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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 205 - (show annotations) (download)
Fri Aug 3 13:18:33 2007 UTC (6 years, 11 months ago) by ph10
File size: 134202 byte(s)
Fix bugs with [\E] and [\Q\E].

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

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