2559 (?imsx-imsx) added the unsetting capability
2560 such a setting is global if at outer level; local otherwise
2561 (?imsx-imsx:) non-capturing groups with option setting
2562 (?(cond)re|re) conditional pattern matching
2563
2564 A backreference to itself in a repeated group matches the previous
2565 captured string.
2566
2567 8. General tidying up of studying (both automatic and via "study")
2568 consequential on the addition of new assertions.
2569
2570 9. As in 5.005, unlimited repeated groups that could match an empty substring
2571 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2572 runtime if any iteration does actually match an empty substring.
2573
2574 10. Include the RunTest script in the distribution.
2575
2576 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2577 discrepancies, some of which were old and were also with respect to 5.004. They
2578 have now been fixed.
2579
2580
2581 Version 1.09 28-Apr-98
2582 ----------------------
2583
2584 1. A negated single character class followed by a quantifier with a minimum
2585 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2586 program crashes, or just wrong answers. This did not apply to negated classes
2587 containing more than one character, or to minima other than one.
2588
2589
2590 Version 1.08 27-Mar-98
2591 ----------------------
2592
2593 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2594
2595 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2596 latter must appear before anything that relies on it in the pattern.
2597
2598
2599 Version 1.07 16-Feb-98
2600 ----------------------
2601
2602 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2603 repeat of a potentially empty string).
2604
2605
2606 Version 1.06 23-Jan-98
2607 ----------------------
2608
2609 1. Added Markus Oberhumer's little patches for C++.
2610
2611 2. Literal strings longer than 255 characters were broken.
2612
2613
2614 Version 1.05 23-Dec-97
2615 ----------------------
2616
2617 1. Negated character classes containing more than one character were failing if
2618 PCRE_CASELESS was set at run time.
2619
2620
2621 Version 1.04 19-Dec-97
2622 ----------------------
2623
2624 1. Corrected the man page, where some "const" qualifiers had been omitted.
2625
2626 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2627 input syntax.
2628
2629 3. Fixed memory leak which occurred when a regex with back references was
2630 matched with an offsets vector that wasn't big enough. The temporary memory
2631 that is used in this case wasn't being freed if the match failed.
2632
2633 4. Tidied pcretest to ensure it frees memory that it gets.
2634
2635 5. Temporary memory was being obtained in the case where the passed offsets
2636 vector was exactly big enough.
2637
2638 6. Corrected definition of offsetof() from change 5 below.
2639
2640 7. I had screwed up change 6 below and broken the rules for the use of
2641 setjmp(). Now fixed.
2642
2643
2644 Version 1.03 18-Dec-97
2645 ----------------------
2646
2647 1. A erroneous regex with a missing opening parenthesis was correctly
2648 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2649 on some systems.
2650
2651 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2652 it was reported that one broken compiler failed on the former because "code" is
2653 also an independent variable.
2654
2655 3. The erroneous regex a[]b caused an array overrun reference.
2656
2657 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2658 fail on data ending with that character. (It was going on too far, and checking
2659 the next character, typically a binary zero.) This was specific to the
2660 optimized code for single-character negative classes.
2661
2662 5. Added a contributed patch from the TIN world which does the following:
2663
2664 + Add an undef for memmove, in case the the system defines a macro for it.
2665
2666 + Add a definition of offsetof(), in case there isn't one. (I don't know
2667 the reason behind this - offsetof() is part of the ANSI standard - but
2668 it does no harm).
2669
2670 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2671 most of the places where whitespace preceded '#'. I have given up and
2672 allowed the remaining 2 cases to be at the margin.
2673
2674 + Rename some variables in pcre to eliminate shadowing. This seems very
2675 pedantic, but does no harm, of course.
2676
2677 6. Moved the call to setjmp() into its own function, to get rid of warnings
2678 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2679
2680 7. Constructs such as \d{8,} were compiling into the equivalent of
2681 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2682 outcome, but in this particular case used more store than had been allocated,
2683 which caused the bug to be discovered because it threw up an internal error.
2684
2685 8. The debugging code in both pcre and pcretest for outputting the compiled
2686 form of a regex was going wrong in the case of back references followed by
2687 curly-bracketed repeats.
2688
2689
2690 Version 1.02 12-Dec-97
2691 ----------------------
2692
2693 1. Typos in pcre.3 and comments in the source fixed.
2694
2695 2. Applied a contributed patch to get rid of places where it used to remove
2696 'const' from variables, and fixed some signed/unsigned and uninitialized
2697 variable warnings.
2698
2699 3. Added the "runtest" target to Makefile.
2700
2701 4. Set default compiler flag to -O2 rather than just -O.
2702
2703
2704 Version 1.01 19-Nov-97
2705 ----------------------
2706
2707 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2708 like /([ab]*)*/, that is, for classes with more than one character in them.
2709
2710 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2711 as /((?>a*))*/ (a PCRE_EXTRA facility).
2712
2713
2714 Version 1.00 18-Nov-97
2715 ----------------------
2716
2717 1. Added compile-time macros to support systems such as SunOS4 which don't have
2718 memmove() or strerror() but have other things that can be used instead.
2719
2720 2. Arranged that "make clean" removes the executables.
2721
2722
2723 Version 0.99 27-Oct-97
2724 ----------------------
2725
2726 1. Fixed bug in code for optimizing classes with only one character. It was
2727 initializing a 32-byte map regardless, which could cause it to run off the end
2728 of the memory it had got.
2729
2730 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2731
2732
2733 Version 0.98 22-Oct-97
2734 ----------------------
2735
2736 1. Fixed bug in code for handling temporary memory usage when there are more
2737 back references than supplied space in the ovector. This could cause segfaults.
2738
2739
2740 Version 0.97 21-Oct-97
2741 ----------------------
2742
2743 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2744
2745 2. Optimized negated single characters not to use a bit map.
2746
2747 3. Brought error texts together as macro definitions; clarified some of them;
2748 fixed one that was wrong - it said "range out of order" when it meant "invalid
2749 escape sequence".
2750
2751 4. Changed some char * arguments to const char *.
2752
2753 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2754
2755 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2756 pcretest.
2757
2758
2759 Version 0.96 16-Oct-97
2760 ----------------------
2761
2762 1. Added a simple "pgrep" utility to the distribution.
2763
2764 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2765 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2766 where "ddd" means "one or more decimal digits".
2767
2768 3. Fixed serious bug. If a pattern had a back reference, but the call to
2769 pcre_exec() didn't supply a large enough ovector to record the related
2770 identifying subpattern, the match always failed. PCRE now remembers the number
2771 of the largest back reference, and gets some temporary memory in which to save
2772 the offsets during matching if necessary, in order to ensure that
2773 backreferences always work.
2774
2775 4. Increased the compatibility with Perl in a number of ways:
2776
2777 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2778 to request this handling. The option can be set at compile or exec time.
2779
2780 (b) $ matches before a terminating newline by default; an option
2781 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2782 mode). The option can be set at compile or exec time.
2783
2784 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2785 the same as Perl's. If the decimal number it represents is less than 10
2786 or there aren't that many previous left capturing parentheses, an octal
2787 escape is read. Inside a character class, it's always an octal escape,
2788 even if it is a single digit.
2789
2790 (d) An escaped but undefined alphabetic character is taken as a literal,
2791 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2792 escapes.
2793
2794 (e) {0} is now permitted. (The previous item is removed from the compiled
2795 pattern).
2796
2797 5. Changed all the names of code files so that the basic parts are no longer
2798 than 10 characters, and abolished the teeny "globals.c" file.
2799
2800 6. Changed the handling of character classes; they are now done with a 32-byte
2801 bit map always.
2802
2803 7. Added the -d and /D options to pcretest to make it possible to look at the
2804 internals of compilation without having to recompile pcre.
2805
2806
2807 Version 0.95 23-Sep-97
2808 ----------------------
2809
2810 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2811 \x20 at the start of a run of normal characters. These were being treated as
2812 real characters, instead of the source characters being re-checked.
2813
2814
2815 Version 0.94 18-Sep-97
2816 ----------------------
2817
2818 1. The functions are now thread-safe, with the caveat that the global variables
2819 containing pointers to malloc() and free() or alternative functions are the
2820 same for all threads.
2821
2822 2. Get pcre_study() to generate a bitmap of initial characters for non-
2823 anchored patterns when this is possible, and use it if passed to pcre_exec().
2824
2825
2826 Version 0.93 15-Sep-97
2827 ----------------------
2828
2829 1. /(b)|(:+)/ was computing an incorrect first character.
2830
2831 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2832 but not actually doing anything yet.
2833
2834 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2835 as Perl does (e.g. [-az] or [az-]).
2836
2837 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2838 all possible positions.
2839
2840 5. Split up into different modules to avoid including unneeded functions in a
2841 compiled binary. However, compile and exec are still in one module. The "study"
2842 function is split off.
2843
2844 6. The character tables are now in a separate module whose source is generated
2845 by an auxiliary program - but can then be edited by hand if required. There are
2846 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2847 toupper() in the code.
2848
2849 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2850 make them global. Abolish the function for setting them, as the caller can now
2851 set them directly.
2852
2853
2854 Version 0.92 11-Sep-97
2855 ----------------------
2856
2857 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2858 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2859
2860 2. Caseless matching was not working in character classes if the characters in
2861 the pattern were in upper case.
2862
2863 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2864
2865 4. Make PCRE_ANCHORED public and accept as a compile option.
2866
2867 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2868 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2869 pass them.
2870
2871 6. Give an error if bad option bits passed at compile or run time.
2872
2873 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2874 pcretest to cause it to pass that flag.
2875
2876 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2877 options, and the first character, if set.
2878
2879 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2880
2881
2882 Version 0.91 10-Sep-97
2883 ----------------------
2884
2885 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2886 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2887
2888 2. PCRE was looping on encountering an indefinitely repeated back reference to
2889 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2890 Perl does - treats the match as successful.
2891
2892 ****

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