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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 224 - (show annotations) (download)
Mon Aug 20 12:58:29 2007 UTC (7 years, 3 months ago) by ph10
File size: 137674 byte(s)
Fix loop for group with possible zero repeat containing \p or \P.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.3 20-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 21. An orphan \E inside a character class could cause a crash.
121
122 22. A repeated capturing bracket such as (A)? could cause a wild memory
123 reference during compilation.
124
125 23. There are several functions in pcre_compile() that scan along a compiled
126 expression for various reasons (e.g. to see if it's fixed length for look
127 behind). There were bugs in these functions when a repeated \p or \P was
128 present in the pattern. These operators have additional parameters compared
129 with \d, etc, and these were not being taken into account when moving along
130 the compiled data. Specifically:
131
132 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
133 length.
134
135 (b) An item such as \pL+ within a repeated group could cause crashes or
136 loops.
137
138 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
139 "reference to non-existent subpattern" error.
140
141 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
142
143 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
144 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
145
146 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
147
148 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
149 character were causing crashes (broken optimization).
150
151 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
152 \p or \P) caused a compile-time loop.
153
154
155 Version 7.2 19-Jun-07
156 ---------------------
157
158 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
159 which is apparently normally available under Windows.
160
161 2. Re-jig the pcregrep tests with different newline settings in an attempt
162 to make them independent of the local environment's newline setting.
163
164 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
165
166 4. Some of the "internals" tests were previously cut out when the link size
167 was not 2, because the output contained actual offsets. The recent new
168 "Z" feature of pcretest means that these can be cut out, making the tests
169 usable with all link sizes.
170
171 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
172 stack recursion. This gives a massive performance boost under BSD, but just
173 a small improvement under Linux. However, it saves one field in the frame
174 in all cases.
175
176 6. Added more features from the forthcoming Perl 5.10:
177
178 (a) (?-n) (where n is a string of digits) is a relative subroutine or
179 recursion call. It refers to the nth most recently opened parentheses.
180
181 (b) (?+n) is also a relative subroutine call; it refers to the nth next
182 to be opened parentheses.
183
184 (c) Conditions that refer to capturing parentheses can be specified
185 relatively, for example, (?(-2)... or (?(+3)...
186
187 (d) \K resets the start of the current match so that everything before
188 is not part of it.
189
190 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
191
192 (f) \g{name} is another synonym - part of Perl 5.10's unification of
193 reference syntax.
194
195 (g) (?| introduces a group in which the numbering of parentheses in each
196 alternative starts with the same number.
197
198 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
199
200 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
201 PCRE_INFO_JCHANGED.
202
203 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
204 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
205 for detecting groups that can match an empty string.
206
207 9. A pattern with a very large number of alternatives (more than several
208 hundred) was running out of internal workspace during the pre-compile
209 phase, where pcre_compile() figures out how much memory will be needed. A
210 bit of new cunning has reduced the workspace needed for groups with
211 alternatives. The 1000-alternative test pattern now uses 12 bytes of
212 workspace instead of running out of the 4096 that are available.
213
214 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
215
216 11. Applied patch from Google to remove an optimization that didn't quite work.
217 The report of the bug said:
218
219 pcrecpp::RE("a*").FullMatch("aaa") matches, while
220 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
221 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
222
223 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
224 it matched the wrong number of bytes.
225
226
227 Version 7.1 24-Apr-07
228 ---------------------
229
230 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
231 that is more "standard", making use of automake and other Autotools. There
232 is some re-arrangement of the files and adjustment of comments consequent
233 on this.
234
235 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
236 for recursive directory scanning broke on some systems because the files
237 are not scanned in any specific order and on different systems the order
238 was different. A call to "sort" has been inserted into RunGrepTest for the
239 approprate test as a short-term fix. In the longer term there may be an
240 alternative.
241
242 3. I had an email from Eric Raymond about problems translating some of PCRE's
243 man pages to HTML (despite the fact that I distribute HTML pages, some
244 people do their own conversions for various reasons). The problems
245 concerned the use of low-level troff macros .br and .in. I have therefore
246 removed all such uses from the man pages (some were redundant, some could
247 be replaced by .nf/.fi pairs). The 132html script that I use to generate
248 HTML has been updated to handle .nf/.fi and to complain if it encounters
249 .br or .in.
250
251 4. Updated comments in configure.ac that get placed in config.h.in and also
252 arranged for config.h to be included in the distribution, with the name
253 config.h.generic, for the benefit of those who have to compile without
254 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
255
256 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
257 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
258 makevp.bat for the current PCRE, using the additional files
259 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
260
261 6. A Windows user reported a minor discrepancy with test 2, which turned out
262 to be caused by a trailing space on an input line that had got lost in his
263 copy. The trailing space was an accident, so I've just removed it.
264
265 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
266 that is needed.
267
268 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
269 as "const" (a) because they are and (b) because it helps the PHP
270 maintainers who have recently made a script to detect big data structures
271 in the php code that should be moved to the .rodata section. I remembered
272 to update Builducptable as well, so it won't revert if ucptable.h is ever
273 re-created.
274
275 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
276 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
277 order to be able to cut out the UTF-8 tables in the latter when UTF-8
278 support is not required. This saves 1.5-2K of code, which is important in
279 some applications.
280
281 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
282 so as not to refer to the tables, even though these functions will never be
283 called when UTF-8 support is disabled. Otherwise there are problems with a
284 shared library.
285
286 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
287
288 (a) It was defining its arguments as char * instead of void *.
289
290 (b) It was assuming that all moves were upwards in memory; this was true
291 a long time ago when I wrote it, but is no longer the case.
292
293 The emulated memove() is provided for those environments that have neither
294 memmove() nor bcopy(). I didn't think anyone used it these days, but that
295 is clearly not the case, as these two bugs were recently reported.
296
297 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
298 and Detrail to create the HTML documentation, the .txt form of the man
299 pages, and it removes trailing spaces from listed files. It also creates
300 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
301 case, it wraps all the #defines with #ifndefs. This script should be run
302 before "make dist".
303
304 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
305 with Unicode property support.
306
307 (a) For a maximizing quantifier, if the two different cases of the
308 character were of different lengths in their UTF-8 codings (there are
309 some cases like this - I found 11), and the matching function had to
310 back up over a mixture of the two cases, it incorrectly assumed they
311 were both the same length.
312
313 (b) When PCRE was configured to use the heap rather than the stack for
314 recursion during matching, it was not correctly preserving the data for
315 the other case of a UTF-8 character when checking ahead for a match
316 while processing a minimizing repeat. If the check also involved
317 matching a wide character, but failed, corruption could cause an
318 erroneous result when trying to check for a repeat of the original
319 character.
320
321 13. Some tidying changes to the testing mechanism:
322
323 (a) The RunTest script now detects the internal link size and whether there
324 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
325 values substituted by "configure". (The RunGrepTest script already did
326 this for UTF-8.) The configure.ac script no longer substitutes the
327 relevant variables.
328
329 (b) The debugging options /B and /D in pcretest show the compiled bytecode
330 with length and offset values. This means that the output is different
331 for different internal link sizes. Test 2 is skipped for link sizes
332 other than 2 because of this, bypassing the problem. Unfortunately,
333 there was also a test in test 3 (the locale tests) that used /B and
334 failed for link sizes other than 2. Rather than cut the whole test out,
335 I have added a new /Z option to pcretest that replaces the length and
336 offset values with spaces. This is now used to make test 3 independent
337 of link size. (Test 2 will be tidied up later.)
338
339 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
340 segmentation fault instead of returning the appropriate error message.
341
342 15. In multiline mode when the newline sequence was set to "any", the pattern
343 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
344 This doesn't seem right; it now treats the CRLF combination as the line
345 ending, and so does not match in that case. It's only a pattern such as ^$
346 that would hit this one: something like ^ABC$ would have failed after \r
347 and then tried again after \r\n.
348
349 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
350 in an attempt to make files that differ only in their line terminators
351 compare equal. This works on Linux.
352
353 17. Under certain error circumstances pcregrep might try to free random memory
354 as it exited. This is now fixed, thanks to valgrind.
355
356 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
357 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
358 was because its rules for how to advance for /g after matching an empty
359 string at the end of a line did not allow for this case. They now check for
360 it specially.
361
362 20. pcretest is supposed to handle patterns and data of any length, by
363 extending its buffers when necessary. It was getting this wrong when the
364 buffer for a data line had to be extended.
365
366 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
367 CRLF as a newline sequence.
368
369 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
370 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
371 I have nevertheless tidied it up.
372
373 23. Added some casts to kill warnings from HP-UX ia64 compiler.
374
375 24. Added a man page for pcre-config.
376
377
378 Version 7.0 19-Dec-06
379 ---------------------
380
381 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
382 moving to gcc 4.1.1.
383
384 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
385 sys/time.h, which is documented as needed for this function. It doesn't
386 seem to matter on Linux, but it showed up on some releases of OS X.
387
388 3. It seems that there are systems where bytes whose values are greater than
389 127 match isprint() in the "C" locale. The "C" locale should be the
390 default when a C program starts up. In most systems, only ASCII printing
391 characters match isprint(). This difference caused the output from pcretest
392 to vary, making some of the tests fail. I have changed pcretest so that:
393
394 (a) When it is outputting text in the compiled version of a pattern, bytes
395 other than 32-126 are always shown as hex escapes.
396
397 (b) When it is outputting text that is a matched part of a subject string,
398 it does the same, unless a different locale has been set for the match
399 (using the /L modifier). In this case, it uses isprint() to decide.
400
401 4. Fixed a major bug that caused incorrect computation of the amount of memory
402 required for a compiled pattern when options that changed within the
403 pattern affected the logic of the preliminary scan that determines the
404 length. The relevant options are -x, and -i in UTF-8 mode. The result was
405 that the computed length was too small. The symptoms of this bug were
406 either the PCRE error "internal error: code overflow" from pcre_compile(),
407 or a glibc crash with a message such as "pcretest: free(): invalid next
408 size (fast)". Examples of patterns that provoked this bug (shown in
409 pcretest format) are:
410
411 /(?-x: )/x
412 /(?x)(?-x: \s*#\s*)/
413 /((?i)[\x{c0}])/8
414 /(?i:[\x{c0}])/8
415
416 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
417 is now done differently.
418
419 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
420 wrapper classes; (b) implement a new function in the C++ scanner that is
421 more efficient than the old way of doing things because it avoids levels of
422 recursion in the regex matching; (c) add a paragraph to the documentation
423 for the FullMatch() function.
424
425 6. The escape sequence \n was being treated as whatever was defined as
426 "newline". Not only was this contrary to the documentation, which states
427 that \n is character 10 (hex 0A), but it also went horribly wrong when
428 "newline" was defined as CRLF. This has been fixed.
429
430 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
431 was being set to -1 for the "end of line" case (supposedly a value that no
432 character can have). Though this value is never used (the check for end of
433 line is "zero bytes in current character"), it caused compiler complaints.
434 I've changed it to 0xffffffff.
435
436 8. In pcre_version.c, the version string was being built by a sequence of
437 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
438 string (as it is for production releases) called a macro with an empty
439 argument. The C standard says the result of this is undefined. The gcc
440 compiler treats it as an empty string (which was what was wanted) but it is
441 reported that Visual C gives an error. The source has been hacked around to
442 avoid this problem.
443
444 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
445 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
446 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
447 of them did).
448
449 10. Originally, pcretest opened its input and output without "b"; then I was
450 told that "b" was needed in some environments, so it was added for release
451 5.0 to both the input and output. (It makes no difference on Unix-like
452 systems.) Later I was told that it is wrong for the input on Windows. I've
453 now abstracted the modes into two macros, to make it easier to fiddle with
454 them, and removed "b" from the input mode under Windows.
455
456 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
457
458 12. Added -help and --help to pcretest as an official way of being reminded
459 of the options.
460
461 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
462 and pcrecpp.cc because they annoy compilers at high warning levels.
463
464 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
465
466 15. Fixed an occurrence of == in configure.ac that should have been = (shell
467 scripts are not C programs :-) and which was not noticed because it works
468 on Linux.
469
470 16. pcretest is supposed to handle any length of pattern and data line (as one
471 line or as a continued sequence of lines) by extending its input buffer if
472 necessary. This feature was broken for very long pattern lines, leading to
473 a string of junk being passed to pcre_compile() if the pattern was longer
474 than about 50K.
475
476 17. I have done a major re-factoring of the way pcre_compile() computes the
477 amount of memory needed for a compiled pattern. Previously, there was code
478 that made a preliminary scan of the pattern in order to do this. That was
479 OK when PCRE was new, but as the facilities have expanded, it has become
480 harder and harder to keep it in step with the real compile phase, and there
481 have been a number of bugs (see for example, 4 above). I have now found a
482 cunning way of running the real compile function in a "fake" mode that
483 enables it to compute how much memory it would need, while actually only
484 ever using a few hundred bytes of working memory and without too many
485 tests of the mode. This should make future maintenance and development
486 easier. A side effect of this work is that the limit of 200 on the nesting
487 depth of parentheses has been removed (though this was never a serious
488 limitation, I suspect). However, there is a downside: pcre_compile() now
489 runs more slowly than before (30% or more, depending on the pattern). I
490 hope this isn't a big issue. There is no effect on runtime performance.
491
492 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
493 newline (only possible for the last line of a file) and it was a
494 pattern that set a locale (followed by /Lsomething), pcretest crashed.
495
496 19. Added additional timing features to pcretest. (1) The -tm option now times
497 matching only, not compiling. (2) Both -t and -tm can be followed, as a
498 separate command line item, by a number that specifies the number of
499 repeats to use when timing. The default is 50000; this gives better
500 precision, but takes uncomfortably long for very large patterns.
501
502 20. Extended pcre_study() to be more clever in cases where a branch of a
503 subpattern has no definite first character. For example, (a*|b*)[cd] would
504 previously give no result from pcre_study(). Now it recognizes that the
505 first character must be a, b, c, or d.
506
507 21. There was an incorrect error "recursive call could loop indefinitely" if
508 a subpattern (or the entire pattern) that was being tested for matching an
509 empty string contained only one non-empty item after a nested subpattern.
510 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
511 incorrectly, because the \d was being skipped in the check.
512
513 22. The pcretest program now has a new pattern option /B and a command line
514 option -b, which is equivalent to adding /B to every pattern. This causes
515 it to show the compiled bytecode, without the additional information that
516 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
517 is the same as /B/I).
518
519 23. A new optimization is now able automatically to treat some sequences such
520 as a*b as a*+b. More specifically, if something simple (such as a character
521 or a simple class like \d) has an unlimited quantifier, and is followed by
522 something that cannot possibly match the quantified thing, the quantifier
523 is automatically "possessified".
524
525 24. A recursive reference to a subpattern whose number was greater than 39
526 went wrong under certain circumstances in UTF-8 mode. This bug could also
527 have affected the operation of pcre_study().
528
529 25. Realized that a little bit of performance could be had by replacing
530 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
531
532 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
533
534 27. Possessive quantifiers such as a++ were previously implemented by turning
535 them into atomic groups such as ($>a+). Now they have their own opcodes,
536 which improves performance. This includes the automatically created ones
537 from 23 above.
538
539 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
540 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
541 the first matched character to be a colon. This applied both to named and
542 numbered groups.
543
544 29. The ucpinternal.h header file was missing its idempotency #ifdef.
545
546 30. I was sent a "project" file called libpcre.a.dev which I understand makes
547 building PCRE on Windows easier, so I have included it in the distribution.
548
549 31. There is now a check in pcretest against a ridiculously large number being
550 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
551 loop, the loop is abandoned.
552
553 32. Forward references to subpatterns in conditions such as (?(2)...) where
554 subpattern 2 is defined later cause pcre_compile() to search forwards in
555 the pattern for the relevant set of parentheses. This search went wrong
556 when there were unescaped parentheses in a character class, parentheses
557 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
558
559 33. "Subroutine" calls and backreferences were previously restricted to
560 referencing subpatterns earlier in the regex. This restriction has now
561 been removed.
562
563 34. Added a number of extra features that are going to be in Perl 5.10. On the
564 whole, these are just syntactic alternatives for features that PCRE had
565 previously implemented using the Python syntax or my own invention. The
566 other formats are all retained for compatibility.
567
568 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
569 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
570 also .NET compatible.
571
572 (b) A recursion or subroutine call to a named group can now be defined as
573 (?&name) as well as (?P>name).
574
575 (c) A backreference to a named group can now be defined as \k<name> or
576 \k'name' as well as (?P=name). The new forms, as well as being in Perl
577 5.10, are also .NET compatible.
578
579 (d) A conditional reference to a named group can now use the syntax
580 (?(<name>) or (?('name') as well as (?(name).
581
582 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
583 groups (named and numbered) that are never evaluated inline, but can be
584 called as "subroutines" from elsewhere. In effect, the DEFINE condition
585 is always false. There may be only one alternative in such a group.
586
587 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
588 as the simple (?(R). The condition is true only if the most recent
589 recursion is that of the given number or name. It does not search out
590 through the entire recursion stack.
591
592 (g) The escape \gN or \g{N} has been added, where N is a positive or
593 negative number, specifying an absolute or relative reference.
594
595 35. Tidied to get rid of some further signed/unsigned compiler warnings and
596 some "unreachable code" warnings.
597
598 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
599 things, this adds five new scripts.
600
601 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
602 There were also incompatibilities regarding the handling of \Q..\E inside
603 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
604 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
605
606 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
607 matches an empty string, and forcibly breaks the loop. There were bugs in
608 this code in non-simple cases. For a pattern such as ^(a()*)* matched
609 against aaaa the result was just "a" rather than "aaaa", for example. Two
610 separate and independent bugs (that affected different cases) have been
611 fixed.
612
613 39. Refactored the code to abolish the use of different opcodes for small
614 capturing bracket numbers. This is a tidy that I avoided doing when I
615 removed the limit on the number of capturing brackets for 3.5 back in 2001.
616 The new approach is not only tidier, it makes it possible to reduce the
617 memory needed to fix the previous bug (38).
618
619 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
620 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
621 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
622 mode.
623
624 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
625 report.
626
627 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
628 copy construction and assignment in the C++ wrapper.
629
630 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
631 couple of bugs that could have given wrong results in the "--newline=crlf"
632 case.
633
634 44. Added a number of casts and did some reorganization of signed/unsigned int
635 variables following suggestions from Dair Grant. Also renamed the variable
636 "this" as "item" because it is a C++ keyword.
637
638 45. Arranged for dftables to add
639
640 #include "pcre_internal.h"
641
642 to pcre_chartables.c because without it, gcc 4.x may remove the array
643 definition from the final binary if PCRE is built into a static library and
644 dead code stripping is activated.
645
646 46. For an unanchored pattern, if a match attempt fails at the start of a
647 newline sequence, and the newline setting is CRLF or ANY, and the next two
648 characters are CRLF, advance by two characters instead of one.
649
650
651 Version 6.7 04-Jul-06
652 ---------------------
653
654 1. In order to handle tests when input lines are enormously long, pcretest has
655 been re-factored so that it automatically extends its buffers when
656 necessary. The code is crude, but this _is_ just a test program. The
657 default size has been increased from 32K to 50K.
658
659 2. The code in pcre_study() was using the value of the re argument before
660 testing it for NULL. (Of course, in any sensible call of the function, it
661 won't be NULL.)
662
663 3. The memmove() emulation function in pcre_internal.h, which is used on
664 systems that lack both memmove() and bcopy() - that is, hardly ever -
665 was missing a "static" storage class specifier.
666
667 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
668 containing an extended class (one that cannot be represented by a bitmap
669 because it contains high-valued characters or Unicode property items, e.g.
670 [\pZ]). Almost always one would set UTF-8 mode when processing such a
671 pattern, but PCRE should not loop if you do not (it no longer does).
672 [Detail: two cases were found: (a) a repeated subpattern containing an
673 extended class; (b) a recursive reference to a subpattern that followed a
674 previous extended class. It wasn't skipping over the extended class
675 correctly when UTF-8 mode was not set.]
676
677 5. A negated single-character class was not being recognized as fixed-length
678 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
679 compile error "lookbehind assertion is not fixed length".
680
681 6. The RunPerlTest auxiliary script was showing an unexpected difference
682 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
683 write a Perl script that can interpret lines of an input file either as
684 byte characters or as UTF-8, which is what "perltest" was being required to
685 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
686 can't do is switch easily at run time between having the "use utf8;" pragma
687 or not. In the end, I fudged it by using the RunPerlTest script to insert
688 "use utf8;" explicitly for the UTF-8 tests.
689
690 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
691 the end of the subject string, contrary to the documentation and to what
692 Perl does. This was true of both matching functions. Now it matches only at
693 the start of the subject and immediately after *internal* newlines.
694
695 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
696 a pointer to an int instead of a pointer to an unsigned long int. This
697 caused problems on 64-bit systems.
698
699 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
700 instance of the 'standard' template library not being so standard".
701
702 10. There was no check on the number of named subpatterns nor the maximum
703 length of a subpattern name. The product of these values is used to compute
704 the size of the memory block for a compiled pattern. By supplying a very
705 long subpattern name and a large number of named subpatterns, the size
706 computation could be caused to overflow. This is now prevented by limiting
707 the length of names to 32 characters, and the number of named subpatterns
708 to 10,000.
709
710 11. Subpatterns that are repeated with specific counts have to be replicated in
711 the compiled pattern. The size of memory for this was computed from the
712 length of the subpattern and the repeat count. The latter is limited to
713 65535, but there was no limit on the former, meaning that integer overflow
714 could in principle occur. The compiled length of a repeated subpattern is
715 now limited to 30,000 bytes in order to prevent this.
716
717 12. Added the optional facility to have named substrings with the same name.
718
719 13. Added the ability to use a named substring as a condition, using the
720 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
721 are numbers (not recommended). Forward references are permitted.
722
723 14. Added forward references in named backreferences (if you see what I mean).
724
725 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
726 pattern could run off the end of the subject. For example, the pattern
727 "(?s)(.{1,5})"8 did this with the subject "ab".
728
729 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
730 PCRE_CASELESS was set when matching characters that were quantified with ?
731 or *.
732
733 17. A character class other than a single negated character that had a minimum
734 but no maximum quantifier - for example [ab]{6,} - was not handled
735 correctly by pce_dfa_exec(). It would match only one character.
736
737 18. A valid (though odd) pattern that looked like a POSIX character
738 class but used an invalid character after [ (for example [[,abc,]]) caused
739 pcre_compile() to give the error "Failed: internal error: code overflow" or
740 in some cases to crash with a glibc free() error. This could even happen if
741 the pattern terminated after [[ but there just happened to be a sequence of
742 letters, a binary zero, and a closing ] in the memory that followed.
743
744 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
745 over the years. Originally (before any Unicode support), just the bottom 8
746 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
747 output from "man perlunicode" includes this:
748
749 The regular expression compiler produces polymorphic opcodes. That
750 is, the pattern adapts to the data and automatically switches to
751 the Unicode character scheme when presented with Unicode data--or
752 instead uses a traditional byte scheme when presented with byte
753 data.
754
755 Sadly, a wide octal escape does not cause a switch, and in a string with
756 no other multibyte characters, these octal escapes are treated as before.
757 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
758 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
759 Unicode string.
760
761 I have not perpetrated such confusion in PCRE. Up till now, it took just
762 the bottom 8 bits, as in old Perl. I have now made octal escapes with
763 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
764 translate to the appropriate multibyte character.
765
766 29. Applied some refactoring to reduce the number of warnings from Microsoft
767 and Borland compilers. This has included removing the fudge introduced
768 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
769 a warning about an unused variable.
770
771 21. PCRE has not included VT (character 0x0b) in the set of whitespace
772 characters since release 4.0, because Perl (from release 5.004) does not.
773 [Or at least, is documented not to: some releases seem to be in conflict
774 with the documentation.] However, when a pattern was studied with
775 pcre_study() and all its branches started with \s, PCRE still included VT
776 as a possible starting character. Of course, this did no harm; it just
777 caused an unnecessary match attempt.
778
779 22. Removed a now-redundant internal flag bit that recorded the fact that case
780 dependency changed within the pattern. This was once needed for "required
781 byte" processing, but is no longer used. This recovers a now-scarce options
782 bit. Also moved the least significant internal flag bit to the most-
783 significant bit of the word, which was not previously used (hangover from
784 the days when it was an int rather than a uint) to free up another bit for
785 the future.
786
787 23. Added support for CRLF line endings as well as CR and LF. As well as the
788 default being selectable at build time, it can now be changed at runtime
789 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
790 specify that it is scanning data with non-default line endings.
791
792 24. Changed the definition of CXXLINK to make it agree with the definition of
793 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
794
795 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
796 recursions. This makes a big different to stack usage for some patterns.
797
798 26. If a subpattern containing a named recursion or subroutine reference such
799 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
800 the space required for the compiled pattern went wrong and gave too small a
801 value. Depending on the environment, this could lead to "Failed: internal
802 error: code overflow at offset 49" or "glibc detected double free or
803 corruption" errors.
804
805 27. Applied patches from Google (a) to support the new newline modes and (b) to
806 advance over multibyte UTF-8 characters in GlobalReplace.
807
808 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
809 difference for some implementation of PCRE in some Windows version.
810
811 29. Added some extra testing facilities to pcretest:
812
813 \q<number> in a data line sets the "match limit" value
814 \Q<number> in a data line sets the "match recursion limt" value
815 -S <number> sets the stack size, where <number> is in megabytes
816
817 The -S option isn't available for Windows.
818
819
820 Version 6.6 06-Feb-06
821 ---------------------
822
823 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
824 in pcreposix.h. I have copied the definition from pcre.h.
825
826 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
827 because pcre.h is no longer a built file.
828
829 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
830 not normally included in the compiled code.
831
832
833 Version 6.5 01-Feb-06
834 ---------------------
835
836 1. When using the partial match feature with pcre_dfa_exec(), it was not
837 anchoring the second and subsequent partial matches at the new starting
838 point. This could lead to incorrect results. For example, with the pattern
839 /1234/, partially matching against "123" and then "a4" gave a match.
840
841 2. Changes to pcregrep:
842
843 (a) All non-match returns from pcre_exec() were being treated as failures
844 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
845 error message is output. Some extra information is given for the
846 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
847 probably the only errors that are likely to be caused by users (by
848 specifying a regex that has nested indefinite repeats, for instance).
849 If there are more than 20 of these errors, pcregrep is abandoned.
850
851 (b) A binary zero was treated as data while matching, but terminated the
852 output line if it was written out. This has been fixed: binary zeroes
853 are now no different to any other data bytes.
854
855 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
856 used to set a locale for matching. The --locale=xxxx long option has
857 been added (no short equivalent) to specify a locale explicitly on the
858 pcregrep command, overriding the environment variables.
859
860 (d) When -B was used with -n, some line numbers in the output were one less
861 than they should have been.
862
863 (e) Added the -o (--only-matching) option.
864
865 (f) If -A or -C was used with -c (count only), some lines of context were
866 accidentally printed for the final match.
867
868 (g) Added the -H (--with-filename) option.
869
870 (h) The combination of options -rh failed to suppress file names for files
871 that were found from directory arguments.
872
873 (i) Added the -D (--devices) and -d (--directories) options.
874
875 (j) Added the -F (--fixed-strings) option.
876
877 (k) Allow "-" to be used as a file name for -f as well as for a data file.
878
879 (l) Added the --colo(u)r option.
880
881 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
882 is not present by default.
883
884 3. A nasty bug was discovered in the handling of recursive patterns, that is,
885 items such as (?R) or (?1), when the recursion could match a number of
886 alternatives. If it matched one of the alternatives, but subsequently,
887 outside the recursion, there was a failure, the code tried to back up into
888 the recursion. However, because of the way PCRE is implemented, this is not
889 possible, and the result was an incorrect result from the match.
890
891 In order to prevent this happening, the specification of recursion has
892 been changed so that all such subpatterns are automatically treated as
893 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
894
895 4. I had overlooked the fact that, in some locales, there are characters for
896 which isalpha() is true but neither isupper() nor islower() are true. In
897 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
898 and ordfeminine) are like this. This affected the treatment of \w and \W
899 when they appeared in character classes, but not when they appeared outside
900 a character class. The bit map for "word" characters is now created
901 separately from the results of isalnum() instead of just taking it from the
902 upper, lower, and digit maps. (Plus the underscore character, of course.)
903
904 5. The above bug also affected the handling of POSIX character classes such as
905 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
906 permanent tables. Instead, the bit maps for such a class were previously
907 created as the appropriate unions of the upper, lower, and digit bitmaps.
908 Now they are created by subtraction from the [[:word:]] class, which has
909 its own bitmap.
910
911 6. The [[:blank:]] character class matches horizontal, but not vertical space.
912 It is created by subtracting the vertical space characters (\x09, \x0a,
913 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
914 subtraction was done in the overall bitmap for a character class, meaning
915 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
916 be recognized. This bug has been fixed.
917
918 7. Patches from the folks at Google:
919
920 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
921 real life, but is still worth protecting against".
922
923 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
924 regular expressions".
925
926 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
927 have it.
928
929 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
930 "configure" and the latter not, in order to fix a problem somebody had
931 with compiling the Arg class on HP-UX.
932
933 (e) Improve the error-handling of the C++ wrapper a little bit.
934
935 (f) New tests for checking recursion limiting.
936
937 8. The pcre_memmove() function, which is used only if the environment does not
938 have a standard memmove() function (and is therefore rarely compiled),
939 contained two bugs: (a) use of int instead of size_t, and (b) it was not
940 returning a result (though PCRE never actually uses the result).
941
942 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
943 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
944 returned instead of calling malloc() with an overflowing number that would
945 most likely cause subsequent chaos.
946
947 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
948
949 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
950 with this option is matched, the nmatch and pmatch options of regexec() are
951 ignored.
952
953 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
954 provided in case anyone wants to the the POSIX interface with UTF-8
955 strings.
956
957 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
958 C++ linking (needed for some HP-UX environments).
959
960 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
961 (unused parameter) and in the pcre_printint() function (omitted "default"
962 switch label when the default is to do nothing).
963
964 15. Added some code to make it possible, when PCRE is compiled as a C++
965 library, to replace subject pointers for pcre_exec() with a smart pointer
966 class, thus making it possible to process discontinuous strings.
967
968 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
969 much the same function. They were added by different people who were trying
970 to make PCRE easy to compile on non-Unix systems. It has been suggested
971 that PCRE_EXPORT be abolished now that there is more automatic apparatus
972 for compiling on Windows systems. I have therefore replaced it with
973 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
974 defaults to "extern" for C or "extern C" for C++, which works fine on
975 Unix-like systems. It is now possible to override the value of PCRE_DATA_
976 SCOPE with something explicit in config.h. In addition:
977
978 (a) pcreposix.h still had just "extern" instead of either of these macros;
979 I have replaced it with PCRE_DATA_SCOPE.
980
981 (b) Functions such as _pcre_xclass(), which are internal to the library,
982 but external in the C sense, all had PCRE_EXPORT in their definitions.
983 This is apparently wrong for the Windows case, so I have removed it.
984 (It makes no difference on Unix-like systems.)
985
986 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
987 of recursive calls to match(). This is different to MATCH_LIMIT because
988 that limits the total number of calls to match(), not all of which increase
989 the depth of recursion. Limiting the recursion depth limits the amount of
990 stack (or heap if NO_RECURSE is set) that is used. The default can be set
991 when PCRE is compiled, and changed at run time. A patch from Google adds
992 this functionality to the C++ interface.
993
994 18. Changes to the handling of Unicode character properties:
995
996 (a) Updated the table to Unicode 4.1.0.
997
998 (b) Recognize characters that are not in the table as "Cn" (undefined).
999
1000 (c) I revised the way the table is implemented to a much improved format
1001 which includes recognition of ranges. It now supports the ranges that
1002 are defined in UnicodeData.txt, and it also amalgamates other
1003 characters into ranges. This has reduced the number of entries in the
1004 table from around 16,000 to around 3,000, thus reducing its size
1005 considerably. I realized I did not need to use a tree structure after
1006 all - a binary chop search is just as efficient. Having reduced the
1007 number of entries, I extended their size from 6 bytes to 8 bytes to
1008 allow for more data.
1009
1010 (d) Added support for Unicode script names via properties such as \p{Han}.
1011
1012 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1013 matching that character.
1014
1015 20. When matching a repeated Unicode property with a minimum greater than zero,
1016 (for example \pL{2,}), PCRE could look past the end of the subject if it
1017 reached it while seeking the minimum number of characters. This could
1018 happen only if some of the characters were more than one byte long, because
1019 there is a check for at least the minimum number of bytes.
1020
1021 21. Refactored the implementation of \p and \P so as to be more general, to
1022 allow for more different types of property in future. This has changed the
1023 compiled form incompatibly. Anybody with saved compiled patterns that use
1024 \p or \P will have to recompile them.
1025
1026 22. Added "Any" and "L&" to the supported property types.
1027
1028 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1029 but give a compile time error if the value is greater than 0xff.
1030
1031 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1032 accidentally not being installed or uninstalled.
1033
1034 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1035 made were to insert the current release number. This seemed silly, because
1036 it made things harder for people building PCRE on systems that don't run
1037 "configure". I have turned pcre.h into a distributed file, no longer built
1038 by "configure", with the version identification directly included. There is
1039 no longer a pcre.h.in file.
1040
1041 However, this change necessitated a change to the pcre-config script as
1042 well. It is built from pcre-config.in, and one of the substitutions was the
1043 release number. I have updated configure.ac so that ./configure now finds
1044 the release number by grepping pcre.h.
1045
1046 26. Added the ability to run the tests under valgrind.
1047
1048
1049 Version 6.4 05-Sep-05
1050 ---------------------
1051
1052 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1053 "--" to be printed when multiple files were scanned, even when none of the
1054 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1055 consider it to be a bug, and have restored the previous behaviour.
1056
1057 2. A couple of code tidies to get rid of compiler warnings.
1058
1059 3. The pcretest program used to cheat by referring to symbols in the library
1060 whose names begin with _pcre_. These are internal symbols that are not
1061 really supposed to be visible externally, and in some environments it is
1062 possible to suppress them. The cheating is now confined to including
1063 certain files from the library's source, which is a bit cleaner.
1064
1065 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1066 file's purpose clearer.
1067
1068 5. Reorganized pcre_ucp_findchar().
1069
1070
1071 Version 6.3 15-Aug-05
1072 ---------------------
1073
1074 1. The file libpcre.pc.in did not have general read permission in the tarball.
1075
1076 2. There were some problems when building without C++ support:
1077
1078 (a) If C++ support was not built, "make install" and "make test" still
1079 tried to test it.
1080
1081 (b) There were problems when the value of CXX was explicitly set. Some
1082 changes have been made to try to fix these, and ...
1083
1084 (c) --disable-cpp can now be used to explicitly disable C++ support.
1085
1086 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1087 backslash in a target when C++ was disabled. This confuses some
1088 versions of "make", apparently. Using an intermediate variable solves
1089 this. (Same for CPP_LOBJ.)
1090
1091 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1092 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1093 necessary on certain architectures.
1094
1095 4. Added a setting of -export-symbols-regex to the link command to remove
1096 those symbols that are exported in the C sense, but actually are local
1097 within the library, and not documented. Their names all begin with
1098 "_pcre_". This is not a perfect job, because (a) we have to except some
1099 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1100 available (and never for static libraries). I have made a note to try to
1101 find a way round (a) in the future.
1102
1103
1104 Version 6.2 01-Aug-05
1105 ---------------------
1106
1107 1. There was no test for integer overflow of quantifier values. A construction
1108 such as {1111111111111111} would give undefined results. What is worse, if
1109 a minimum quantifier for a parenthesized subpattern overflowed and became
1110 negative, the calculation of the memory size went wrong. This could have
1111 led to memory overwriting.
1112
1113 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1114
1115 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1116 operating environments where this matters.
1117
1118 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1119 PCRE options from within the C++ wrapper.
1120
1121 5. Named capturing subpatterns were not being correctly counted when a pattern
1122 was compiled. This caused two problems: (a) If there were more than 100
1123 such subpatterns, the calculation of the memory needed for the whole
1124 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1125 back references of the form \12, where the number was greater than 9, were
1126 not recognized as back references, even though there were sufficient
1127 previous subpatterns.
1128
1129 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1130 versions of gcc, e.g. 2.95.4.
1131
1132
1133 Version 6.1 21-Jun-05
1134 ---------------------
1135
1136 1. There was one reference to the variable "posix" in pcretest.c that was not
1137 surrounded by "#if !defined NOPOSIX".
1138
1139 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1140 the cross-check on the old pcre_info() function, for the benefit of the
1141 cut-down version of PCRE that is currently imported into Exim.
1142
1143 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1144 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1145 patterns that start (?i) but I don't think that matters. The use of (?i) is
1146 just an example; this all applies to the other options as well.
1147
1148 4. Since libtool seems to echo the compile commands it is issuing, the output
1149 from "make" can be reduced a bit by putting "@" in front of each libtool
1150 compile command.
1151
1152 5. Patch from the folks at Google for configure.in to be a bit more thorough
1153 in checking for a suitable C++ installation before trying to compile the
1154 C++ stuff. This should fix a reported problem when a compiler was present,
1155 but no suitable headers.
1156
1157 6. The man pages all had just "PCRE" as their title. I have changed them to
1158 be the relevant file name. I have also arranged that these names are
1159 retained in the file doc/pcre.txt, which is a concatenation in text format
1160 of all the man pages except the little individual ones for each function.
1161
1162 7. The NON-UNIX-USE file had not been updated for the different set of source
1163 files that come with release 6. I also added a few comments about the C++
1164 wrapper.
1165
1166
1167 Version 6.0 07-Jun-05
1168 ---------------------
1169
1170 1. Some minor internal re-organization to help with my DFA experiments.
1171
1172 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1173 didn't matter for the library itself when fully configured, but did matter
1174 when compiling without UCP support, or within Exim, where the ucp files are
1175 not imported.
1176
1177 3. Refactoring of the library code to split up the various functions into
1178 different source modules. The addition of the new DFA matching code (see
1179 below) to a single monolithic source would have made it really too
1180 unwieldy, quite apart from causing all the code to be include in a
1181 statically linked application, when only some functions are used. This is
1182 relevant even without the DFA addition now that patterns can be compiled in
1183 one application and matched in another.
1184
1185 The downside of splitting up is that there have to be some external
1186 functions and data tables that are used internally in different modules of
1187 the library but which are not part of the API. These have all had their
1188 names changed to start with "_pcre_" so that they are unlikely to clash
1189 with other external names.
1190
1191 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1192 a different (DFA) algorithm. Although it is slower than the original
1193 function, it does have some advantages for certain types of matching
1194 problem.
1195
1196 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1197 including restarting after a partial match.
1198
1199 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1200 defined when compiling for Windows was sent to me. I have put it into the
1201 code, though I have no means of testing or verifying it.
1202
1203 7. Added the pcre_refcount() auxiliary function.
1204
1205 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1206 match before or at the first newline in the subject string. In pcretest,
1207 the /f option on a pattern can be used to set this.
1208
1209 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1210 would behave wrongly. This has been present in PCRE since release 4.0.
1211
1212 10. A number of changes to the pcregrep command:
1213
1214 (a) Refactored how -x works; insert ^(...)$ instead of setting
1215 PCRE_ANCHORED and checking the length, in preparation for adding
1216 something similar for -w.
1217
1218 (b) Added the -w (match as a word) option.
1219
1220 (c) Refactored the way lines are read and buffered so as to have more
1221 than one at a time available.
1222
1223 (d) Implemented a pcregrep test script.
1224
1225 (e) Added the -M (multiline match) option. This allows patterns to match
1226 over several lines of the subject. The buffering ensures that at least
1227 8K, or the rest of the document (whichever is the shorter) is available
1228 for matching (and similarly the previous 8K for lookbehind assertions).
1229
1230 (f) Changed the --help output so that it now says
1231
1232 -w, --word-regex(p)
1233
1234 instead of two lines, one with "regex" and the other with "regexp"
1235 because that confused at least one person since the short forms are the
1236 same. (This required a bit of code, as the output is generated
1237 automatically from a table. It wasn't just a text change.)
1238
1239 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1240 option but starts with a hyphen. Could be a pattern or a path name
1241 starting with a hyphen, for instance.
1242
1243 (h) "-" can be given as a file name to represent stdin.
1244
1245 (i) When file names are being printed, "(standard input)" is used for
1246 the standard input, for compatibility with GNU grep. Previously
1247 "<stdin>" was used.
1248
1249 (j) The option --label=xxx can be used to supply a name to be used for
1250 stdin when file names are being printed. There is no short form.
1251
1252 (k) Re-factored the options decoding logic because we are going to add
1253 two more options that take data. Such options can now be given in four
1254 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1255
1256 (l) Added the -A, -B, and -C options for requesting that lines of context
1257 around matches be printed.
1258
1259 (m) Added the -L option to print the names of files that do not contain
1260 any matching lines, that is, the complement of -l.
1261
1262 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1263 continue to scan other files.
1264
1265 (o) The -s option was incorrectly implemented. For compatibility with other
1266 greps, it now suppresses the error message for a non-existent or non-
1267 accessible file (but not the return code). There is a new option called
1268 -q that suppresses the output of matching lines, which was what -s was
1269 previously doing.
1270
1271 (p) Added --include and --exclude options to specify files for inclusion
1272 and exclusion when recursing.
1273
1274 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1275 Hopefully, it now does.
1276
1277 12. Missing cast in pcre_study().
1278
1279 13. Added an "uninstall" target to the makefile.
1280
1281 14. Replaced "extern" in the function prototypes in Makefile.in with
1282 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1283 world, but is set differently for Windows.
1284
1285 15. Added a second compiling function called pcre_compile2(). The only
1286 difference is that it has an extra argument, which is a pointer to an
1287 integer error code. When there is a compile-time failure, this is set
1288 non-zero, in addition to the error test pointer being set to point to an
1289 error message. The new argument may be NULL if no error number is required
1290 (but then you may as well call pcre_compile(), which is now just a
1291 wrapper). This facility is provided because some applications need a
1292 numeric error indication, but it has also enabled me to tidy up the way
1293 compile-time errors are handled in the POSIX wrapper.
1294
1295 16. Added VPATH=.libs to the makefile; this should help when building with one
1296 prefix path and installing with another. (Or so I'm told by someone who
1297 knows more about this stuff than I do.)
1298
1299 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1300 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1301 match everything, including newlines. This is not POSIX-compatible, but
1302 somebody wanted the feature. From pcretest it can be activated by using
1303 both the P and the s flags.
1304
1305 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1306
1307 19. libpcre.pc was being incorrectly installed as executable.
1308
1309 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1310 it now also looks for '\r' so that it will work unmodified on Windows.
1311
1312 21. Added Google's contributed C++ wrapper to the distribution.
1313
1314 22. Added some untidy missing memory free() calls in pcretest, to keep
1315 Electric Fence happy when testing.
1316
1317
1318
1319 Version 5.0 13-Sep-04
1320 ---------------------
1321
1322 1. Internal change: literal characters are no longer packed up into items
1323 containing multiple characters in a single byte-string. Each character
1324 is now matched using a separate opcode. However, there may be more than one
1325 byte in the character in UTF-8 mode.
1326
1327 2. The pcre_callout_block structure has two new fields: pattern_position and
1328 next_item_length. These contain the offset in the pattern to the next match
1329 item, and its length, respectively.
1330
1331 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1332 insertion of callouts before each pattern item. Added the /C option to
1333 pcretest to make use of this.
1334
1335 4. On the advice of a Windows user, the lines
1336
1337 #if defined(_WIN32) || defined(WIN32)
1338 _setmode( _fileno( stdout ), 0x8000 );
1339 #endif /* defined(_WIN32) || defined(WIN32) */
1340
1341 have been added to the source of pcretest. This apparently does useful
1342 magic in relation to line terminators.
1343
1344 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1345 for the benefit of those environments where the "b" makes a difference.
1346
1347 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1348 to know about it. I have put a hack into configure.in that adds in code
1349 to set GCC=yes if CC=icc. This seems to end up at a point in the
1350 generated configure script that is early enough to affect the setting of
1351 compiler options, which is what is needed, but I have no means of testing
1352 whether it really works. (The user who reported this had patched the
1353 generated configure script, which of course I cannot do.)
1354
1355 LATER: After change 22 below (new libtool files), the configure script
1356 seems to know about icc (and also ecc). Therefore, I have commented out
1357 this hack in configure.in.
1358
1359 7. Added support for pkg-config (2 patches were sent in).
1360
1361 8. Negated POSIX character classes that used a combination of internal tables
1362 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1363 [[:^ascii]]. Typically, they would match almost any characters. The other
1364 POSIX classes were not broken in this way.
1365
1366 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1367 to find the match, as PCRE was deluded into thinking that the match had to
1368 start at the start point or following a newline. The same bug applied to
1369 patterns with negative forward assertions or any backward assertions
1370 preceding ".*" at the start, unless the pattern required a fixed first
1371 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1372
1373 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1374 starting at the last subject character, bytes beyond the end of the subject
1375 string were read.
1376
1377 11. Renamed the variable "class" as "classbits" to make life easier for C++
1378 users. (Previously there was a macro definition, but it apparently wasn't
1379 enough.)
1380
1381 12. Added the new field "tables" to the extra data so that tables can be passed
1382 in at exec time, or the internal tables can be re-selected. This allows
1383 a compiled regex to be saved and re-used at a later time by a different
1384 program that might have everything at different addresses.
1385
1386 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1387 -R library as well as a -L library.
1388
1389 14. The debugging options of pcretest (-d on the command line or D on a
1390 pattern) showed incorrect output for anything following an extended class
1391 that contained multibyte characters and which was followed by a quantifier.
1392
1393 15. Added optional support for general category Unicode character properties
1394 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1395 support. It adds about 90K to the size of the library. The meanings of the
1396 inbuilt class escapes such as \d and \s have NOT been changed.
1397
1398 16. Updated pcredemo.c to include calls to free() to release the memory for the
1399 compiled pattern.
1400
1401 17. The generated file chartables.c was being created in the source directory
1402 instead of in the building directory. This caused the build to fail if the
1403 source directory was different from the building directory, and was
1404 read-only.
1405
1406 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1407 file. No doubt somebody will tell me if they don't make sense... Also added
1408 Dan Mooney's comments about building on OpenVMS.
1409
1410 19. Added support for partial matching via the PCRE_PARTIAL option for
1411 pcre_exec() and the \P data escape in pcretest.
1412
1413 20. Extended pcretest with 3 new pattern features:
1414
1415 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1416 write the compiled pattern to the file whose name is "rest-of-line".
1417 This is a straight binary dump of the data, with the saved pointer to
1418 the character tables forced to be NULL. The study data, if any, is
1419 written too. After writing, pcretest reads a new pattern.
1420
1421 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1422 compiled pattern from the given file. There must not be any
1423 occurrences of "<" in the file name (pretty unlikely); if there are,
1424 pcretest will instead treat the initial "<" as a pattern delimiter.
1425 After reading in the pattern, pcretest goes on to read data lines as
1426 usual.
1427
1428 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1429 and 16-bit fields in a compiled pattern, to simulate a pattern that
1430 was compiled on a host of opposite endianness.
1431
1432 21. The pcre-exec() function can now cope with patterns that were compiled on
1433 hosts of opposite endianness, with this restriction:
1434
1435 As for any compiled expression that is saved and used later, the tables
1436 pointer field cannot be preserved; the extra_data field in the arguments
1437 to pcre_exec() should be used to pass in a tables address if a value
1438 other than the default internal tables were used at compile time.
1439
1440 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1441 now diagnosed as an error. Previously, most of the time, a negative number
1442 would have been treated as zero, but if in addition "ovector" was passed as
1443 NULL, a crash could occur.
1444
1445 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1446 new versions from the libtool 1.5 distribution (the last one is a copy of
1447 a file called libtool.m4). This seems to have fixed the need to patch
1448 "configure" to support Darwin 1.3 (which I used to do). However, I still
1449 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1450 workstation).
1451
1452 24. Changed the PCRE licence to be the more standard "BSD" licence.
1453
1454
1455 Version 4.5 01-Dec-03
1456 ---------------------
1457
1458 1. There has been some re-arrangement of the code for the match() function so
1459 that it can be compiled in a version that does not call itself recursively.
1460 Instead, it keeps those local variables that need separate instances for
1461 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1462 needs to "recurse". Keeping track of where control must go is done by means
1463 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1464 hide most of the details from the main code, and operates only if
1465 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1466 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1467 operating.
1468
1469 To make it easier for callers to provide specially tailored get/free
1470 functions for this usage, two new functions, pcre_stack_malloc, and
1471 pcre_stack_free, are used. They are always called in strict stacking order,
1472 and the size of block requested is always the same.
1473
1474 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1475 PCRE has been compiled to use the stack or the heap for recursion. The
1476 -C option of pcretest uses this to show which version is compiled.
1477
1478 A new data escape \S, is added to pcretest; it causes the amounts of store
1479 obtained and freed by both kinds of malloc/free at match time to be added
1480 to the output.
1481
1482 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1483 what's available on my current Linux desktop machine.
1484
1485 3. When matching a UTF-8 string, the test for a valid string at the start has
1486 been extended. If start_offset is not zero, PCRE now checks that it points
1487 to a byte that is the start of a UTF-8 character. If not, it returns
1488 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1489 this is necessary because there may be backward assertions in the pattern.
1490 When matching the same subject several times, it may save resources to use
1491 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1492
1493 4. The code for checking the validity of UTF-8 strings has been tightened so
1494 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1495 containing "overlong sequences".
1496
1497 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1498 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1499 should have been "&", but it just so happened that all the cases this let
1500 through by mistake were picked up later in the function.
1501
1502 6. I had used a variable called "isblank" - this is a C99 function, causing
1503 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1504
1505 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1506 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1507 so the version is shown; (c) stop "make test" echoing "./RunTest".
1508
1509 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1510
1511 9. The prototype for memmove() for systems that don't have it was using
1512 size_t, but the inclusion of the header that defines size_t was later. I've
1513 moved the #includes for the C headers earlier to avoid this.
1514
1515 10. Added some adjustments to the code to make it easier to compiler on certain
1516 special systems:
1517
1518 (a) Some "const" qualifiers were missing.
1519 (b) Added the macro EXPORT before all exported functions; by default this
1520 is defined to be empty.
1521 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1522 that it reads its output file name as an argument instead of writing
1523 to the standard output and assuming this can be redirected.
1524
1525 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1526 class containing characters with values greater than 255, PCRE compilation
1527 went into a loop.
1528
1529 12. A recursive reference to a subpattern that was within another subpattern
1530 that had a minimum quantifier of zero caused PCRE to crash. For example,
1531 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1532 recursion. If the recursively-called subpattern itself had a zero repeat,
1533 that was OK.
1534
1535 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1536 buffer into which it was copied (for escape processing) was still set at
1537 1024, so long lines caused crashes.
1538
1539 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1540 "internal error: code overflow...". This applied to any character class
1541 that was followed by a possessive quantifier.
1542
1543 15. Modified the Makefile to add libpcre.la as a prerequisite for
1544 libpcreposix.la because I was told this is needed for a parallel build to
1545 work.
1546
1547 16. If a pattern that contained .* following optional items at the start was
1548 studied, the wrong optimizing data was generated, leading to matching
1549 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1550 matching string must start with a or b or c. The correct conclusion for
1551 this pattern is that a match can start with any character.
1552
1553
1554 Version 4.4 13-Aug-03
1555 ---------------------
1556
1557 1. In UTF-8 mode, a character class containing characters with values between
1558 127 and 255 was not handled correctly if the compiled pattern was studied.
1559 In fixing this, I have also improved the studying algorithm for such
1560 classes (slightly).
1561
1562 2. Three internal functions had redundant arguments passed to them. Removal
1563 might give a very teeny performance improvement.
1564
1565 3. Documentation bug: the value of the capture_top field in a callout is *one
1566 more than* the number of the hightest numbered captured substring.
1567
1568 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1569 in incorrectly linking with a previously installed version. They now link
1570 explicitly with libpcre.la.
1571
1572 5. configure.in no longer needs to recognize Cygwin specially.
1573
1574 6. A problem in pcre.in for Windows platforms is fixed.
1575
1576 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1577 pcretest, it used to include the size of the study block as part of its
1578 output. Unfortunately, the structure contains a field that has a different
1579 size on different hardware architectures. This meant that the tests that
1580 showed this size failed. As the block is currently always of a fixed size,
1581 this information isn't actually particularly useful in pcretest output, so
1582 I have just removed it.
1583
1584 8. Three pre-processor statements accidentally did not start in column 1.
1585 Sadly, there are *still* compilers around that complain, even though
1586 standard C has not required this for well over a decade. Sigh.
1587
1588 9. In pcretest, the code for checking callouts passed small integers in the
1589 callout_data field, which is a void * field. However, some picky compilers
1590 complained about the casts involved for this on 64-bit systems. Now
1591 pcretest passes the address of the small integer instead, which should get
1592 rid of the warnings.
1593
1594 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1595 both compile and run time, and gives an error if an invalid UTF-8 sequence
1596 is found. There is a option for disabling this check in cases where the
1597 string is known to be correct and/or the maximum performance is wanted.
1598
1599 11. In response to a bug report, I changed one line in Makefile.in from
1600
1601 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1602 to
1603 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1604
1605 to look similar to other lines, but I have no way of telling whether this
1606 is the right thing to do, as I do not use Windows. No doubt I'll get told
1607 if it's wrong...
1608
1609
1610 Version 4.3 21-May-03
1611 ---------------------
1612
1613 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1614 Makefile.
1615
1616 2. Some refactoring to improve the quality of the code:
1617
1618 (i) The utf8_table... variables are now declared "const".
1619
1620 (ii) The code for \cx, which used the "case flipping" table to upper case
1621 lower case letters, now just substracts 32. This is ASCII-specific,
1622 but the whole concept of \cx is ASCII-specific, so it seems
1623 reasonable.
1624
1625 (iii) PCRE was using its character types table to recognize decimal and
1626 hexadecimal digits in the pattern. This is silly, because it handles
1627 only 0-9, a-f, and A-F, but the character types table is locale-
1628 specific, which means strange things might happen. A private
1629 table is now used for this - though it costs 256 bytes, a table is
1630 much faster than multiple explicit tests. Of course, the standard
1631 character types table is still used for matching digits in subject
1632 strings against \d.
1633
1634 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1635 ending in _t are). So I've renamed it as ESC_tee.
1636
1637 3. The first argument for regexec() in the POSIX wrapper should have been
1638 defined as "const".
1639
1640 4. Changed pcretest to use malloc() for its buffers so that they can be
1641 Electric Fenced for debugging.
1642
1643 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1644 to read one or more bytes before the start of the subject string. Often this
1645 had no effect on PCRE's behaviour, but in some circumstances it could
1646 provoke a segmentation fault.
1647
1648 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1649 to try to read one or more bytes before the start of the subject string.
1650
1651 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1652 UTF-8 support could misbehave in various ways if the subject string
1653 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1654 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1655 back over UTF-8 characters.)
1656
1657
1658 Version 4.2 14-Apr-03
1659 ---------------------
1660
1661 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1662
1663 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1664 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1665 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1666 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1667 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1668 and BUILD_EXEEXT
1669 Note: automatic setting of the BUILD variables is not yet working
1670 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1671 compile-time but not at link-time
1672 [LINK]: use for linking executables only
1673 make different versions for Windows and non-Windows
1674 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1675 libraries
1676 [LINK_FOR_BUILD]: new variable
1677 [OBJEXT]: use throughout
1678 [EXEEXT]: use throughout
1679 <winshared>: new target
1680 <wininstall>: new target
1681 <dftables.o>: use native compiler
1682 <dftables>: use native linker
1683 <install>: handle Windows platform correctly
1684 <clean>: ditto
1685 <check>: ditto
1686 copy DLL to top builddir before testing
1687
1688 As part of these changes, -no-undefined was removed again. This was reported
1689 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1690 in any case.
1691
1692 3. Some tidies to get rid of compiler warnings:
1693
1694 . In the match_data structure, match_limit was an unsigned long int, whereas
1695 match_call_count was an int. I've made them both unsigned long ints.
1696
1697 . In pcretest the fact that a const uschar * doesn't automatically cast to
1698 a void * provoked a warning.
1699
1700 . Turning on some more compiler warnings threw up some "shadow" variables
1701 and a few more missing casts.
1702
1703 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1704 option, a class that contained a single character with a value between 128
1705 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1706
1707 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1708 option, a class that contained several characters, but with at least one
1709 whose value was between 128 and 255 caused PCRE to crash.
1710
1711
1712 Version 4.1 12-Mar-03
1713 ---------------------
1714
1715 1. Compiling with gcc -pedantic found a couple of places where casts were
1716 needed, and a string in dftables.c that was longer than standard compilers are
1717 required to support.
1718
1719 2. Compiling with Sun's compiler found a few more places where the code could
1720 be tidied up in order to avoid warnings.
1721
1722 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1723 first of these names is deprecated in the latest Autoconf in favour of the name
1724 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1725 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1726 analogy I have changed it to CFLAGS_FOR_BUILD.
1727
1728 4. Added -no-undefined to the linking command in the Makefile, because this is
1729 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1730 linking step for the pcreposix library.
1731
1732 5. PCRE was failing to diagnose the case of two named groups with the same
1733 name.
1734
1735 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1736 literal character that is needed in the subject for a match, and scans along to
1737 ensure that it is present before embarking on the full matching process. This
1738 saves time in cases of nested unlimited repeats that are never going to match.
1739 Problem: the scan can take a lot of time if the subject is very long (e.g.
1740 megabytes), thus penalizing straightforward matches. It is now done only if the
1741 amount of subject to be scanned is less than 1000 bytes.
1742
1743 7. A lesser problem with the same optimization is that it was recording the
1744 first character of an anchored pattern as "needed", thus provoking a search
1745 right along the subject, even when the first match of the pattern was going to
1746 fail. The "needed" character is now not set for anchored patterns, unless it
1747 follows something in the pattern that is of non-fixed length. Thus, it still
1748 fulfils its original purpose of finding quick non-matches in cases of nested
1749 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1750
1751
1752 Version 4.0 17-Feb-03
1753 ---------------------
1754
1755 1. If a comment in an extended regex that started immediately after a meta-item
1756 extended to the end of string, PCRE compiled incorrect data. This could lead to
1757 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1758
1759 2. Moved to autoconf 2.53 and libtool 1.4.2.
1760
1761 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1762 the special perltest8 script is no longer needed - all the tests can be run
1763 from a single perltest script.
1764
1765 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1766 by \s. It has now been removed in PCRE. This means it isn't recognized as
1767 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1768 class [:space:] *does* include VT, thereby creating a mess.
1769
1770 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1771 space and tab.
1772
1773 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1774 its new features into the main test script, reducing the number of scripts.
1775
1776 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1777 were backward compatible, and made the (?i) apply to the whole pattern, as if
1778 /i were given. Now it behaves more logically, and applies the option setting
1779 only to what follows. PCRE has been changed to follow suit. However, if it
1780 finds options settings right at the start of the pattern, it extracts them into
1781 the global options, as before. Thus, they show up in the info data.
1782
1783 8. Added support for the \Q...\E escape sequence. Characters in between are
1784 treated as literals. This is slightly different from Perl in that $ and @ are
1785 also handled as literals inside the quotes. In Perl, they will cause variable
1786 interpolation. Note the following examples:
1787
1788 Pattern PCRE matches Perl matches
1789
1790 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1791 \Qabc\$xyz\E abc\$xyz abc\$xyz
1792 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1793
1794 For compatibility with Perl, \Q...\E sequences are recognized inside character
1795 classes as well as outside them.
1796
1797 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1798 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1799 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1800 signed/unsigned warnings.
1801
1802 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1803 option for pcretest, so I've replaced it by a simple function that does just
1804 that job.
1805
1806 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1807 "pcregrep -".
1808
1809 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1810 Java package. This provides some syntactic sugar for simple cases of what my
1811 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1812 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1813 item, you can use this simplified notation. Note that only makes sense with
1814 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1815 greediness, whatever the setting of the PCRE_UNGREEDY option.
1816
1817 13. A change of greediness default within a pattern was not taking effect at
1818 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1819 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1820 was abstracted outside.
1821
1822 14. PCRE now supports the \G assertion. It is true when the current matching
1823 position is at the start point of the match. This differs from \A when the
1824 starting offset is non-zero. Used with the /g option of pcretest (or similar
1825 code), it works in the same way as it does for Perl's /g option. If all
1826 alternatives of a regex begin with \G, the expression is anchored to the start
1827 match position, and the "anchored" flag is set in the compiled expression.
1828
1829 15. Some bugs concerning the handling of certain option changes within patterns
1830 have been fixed. These applied to options other than (?ims). For example,
1831 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1832 been the other way round. Some of this was related to change 7 above.
1833
1834 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1835 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1836 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1837 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1838
1839 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1840 mode. Unlike ".", it always matches newline, whatever the setting of
1841 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1842 assertions. Perl allows it, but it doesn't (in general) work because it can't
1843 calculate the length of the lookbehind. At least, that's the case for Perl
1844 5.8.0 - I've been told they are going to document that it doesn't work in
1845 future.
1846
1847 18. Added an error diagnosis for escapes that PCRE does not support: these are
1848 \L, \l, \N, \P, \p, \U, \u, and \X.
1849
1850 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1851 reading past the end of the pattern in cases such as /[abcd/.
1852
1853 20. PCRE was getting more memory than necessary for patterns with classes that
1854 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1855
1856 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1857 compiling PCRE for use with Virtual Pascal.
1858
1859 22. Small fix to the Makefile to make it work properly if the build is done
1860 outside the source tree.
1861
1862 23. Added a new extension: a condition to go with recursion. If a conditional
1863 subpattern starts with (?(R) the "true" branch is used if recursion has
1864 happened, whereas the "false" branch is used only at the top level.
1865
1866 24. When there was a very long string of literal characters (over 255 bytes
1867 without UTF support, over 250 bytes with UTF support), the computation of how
1868 much memory was required could be incorrect, leading to segfaults or other
1869 strange effects.
1870
1871 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1872 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1873 there was a subsequent back reference to those brackets. This meant that, for
1874 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1875 possible to check for precisely this case. All we can do is abandon the
1876 optimization if .* occurs inside capturing brackets when there are any back
1877 references whatsoever. (See below for a better fix that came later.)
1878
1879 26. The handling of the optimization for finding the first character of a
1880 non-anchored pattern, and for finding a character that is required later in the
1881 match were failing in some cases. This didn't break the matching; it just
1882 failed to optimize when it could. The way this is done has been re-implemented.
1883
1884 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1885
1886 28. Added a new feature that provides some of the functionality that Perl
1887 provides with (?{...}). The facility is termed a "callout". The way it is done
1888 in PCRE is for the caller to provide an optional function, by setting
1889 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1890 global variable. By default it is unset, which disables all calling out. To get
1891 the function called, the regex must include (?C) at appropriate points. This
1892 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1893 This provides a means of identifying different callout points. When PCRE
1894 reaches such a point in the regex, if pcre_callout has been set, the external
1895 function is called. It is provided with data in a structure called
1896 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1897 matching continues; if it returns a non-zero value, the match at the current
1898 point fails. However, backtracking will occur if possible. [This was changed
1899 later and other features added - see item 49 below.]
1900
1901 29. pcretest is upgraded to test the callout functionality. It provides a
1902 callout function that displays information. By default, it shows the start of
1903 the match and the current position in the text. There are some new data escapes
1904 to vary what happens:
1905
1906 \C+ in addition, show current contents of captured substrings
1907 \C- do not supply a callout function
1908 \C!n return 1 when callout number n is reached
1909 \C!n!m return 1 when callout number n is reached for the mth time
1910
1911 30. If pcregrep was called with the -l option and just a single file name, it
1912 output "<stdin>" if a match was found, instead of the file name.
1913
1914 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1915 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1916 pcre_exec(). This saves a malloc/free per call. The default value of
1917 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1918 when configuring.
1919
1920 32. The default maximum size of a compiled pattern is 64K. There have been a
1921 few cases of people hitting this limit. The code now uses macros to handle the
1922 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1923 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1924 configuring. Tests 2 and 5 work only with 2-byte links because they output
1925 debugging information about compiled patterns.
1926
1927 33. Internal code re-arrangements:
1928
1929 (a) Moved the debugging function for printing out a compiled regex into
1930 its own source file (printint.c) and used #include to pull it into
1931 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1932 separate copies.
1933
1934 (b) Defined the list of op-code names for debugging as a macro in
1935 internal.h so that it is next to the definition of the opcodes.
1936
1937 (c) Defined a table of op-code lengths for simpler skipping along compiled
1938 code. This is again a macro in internal.h so that it is next to the
1939 definition of the opcodes.
1940
1941 34. Added support for recursive calls to individual subpatterns, along the
1942 lines of Robin Houston's patch (but implemented somewhat differently).
1943
1944 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1945 allow it to read and process whole directories in Win32. This code was
1946 contributed by Lionel Fourquaux; it has not been tested by me.
1947
1948 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1949 used to name a group. Names consist of alphanumerics and underscores, and must
1950 be unique. Back references use the syntax (?P=name) and recursive calls use
1951 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1952 numbers. The function pcre_fullinfo() can be used after compilation to extract
1953 a name/number map. There are three relevant calls:
1954
1955 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1956 PCRE_INFO_NAMECOUNT yields the number of entries
1957 PCRE_INFO_NAMETABLE yields a pointer to the map.
1958
1959 The map is a vector of fixed-size entries. The size of each entry depends on
1960 the length of the longest name used. The first two bytes of each entry are the
1961 group number, most significant byte first. There follows the corresponding
1962 name, zero terminated. The names are in alphabetical order.
1963
1964 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1965 case instead of 255. Making it the same both with and without UTF-8 support
1966 means that the same test output works with both.
1967
1968 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1969 calling malloc() with a zero argument.
1970
1971 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1972 optimization. I've improved things by keeping a bitmap of backreferences with
1973 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1974 fact referenced, the optimization can be applied. It is unlikely that a
1975 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1976 the match to follow \n) will appear inside brackets with a number greater than
1977 31, but if it does, any back reference > 31 suppresses the optimization.
1978
1979 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1980 of disabling numbered capturing parentheses. Any opening parenthesis that is
1981 not followed by ? behaves as if it were followed by ?: but named parentheses
1982 can still be used for capturing (and they will acquire numbers in the usual
1983 way).
1984
1985 41. Redesigned the return codes from the match() function into yes/no/error so
1986 that errors can be passed back from deep inside the nested calls. A malloc
1987 failure while inside a recursive subpattern call now causes the
1988 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1989
1990 42. It is now possible to set a limit on the number of times the match()
1991 function is called in a call to pcre_exec(). This facility makes it possible to
1992 limit the amount of recursion and backtracking, though not in a directly
1993 obvious way, because the match() function is used in a number of different
1994 circumstances. The count starts from zero for each position in the subject
1995 string (for non-anchored patterns). The default limit is, for compatibility, a
1996 large number, namely 10 000 000. You can change this in two ways:
1997
1998 (a) When configuring PCRE before making, you can use --with-match-limit=n
1999 to set a default value for the compiled library.
2000
2001 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2002 a different value is set. See 45 below.
2003
2004 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2005
2006 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2007 of things that can be changed at compile time. The first argument specifies
2008 what is wanted and the second points to where the information is to be placed.
2009 The current list of available information is:
2010
2011 PCRE_CONFIG_UTF8
2012
2013 The output is an integer that is set to one if UTF-8 support is available;
2014 otherwise it is set to zero.
2015
2016 PCRE_CONFIG_NEWLINE
2017
2018 The output is an integer that it set to the value of the code that is used for
2019 newline. It is either LF (10) or CR (13).
2020
2021 PCRE_CONFIG_LINK_SIZE
2022
2023 The output is an integer that contains the number of bytes used for internal
2024 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2025
2026 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2027
2028 The output is an integer that contains the threshold above which the POSIX
2029 interface uses malloc() for output vectors. See item 31 above.
2030
2031 PCRE_CONFIG_MATCH_LIMIT
2032
2033 The output is an unsigned integer that contains the default limit of the number
2034 of match() calls in a pcre_exec() execution. See 42 above.
2035
2036 44. pcretest has been upgraded by the addition of the -C option. This causes it
2037 to extract all the available output from the new pcre_config() function, and to
2038 output it. The program then exits immediately.
2039
2040 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2041 order to support additional features. One way would have been to define
2042 pcre_exec2() (for example) with extra arguments, but this would not have been
2043 extensible, and would also have required all calls to the original function to
2044 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2045 is used for passing in "extra" data from pcre_study().
2046
2047 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2048 contains the following fields:
2049
2050 flags a bitmap indicating which of the following fields are set
2051 study_data opaque data from pcre_study()
2052 match_limit a way of specifying a limit on match() calls for a specific
2053 call to pcre_exec()
2054 callout_data data for callouts (see 49 below)
2055
2056 The flag bits are also defined in pcre.h, and are
2057
2058 PCRE_EXTRA_STUDY_DATA
2059 PCRE_EXTRA_MATCH_LIMIT
2060 PCRE_EXTRA_CALLOUT_DATA
2061
2062 The pcre_study() function now returns one of these new pcre_extra blocks, with
2063 the actual study data pointed to by the study_data field, and the
2064 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2065 before. That is, this change is entirely upwards-compatible and requires no
2066 change to existing code.
2067
2068 If you want to pass in additional data to pcre_exec(), you can either place it
2069 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2070 block.
2071
2072 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2073 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2074 times with different match limits, until it finds the minimum value needed for
2075 pcre_exec() to complete. The value is then output. This can be instructive; for
2076 most simple matches the number is quite small, but for pathological cases it
2077 gets very large very quickly.
2078
2079 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2080 returns the size of the data block pointed to by the study_data field in a
2081 pcre_extra block, that is, the value that was passed as the argument to
2082 pcre_malloc() when PCRE was getting memory in which to place the information
2083 created by pcre_study(). The fourth argument should point to a size_t variable.
2084 pcretest has been extended so that this information is shown after a successful
2085 pcre_study() call when information about the compiled regex is being displayed.
2086
2087 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2088 because what follows is always an absolute path. (Later: it turns out that this
2089 is more than cosmetic for MinGW, because it doesn't like empty path
2090 components.)
2091
2092 49. Some changes have been made to the callout feature (see 28 above):
2093
2094 (i) A callout function now has three choices for what it returns:
2095
2096 0 => success, carry on matching
2097 > 0 => failure at this point, but backtrack if possible
2098 < 0 => serious error, return this value from pcre_exec()
2099
2100 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2101 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2102 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2103 use by callout functions. It will never be used by PCRE itself.
2104
2105 (ii) The pcre_extra structure (see 45 above) has a void * field called
2106 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2107 pcre_callout_block structure has a field of the same name. The contents of
2108 the field passed in the pcre_extra structure are passed to the callout
2109 function in the corresponding field in the callout block. This makes it
2110 easier to use the same callout-containing regex from multiple threads. For
2111 testing, the pcretest program has a new data escape
2112
2113 \C*n pass the number n (may be negative) as callout_data
2114
2115 If the callout function in pcretest receives a non-zero value as
2116 callout_data, it returns that value.
2117
2118 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2119 there were some redundant $(CFLAGS) in commands that are now specified as
2120 $(LINK), which already includes $(CFLAGS).
2121
2122 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2123 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2124 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2125 one-byte characters throughout. Note that case-insensitive matching applies
2126 only to characters whose values are less than 256. PCRE doesn't support the
2127 notion of cases for higher-valued characters.
2128
2129 (i) A character class whose characters are all within 0-255 is handled as
2130 a bit map, and the map is inverted for negative classes. Previously, a
2131 character > 255 always failed to match such a class; however it should
2132 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2133
2134 (ii) A negated character class with a single character < 255 is coded as
2135 "not this character" (OP_NOT). This wasn't working properly when the test
2136 character was multibyte, either singly or repeated.
2137
2138 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2139 mode, for example: \x{100}{2,3}.
2140
2141 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2142 singly or repeated) now correctly test multibyte characters. However,
2143 PCRE doesn't recognize any characters with values greater than 255 as
2144 digits, spaces, or word characters. Such characters always match \D, \S,
2145 and \W, and never match \d, \s, or \w.
2146
2147 (v) Classes may now contain characters and character ranges with values
2148 greater than 255. For example: [ab\x{100}-\x{400}].
2149
2150 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2151 PCRE in UTF-8 mode.
2152
2153 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2154 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2155 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2156 value.)
2157
2158 53. The single man page has become too large. I have therefore split it up into
2159 a number of separate man pages. These also give rise to individual HTML pages;
2160 these are now put in a separate directory, and there is an index.html page that
2161 lists them all. Some hyperlinking between the pages has been installed.
2162
2163 54. Added convenience functions for handling named capturing parentheses.
2164
2165 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2166 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2167 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2168 are faulted.
2169
2170 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2171 calling configure. These values are used when compiling the dftables.c program
2172 which is run to generate the source of the default character tables. They
2173 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2174 you will need to set these values.
2175
2176 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2177
2178
2179 Version 3.9 02-Jan-02
2180 ---------------------
2181
2182 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2183
2184 2. If --disable-static was given, the building process failed when trying to
2185 build pcretest and pcregrep. (For some reason it was using libtool to compile
2186 them, which is not right, as they aren't part of the library.)
2187
2188
2189 Version 3.8 18-Dec-01
2190 ---------------------
2191
2192 1. The experimental UTF-8 code was completely screwed up. It was packing the
2193 bytes in the wrong order. How dumb can you get?
2194
2195
2196 Version 3.7 29-Oct-01
2197 ---------------------
2198
2199 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2200 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2201 this didn't happen under Solaris 8, where I normally test things.
2202
2203 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2204 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2205 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2206 this fix an hour or so after the initial 3.7 release.)
2207
2208
2209 Version 3.6 23-Oct-01
2210 ---------------------
2211
2212 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2213 offsets passed as NULL with zero offset count.
2214
2215 2. The config.guess and config.sub files had not been updated when I moved to
2216 the latest autoconf.
2217
2218
2219 Version 3.5 15-Aug-01
2220 ---------------------
2221
2222 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2223 had been forgotten.
2224
2225 2. By using declared but undefined structures, we can avoid using "void"
2226 definitions in pcre.h while keeping the internal definitions of the structures
2227 private.
2228
2229 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2230 user point of view, this means that both static and shared libraries are built
2231 by default, but this can be individually controlled. More of the work of
2232 handling this static/shared cases is now inside libtool instead of PCRE's make
2233 file.
2234
2235 4. The pcretest utility is now installed along with pcregrep because it is
2236 useful for users (to test regexs) and by doing this, it automatically gets
2237 relinked by libtool. The documentation has been turned into a man page, so
2238 there are now .1, .txt, and .html versions in /doc.
2239
2240 5. Upgrades to pcregrep:
2241 (i) Added long-form option names like gnu grep.
2242 (ii) Added --help to list all options with an explanatory phrase.
2243 (iii) Added -r, --recursive to recurse into sub-directories.
2244 (iv) Added -f, --file to read patterns from a file.
2245
2246 6. pcre_exec() was referring to its "code" argument before testing that
2247 argument for NULL (and giving an error if it was NULL).
2248
2249 7. Upgraded Makefile.in to allow for compiling in a different directory from
2250 the source directory.
2251
2252 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2253 options bits, the pointer it was passed was to an int instead of to an unsigned
2254 long int. This mattered only on 64-bit systems.
2255
2256 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2257 generated) instead of pcre.in, which it its source. Also made the same change
2258 in several of the .c files.
2259
2260 10. A new release of gcc defines printf() as a macro, which broke pcretest
2261 because it had an ifdef in the middle of a string argument for printf(). Fixed
2262 by using separate calls to printf().
2263
2264 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2265 script, to force use of CR or LF instead of \n in the source. On non-Unix
2266 systems, the value can be set in config.h.
2267
2268 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2269 absolute limit. Changed the text of the error message to make this clear, and
2270 likewise updated the man page.
2271
2272 13. The limit of 99 on the number of capturing subpatterns has been removed.
2273 The new limit is 65535, which I hope will not be a "real" limit.
2274
2275
2276 Version 3.4 22-Aug-00
2277 ---------------------
2278
2279 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2280
2281 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2282
2283
2284 Version 3.3 01-Aug-00
2285 ---------------------
2286
2287 1. If an octal character was given, but the value was greater than \377, it
2288 was not getting masked to the least significant bits, as documented. This could
2289 lead to crashes in some systems.
2290
2291 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2292 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2293
2294 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2295 These just pass their arguments on to (pcre_free)(), but they are provided
2296 because some uses of PCRE bind it to non-C systems that can call its functions,
2297 but cannot call free() or pcre_free() directly.
2298
2299 4. Add "make test" as a synonym for "make check". Corrected some comments in
2300 the Makefile.
2301
2302 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2303 Makefile.
2304
2305 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2306 command called pgrep for grepping around the active processes.
2307
2308 7. Added the beginnings of support for UTF-8 character strings.
2309
2310 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2311 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2312 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2313 out for the ar command.)
2314
2315
2316 Version 3.2 12-May-00
2317 ---------------------
2318
2319 This is purely a bug fixing release.
2320
2321 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2322 of ZA. This was just one example of several cases that could provoke this bug,
2323 which was introduced by change 9 of version 2.00. The code for breaking
2324 infinite loops after an iteration that matches an empty string was't working
2325 correctly.
2326
2327 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2328 when matched against abbab (for example). After matching an empty string, it
2329 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2330 caused it to match further down the string than it should.
2331
2332 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2333 was there because it doesn't seem to be needed, and it causes trouble on some
2334 systems, as it is not a Standard C header. It has been removed.
2335
2336 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2337 were reported on the Macintosh. The changes were from
2338
2339 while ((c = *(++ptr)) != 0 && c != '\n');
2340 to
2341 while ((c = *(++ptr)) != 0 && c != '\n') ;
2342
2343 Totally extraordinary, but if that's what it takes...
2344
2345 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2346 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2347 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2348 assumes the way PCRE uses memmove() (always moving upwards).
2349
2350 6. PCRE is being used in one environment where strchr() is not available. There
2351 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2352 faster code anyway.
2353
2354
2355 Version 3.1 09-Feb-00
2356 ---------------------
2357
2358 The only change in this release is the fixing of some bugs in Makefile.in for
2359 the "install" target:
2360
2361 (1) It was failing to install pcreposix.h.
2362
2363 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2364
2365
2366 Version 3.0 01-Feb-00
2367 ---------------------
2368
2369 1. Add support for the /+ modifier to perltest (to output $` like it does in
2370 pcretest).
2371
2372 2. Add support for the /g modifier to perltest.
2373
2374 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2375 matches null strings.
2376
2377 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2378 pattern. Perl treats empty patterns specially - it reuses the most recent
2379 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2380 effect.
2381
2382 5. The POSIX interface was broken in that it was just handing over the POSIX
2383 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2384 required a bigger vector, with some working space on the end. This means that
2385 the POSIX wrapper now has to get and free some memory, and copy the results.
2386
2387 6. Added some simple autoconf support, placing the test data and the
2388 documentation in separate directories, re-organizing some of the
2389 information files, and making it build pcre-config (a GNU standard). Also added
2390 libtool support for building PCRE as a shared library, which is now the
2391 default.
2392
2393 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2394 09 are not valid octal constants. Single digits will be used for minor values
2395 less than 10.
2396
2397 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2398 existing programs that set these in the POSIX interface can use PCRE without
2399 modification.
2400
2401 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2402 return all that pcre_info() returns, plus additional data. The pcre_info()
2403 function is retained for compatibility, but is considered to be obsolete.
2404
2405 10. Added experimental recursion feature (?R) to handle one common case that
2406 Perl 5.6 will be able to do with (?p{...}).
2407
2408 11. Added support for POSIX character classes like [:alpha:], which Perl is
2409 adopting.
2410
2411
2412 Version 2.08 31-Aug-99
2413 ----------------------
2414
2415 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2416 trying to match at the startoffset position, but instead was moving forward to
2417 the next newline as if a previous match had failed.
2418
2419 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2420 and could get into a loop if a null string was matched other than at the start
2421 of the subject.
2422
2423 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2424 be distinguished at compile time, and for completeness also added PCRE_DATE.
2425
2426 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2427 in GnuWin32 environments.
2428
2429
2430 Version 2.07 29-Jul-99
2431 ----------------------
2432
2433 1. The documentation is now supplied in plain text form and HTML as well as in
2434 the form of man page sources.
2435
2436 2. C++ compilers don't like assigning (void *) values to other pointer types.
2437 In particular this affects malloc(). Although there is no problem in Standard
2438 C, I've put in casts to keep C++ compilers happy.
2439
2440 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2441 should be (const char *).
2442
2443 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2444 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2445 However, I haven't made this a standard facility. The documentation doesn't
2446 mention it, and the Makefile doesn't support it.
2447
2448 5. The Makefile now contains an "install" target, with editable destinations at
2449 the top of the file. The pcretest program is not installed.
2450
2451 6. pgrep -V now gives the PCRE version number and date.
2452
2453 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2454 causing the entire string to be ignored, instead of just the last character.
2455
2456 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2457 non-matching string, it can take a very, very long time, even for strings of
2458 quite modest length, because of the nested recursion. PCRE now does better in
2459 some of these cases. It does this by remembering the last required literal
2460 character in the pattern, and pre-searching the subject to ensure it is present
2461 before running the real match. In other words, it applies a heuristic to detect
2462 some types of certain failure quickly, and in the above example, if presented
2463 with a string that has no trailing " it gives "no match" very quickly.
2464
2465 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2466 other alternatives are tried instead.
2467
2468
2469 Version 2.06 09-Jun-99
2470 ----------------------
2471
2472 1. Change pcretest's output for amount of store used to show just the code
2473 space, because the remainder (the data block) varies in size between 32-bit and
2474 64-bit systems.
2475
2476 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2477 start matching at. This allows lookbehinds to work when searching for multiple
2478 occurrences in a string.
2479
2480 3. Added additional options to pcretest for testing multiple occurrences:
2481
2482 /+ outputs the rest of the string that follows a match
2483 /g loops for multiple occurrences, using the new startoffset argument
2484 /G loops for multiple occurrences by passing an incremented pointer
2485
2486 4. PCRE wasn't doing the "first character" optimization for patterns starting
2487 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2488 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2489 the letter 'x'. On long subject strings, this gives a significant speed-up.
2490
2491
2492 Version 2.05 21-Apr-99
2493 ----------------------
2494
2495 1. Changed the type of magic_number from int to long int so that it works
2496 properly on 16-bit systems.
2497
2498 2. Fixed a bug which caused patterns starting with .* not to work correctly
2499 when the subject string contained newline characters. PCRE was assuming
2500 anchoring for such patterns in all cases, which is not correct because .* will
2501 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2502 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2503 must be retried after every newline in the subject.
2504
2505
2506 Version 2.04 18-Feb-99
2507 ----------------------
2508
2509 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2510 computation of the store needed to hold the pattern was incorrect (too large).
2511 If such patterns were nested a few deep, this could multiply and become a real
2512 problem.
2513
2514 2. Added /M option to pcretest to show the memory requirement of a specific
2515 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2516
2517 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2518 compiled in such a way that the backtracking after subsequent failure was
2519 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2520 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2521
2522
2523 Version 2.03 02-Feb-99
2524 ----------------------
2525
2526 1. Fixed typo and small mistake in man page.
2527
2528 2. Added 4th condition (GPL supersedes if conflict) and created separate
2529 LICENCE file containing the conditions.
2530
2531 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2532 Perl, that is the internal \ allows the delimiter to be included in the
2533 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2534 the final delimiter, add \ to the end of the pattern (to test the error).
2535
2536 4. Added the convenience functions for extracting substrings after a successful
2537 match. Updated pcretest to make it able to test these functions.
2538
2539
2540 Version 2.02 14-Jan-99
2541 ----------------------
2542
2543 1. Initialized the working variables associated with each extraction so that
2544 their saving and restoring doesn't refer to uninitialized store.
2545
2546 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2547 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2548 fix the problem.
2549
2550 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2551 calls, and wasn't printing the correct value for compiling calls. Increased the
2552 default value of LOOPREPEAT, and the number of significant figures in the
2553 times.
2554
2555 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2556
2557 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2558 a building problem on Windows NT with a FAT file system.
2559
2560
2561 Version 2.01 21-Oct-98
2562 ----------------------
2563
2564 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2565 to character tables built by pcre_maketables() in the current locale. If NULL
2566 is passed, the default tables are used.
2567
2568
2569 Version 2.00 24-Sep-98
2570 ----------------------
2571
2572 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2573 it any more.
2574
2575 2. Allow quantification of (?>) groups, and make it work correctly.
2576
2577 3. The first character computation wasn't working for (?>) groups.
2578
2579 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2580 end of the subject) and add 5.005's \z, which really does match only at the
2581 very end of the subject.
2582
2583 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2584
2585 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2586 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2587 localized options. All options to pcre_study() were also removed.
2588
2589 7. Add other new features from 5.005:
2590
2591 $(?<= positive lookbehind
2592 $(?<! negative lookbehind
2593 (?imsx-imsx) added the unsetting capability
2594 such a setting is global if at outer level; local otherwise
2595 (?imsx-imsx:) non-capturing groups with option setting
2596 (?(cond)re|re) conditional pattern matching
2597
2598 A backreference to itself in a repeated group matches the previous
2599 captured string.
2600
2601 8. General tidying up of studying (both automatic and via "study")
2602 consequential on the addition of new assertions.
2603
2604 9. As in 5.005, unlimited repeated groups that could match an empty substring
2605 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2606 runtime if any iteration does actually match an empty substring.
2607
2608 10. Include the RunTest script in the distribution.
2609
2610 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2611 discrepancies, some of which were old and were also with respect to 5.004. They
2612 have now been fixed.
2613
2614
2615 Version 1.09 28-Apr-98
2616 ----------------------
2617
2618 1. A negated single character class followed by a quantifier with a minimum
2619 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2620 program crashes, or just wrong answers. This did not apply to negated classes
2621 containing more than one character, or to minima other than one.
2622
2623
2624 Version 1.08 27-Mar-98
2625 ----------------------
2626
2627 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2628
2629 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2630 latter must appear before anything that relies on it in the pattern.
2631
2632
2633 Version 1.07 16-Feb-98
2634 ----------------------
2635
2636 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2637 repeat of a potentially empty string).
2638
2639
2640 Version 1.06 23-Jan-98
2641 ----------------------
2642
2643 1. Added Markus Oberhumer's little patches for C++.
2644
2645 2. Literal strings longer than 255 characters were broken.
2646
2647
2648 Version 1.05 23-Dec-97
2649 ----------------------
2650
2651 1. Negated character classes containing more than one character were failing if
2652 PCRE_CASELESS was set at run time.
2653
2654
2655 Version 1.04 19-Dec-97
2656 ----------------------
2657
2658 1. Corrected the man page, where some "const" qualifiers had been omitted.
2659
2660 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2661 input syntax.
2662
2663 3. Fixed memory leak which occurred when a regex with back references was
2664 matched with an offsets vector that wasn't big enough. The temporary memory
2665 that is used in this case wasn't being freed if the match failed.
2666
2667 4. Tidied pcretest to ensure it frees memory that it gets.
2668
2669 5. Temporary memory was being obtained in the case where the passed offsets
2670 vector was exactly big enough.
2671
2672 6. Corrected definition of offsetof() from change 5 below.
2673
2674 7. I had screwed up change 6 below and broken the rules for the use of
2675 setjmp(). Now fixed.
2676
2677
2678 Version 1.03 18-Dec-97
2679 ----------------------
2680
2681 1. A erroneous regex with a missing opening parenthesis was correctly
2682 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2683 on some systems.
2684
2685 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2686 it was reported that one broken compiler failed on the former because "code" is
2687 also an independent variable.
2688
2689 3. The erroneous regex a[]b caused an array overrun reference.
2690
2691 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2692 fail on data ending with that character. (It was going on too far, and checking
2693 the next character, typically a binary zero.) This was specific to the
2694 optimized code for single-character negative classes.
2695
2696 5. Added a contributed patch from the TIN world which does the following:
2697
2698 + Add an undef for memmove, in case the the system defines a macro for it.
2699
2700 + Add a definition of offsetof(), in case there isn't one. (I don't know
2701 the reason behind this - offsetof() is part of the ANSI standard - but
2702 it does no harm).
2703
2704 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2705 most of the places where whitespace preceded '#'. I have given up and
2706 allowed the remaining 2 cases to be at the margin.
2707
2708 + Rename some variables in pcre to eliminate shadowing. This seems very
2709 pedantic, but does no harm, of course.
2710
2711 6. Moved the call to setjmp() into its own function, to get rid of warnings
2712 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2713
2714 7. Constructs such as \d{8,} were compiling into the equivalent of
2715 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2716 outcome, but in this particular case used more store than had been allocated,
2717 which caused the bug to be discovered because it threw up an internal error.
2718
2719 8. The debugging code in both pcre and pcretest for outputting the compiled
2720 form of a regex was going wrong in the case of back references followed by
2721 curly-bracketed repeats.
2722
2723
2724 Version 1.02 12-Dec-97
2725 ----------------------
2726
2727 1. Typos in pcre.3 and comments in the source fixed.
2728
2729 2. Applied a contributed patch to get rid of places where it used to remove
2730 'const' from variables, and fixed some signed/unsigned and uninitialized
2731 variable warnings.
2732
2733 3. Added the "runtest" target to Makefile.
2734
2735 4. Set default compiler flag to -O2 rather than just -O.
2736
2737
2738 Version 1.01 19-Nov-97
2739 ----------------------
2740
2741 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2742 like /([ab]*)*/, that is, for classes with more than one character in them.
2743
2744 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2745 as /((?>a*))*/ (a PCRE_EXTRA facility).
2746
2747
2748 Version 1.00 18-Nov-97
2749 ----------------------
2750
2751 1. Added compile-time macros to support systems such as SunOS4 which don't have
2752 memmove() or strerror() but have other things that can be used instead.
2753
2754 2. Arranged that "make clean" removes the executables.
2755
2756
2757 Version 0.99 27-Oct-97
2758 ----------------------
2759
2760 1. Fixed bug in code for optimizing classes with only one character. It was
2761 initializing a 32-byte map regardless, which could cause it to run off the end
2762 of the memory it had got.
2763
2764 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2765
2766
2767 Version 0.98 22-Oct-97
2768 ----------------------
2769
2770 1. Fixed bug in code for handling temporary memory usage when there are more
2771 back references than supplied space in the ovector. This could cause segfaults.
2772
2773
2774 Version 0.97 21-Oct-97
2775 ----------------------
2776
2777 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2778
2779 2. Optimized negated single characters not to use a bit map.
2780
2781 3. Brought error texts together as macro definitions; clarified some of them;
2782 fixed one that was wrong - it said "range out of order" when it meant "invalid
2783 escape sequence".
2784
2785 4. Changed some char * arguments to const char *.
2786
2787 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2788
2789 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2790 pcretest.
2791
2792
2793 Version 0.96 16-Oct-97
2794 ----------------------
2795
2796 1. Added a simple "pgrep" utility to the distribution.
2797
2798 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2799 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2800 where "ddd" means "one or more decimal digits".
2801
2802 3. Fixed serious bug. If a pattern had a back reference, but the call to
2803 pcre_exec() didn't supply a large enough ovector to record the related
2804 identifying subpattern, the match always failed. PCRE now remembers the number
2805 of the largest back reference, and gets some temporary memory in which to save
2806 the offsets during matching if necessary, in order to ensure that
2807 backreferences always work.
2808
2809 4. Increased the compatibility with Perl in a number of ways:
2810
2811 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2812 to request this handling. The option can be set at compile or exec time.
2813
2814 (b) $ matches before a terminating newline by default; an option
2815 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2816 mode). The option can be set at compile or exec time.
2817
2818 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2819 the same as Perl's. If the decimal number it represents is less than 10
2820 or there aren't that many previous left capturing parentheses, an octal
2821 escape is read. Inside a character class, it's always an octal escape,
2822 even if it is a single digit.
2823
2824 (d) An escaped but undefined alphabetic character is taken as a literal,
2825 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2826 escapes.
2827
2828 (e) {0} is now permitted. (The previous item is removed from the compiled
2829 pattern).
2830
2831 5. Changed all the names of code files so that the basic parts are no longer
2832 than 10 characters, and abolished the teeny "globals.c" file.
2833
2834 6. Changed the handling of character classes; they are now done with a 32-byte
2835 bit map always.
2836
2837 7. Added the -d and /D options to pcretest to make it possible to look at the
2838 internals of compilation without having to recompile pcre.
2839
2840
2841 Version 0.95 23-Sep-97
2842 ----------------------
2843
2844 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2845 \x20 at the start of a run of normal characters. These were being treated as
2846 real characters, instead of the source characters being re-checked.
2847
2848
2849 Version 0.94 18-Sep-97
2850 ----------------------
2851
2852 1. The functions are now thread-safe, with the caveat that the global variables
2853 containing pointers to malloc() and free() or alternative functions are the
2854 same for all threads.
2855
2856 2. Get pcre_study() to generate a bitmap of initial characters for non-
2857 anchored patterns when this is possible, and use it if passed to pcre_exec().
2858
2859
2860 Version 0.93 15-Sep-97
2861 ----------------------
2862
2863 1. /(b)|(:+)/ was computing an incorrect first character.
2864
2865 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2866 but not actually doing anything yet.
2867
2868 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2869 as Perl does (e.g. [-az] or [az-]).
2870
2871 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2872 all possible positions.
2873
2874 5. Split up into different modules to avoid including unneeded functions in a
2875 compiled binary. However, compile and exec are still in one module. The "study"
2876 function is split off.
2877
2878 6. The character tables are now in a separate module whose source is generated
2879 by an auxiliary program - but can then be edited by hand if required. There are
2880 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2881 toupper() in the code.
2882
2883 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2884 make them global. Abolish the function for setting them, as the caller can now
2885 set them directly.
2886
2887
2888 Version 0.92 11-Sep-97
2889 ----------------------
2890
2891 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2892 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2893
2894 2. Caseless matching was not working in character classes if the characters in
2895 the pattern were in upper case.
2896
2897 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2898
2899 4. Make PCRE_ANCHORED public and accept as a compile option.
2900
2901 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2902 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2903 pass them.
2904
2905 6. Give an error if bad option bits passed at compile or run time.
2906
2907 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2908 pcretest to cause it to pass that flag.
2909
2910 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2911 options, and the first character, if set.
2912
2913 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2914
2915
2916 Version 0.91 10-Sep-97
2917 ----------------------
2918
2919 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2920 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2921
2922 2. PCRE was looping on encountering an indefinitely repeated back reference to
2923 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2924 Perl does - treats the match as successful.
2925
2926 ****

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