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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 208 - (show annotations) (download)
Mon Aug 6 15:23:29 2007 UTC (7 years, 1 month ago) by ph10
File size: 134806 byte(s)
Added a pcresyntax man page; tidied some others.

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

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