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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 211 - (show annotations) (download)
Thu Aug 9 09:52:43 2007 UTC (7 years, 3 months ago) by ph10
File size: 135354 byte(s)
Update UTF-8 validity check and documentation.

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

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