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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 614 - (show annotations) (download)
Sat Jul 9 10:48:16 2011 UTC (3 years, 1 month ago) by ph10
File size: 205080 byte(s)
Fix atomic group and assertion capturing problems.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.13 30-Apr-2011
5 ------------------------
6
7 1. The Unicode data tables have been updated to Unicode 6.0.0.
8
9 2. Two minor typos in pcre_internal.h have been fixed.
10
11 3. Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12 pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13 in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14
15 4. There were a number of related bugs in the code for matching backrefences
16 caselessly in UTF-8 mode when codes for the characters concerned were
17 different numbers of bytes. For example, U+023A and U+2C65 are an upper
18 and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19 (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20 code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21 2-byte code at the end of the subject (it thought there wasn't enough data
22 left).
23
24 5. Comprehensive information about what went wrong is now returned by
25 pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26 as the output vector has at least 2 elements. The offset of the start of
27 the failing character and a reason code are placed in the vector.
28
29 6. When the UTF-8 string check fails for pcre_compile(), the offset that is
30 now returned is for the first byte of the failing character, instead of the
31 last byte inspected. This is an incompatible change, but I hope it is small
32 enough not to be a problem. It makes the returned offset consistent with
33 pcre_exec() and pcre_dfa_exec().
34
35 7. pcretest now gives a text phrase as well as the error number when
36 pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37 failure, the offset and reason code are output.
38
39 8. When \R was used with a maximizing quantifier it failed to skip backwards
40 over a \r\n pair if the subsequent match failed. Instead, it just skipped
41 back over a single character (\n). This seems wrong (because it treated the
42 two characters as a single entity when going forwards), conflicts with the
43 documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44 behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45 has been changed.
46
47 9. Some internal refactoring has changed the processing so that the handling
48 of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49 time (the PCRE_DOTALL option was changed this way some time ago: version
50 7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51 which was always a bit of a fudge. It also means that there is one less
52 argument for the match() function, which reduces its stack requirements
53 slightly. This change also fixes an incompatibility with Perl: the pattern
54 (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55
56 10. More internal refactoring has drastically reduced the number of recursive
57 calls to match() for possessively repeated groups such as (abc)++ when
58 using pcre_exec().
59
60 11. While implementing 10, a number of bugs in the handling of groups were
61 discovered and fixed:
62
63 (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64 (a|)*(?1) gave a compile-time internal error.
65 ((a|)+)+ did not notice that the outer group could match an empty string.
66 (^a|^)+ was not marked as anchored.
67 (.*a|.*)+ was not marked as matching at start or after a newline.
68
69 12. Yet more internal refactoring has removed another argument from the match()
70 function. Special calls to this function are now indicated by setting a
71 value in a variable in the "match data" data block.
72
73 13. Be more explicit in pcre_study() instead of relying on "default" for
74 opcodes that mean there is no starting character; this means that when new
75 ones are added and accidentally left out of pcre_study(), testing should
76 pick them up.
77
78 14. The -s option of pcretest has been documented for ages as being an old
79 synonym of -m (show memory usage). I have changed it to mean "force study
80 for every regex", that is, assume /S for every regex. This is similar to -i
81 and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82 using it. It makes it easier to run collections of tests with and without
83 study enabled, and thereby test pcre_study() more easily. All the standard
84 tests are now run with and without -s (but some patterns can be marked as
85 "never study" - see 20 below).
86
87 15. When (*ACCEPT) was used in a subpattern that was called recursively, the
88 restoration of the capturing data to the outer values was not happening
89 correctly.
90
91 16. If a recursively called subpattern ended with (*ACCEPT) and matched an
92 empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
93 pattern had matched an empty string, and so incorrectly returned a no
94 match.
95
96 17. There was optimizing code for the last branch of non-capturing parentheses,
97 and also for the obeyed branch of a conditional subexpression, which used
98 tail recursion to cut down on stack usage. Unfortunately, not that there is
99 the possibility of (*THEN) occurring in these branches, tail recursion is
100 no longer possible because the return has to be checked for (*THEN). These
101 two optimizations have therefore been removed.
102
103 18. If a pattern containing \R was studied, it was assumed that \R always
104 matched two bytes, thus causing the minimum subject length to be
105 incorrectly computed because \R can also match just one byte.
106
107 19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108 was incorrectly computed.
109
110 20. If /S is present twice on a test pattern in pcretest input, it *disables*
111 studying, thereby overriding the use of -s on the command line. This is
112 necessary for one or two tests to keep the output identical in both cases.
113
114 21. When (*ACCEPT) was used in an assertion that matched an empty string and
115 PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
116
117 22. When an atomic group that contained a capturing parenthesis was
118 successfully matched, but the branch in which it appeared failed, the
119 capturing was not being forgotten if a higher numbered group was later
120 captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
121 group 1 to "a", when in fact it should be unset. This applied to multi-
122 branched capturing and non-capturing groups, repeated or not, and also to
123 positive assertions (capturing in negative assertions is not well defined
124 in PCRE) and also to nested atomic groups.
125
126
127 Version 8.12 15-Jan-2011
128 ------------------------
129
130 1. Fixed some typos in the markup of the man pages, and wrote a script that
131 checks for such things as part of the documentation building process.
132
133 2. On a big-endian 64-bit system, pcregrep did not correctly process the
134 --match-limit and --recursion-limit options (added for 8.11). In
135 particular, this made one of the standard tests fail. (The integer value
136 went into the wrong half of a long int.)
137
138 3. If the --colour option was given to pcregrep with -v (invert match), it
139 did strange things, either producing crazy output, or crashing. It should,
140 of course, ignore a request for colour when reporting lines that do not
141 match.
142
143 4. Another pcregrep bug caused similar problems if --colour was specified with
144 -M (multiline) and the pattern match finished with a line ending.
145
146 5. In pcregrep, when a pattern that ended with a literal newline sequence was
147 matched in multiline mode, the following line was shown as part of the
148 match. This seems wrong, so I have changed it.
149
150 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
151 the check for further matches in the same line (so they could be coloured)
152 to overrun the end of the current line. If another match was found, it was
153 incorrectly shown (and then shown again when found in the next line).
154
155 7. If pcregrep was compiled under Windows, there was a reference to the
156 function pcregrep_exit() before it was defined. I am assuming this was
157 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
158 reported by a user. I've moved the definition above the reference.
159
160
161 Version 8.11 10-Dec-2010
162 ------------------------
163
164 1. (*THEN) was not working properly if there were untried alternatives prior
165 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
166 backtracked to try for "b" instead of moving to the next alternative branch
167 at the same level (in this case, to look for "c"). The Perl documentation
168 is clear that when (*THEN) is backtracked onto, it goes to the "next
169 alternative in the innermost enclosing group".
170
171 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
172 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
173 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
174 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
175 (*THEN).
176
177 3. If \s appeared in a character class, it removed the VT character from
178 the class, even if it had been included by some previous item, for example
179 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
180 of \s, but is part of the POSIX "space" class.)
181
182 4. A partial match never returns an empty string (because you can always
183 match an empty string at the end of the subject); however the checking for
184 an empty string was starting at the "start of match" point. This has been
185 changed to the "earliest inspected character" point, because the returned
186 data for a partial match starts at this character. This means that, for
187 example, /(?<=abc)def/ gives a partial match for the subject "abc"
188 (previously it gave "no match").
189
190 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
191 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
192 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
193 has an implication that the given string is incomplete (because a partial
194 match is preferred over a full match). For this reason, these items now
195 give a partial match in this situation. [Aside: previously, the one case
196 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
197 match rather than a full match, which was wrong by the old rules, but is
198 now correct.]
199
200 6. There was a bug in the handling of #-introduced comments, recognized when
201 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
202 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
203 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
204 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
205 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
206 places in pcre_compile().
207
208 7. Related to (6) above, when pcre_compile() was skipping #-introduced
209 comments when looking ahead for named forward references to subpatterns,
210 the only newline sequence it recognized was NL. It now handles newlines
211 according to the set newline convention.
212
213 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
214 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
215 cater for a lack of strerror(). These oversights have been fixed.
216
217 9. Added --match-limit and --recursion-limit to pcregrep.
218
219 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
220
221 11. When the -o option was used, pcregrep was setting a return code of 1, even
222 when matches were found, and --line-buffered was not being honoured.
223
224 12. Added an optional parentheses number to the -o and --only-matching options
225 of pcregrep.
226
227 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
228 can match an empty string. The code to do it in pcretest and pcredemo
229 needed fixing:
230
231 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
232 only one byte after an empty string match just before CRLF (this case
233 just got forgotten; "any" and "anycrlf" were OK).
234
235 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
236 mode when an empty string match preceded an ASCII character followed by
237 a non-ASCII character. (The code for advancing by one character rather
238 than one byte was nonsense.)
239
240 (c) The pcredemo.c sample program did not have any code at all to handle
241 the cases when CRLF is a valid newline sequence.
242
243 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
244 as a starting offset was within the subject string. There is now a new
245 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
246 negative or greater than the length of the string. In order to test this,
247 pcretest is extended to allow the setting of negative starting offsets.
248
249 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
250 starting offset points to the beginning of a UTF-8 character was
251 unnecessarily clumsy. I tidied it up.
252
253 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
254 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
255
256 17. Nobody had reported that the --include_dir option, which was added in
257 release 7.7 should have been called --include-dir (hyphen, not underscore)
258 for compatibility with GNU grep. I have changed it to --include-dir, but
259 left --include_dir as an undocumented synonym, and the same for
260 --exclude-dir, though that is not available in GNU grep, at least as of
261 release 2.5.4.
262
263 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
264 characters from a string of bytes) have been redefined so as not to use
265 loops, in order to improve performance in some environments. At the same
266 time, I abstracted some of the common code into auxiliary macros to save
267 repetition (this should not affect the compiled code).
268
269 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
270 compile-time error is now given if \c is not followed by an ASCII
271 character, that is, a byte less than 128. (In EBCDIC mode, the code is
272 different, and any byte value is allowed.)
273
274 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
275 START_OPTIMIZE option, which is now allowed at compile time - but just
276 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
277 to pcregrep and other applications that have no direct access to PCRE
278 options. The new /Y option in pcretest sets this option when calling
279 pcre_compile().
280
281 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
282 back references. Groups containing recursive back references were forced to
283 be atomic by that change, but in the case of named groups, the amount of
284 memory required was incorrectly computed, leading to "Failed: internal
285 error: code overflow". This has been fixed.
286
287 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
288 pcretest.c, to avoid build problems in some Borland environments.
289
290
291 Version 8.10 25-Jun-2010
292 ------------------------
293
294 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
295 THEN.
296
297 2. (*ACCEPT) was not working when inside an atomic group.
298
299 3. Inside a character class, \B is treated as a literal by default, but
300 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
301 causes the error). The code is unchanged, but I tidied the documentation.
302
303 4. Inside a character class, PCRE always treated \R and \X as literals,
304 whereas Perl faults them if its -w option is set. I have changed PCRE so
305 that it faults them when PCRE_EXTRA is set.
306
307 5. Added support for \N, which always matches any character other than
308 newline. (It is the same as "." when PCRE_DOTALL is not set.)
309
310 6. When compiling pcregrep with newer versions of gcc which may have
311 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
312 declared with attribute warn_unused_result" were given. Just casting the
313 result to (void) does not stop the warnings; a more elaborate fudge is
314 needed. I've used a macro to implement this.
315
316 7. Minor change to pcretest.c to avoid a compiler warning.
317
318 8. Added four artifical Unicode properties to help with an option to make
319 \s etc use properties (see next item). The new properties are: Xan
320 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
321
322 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
323 use Unicode properties. (*UCP) at the start of a pattern can be used to set
324 this option. Modified pcretest to add /W to test this facility. Added
325 REG_UCP to make it available via the POSIX interface.
326
327 10. Added --line-buffered to pcregrep.
328
329 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
330 studied, and the match started with a letter with a code point greater than
331 127 whose first byte was different to the first byte of the other case of
332 the letter, the other case of this starting letter was not recognized
333 (#976).
334
335 12. If a pattern that was studied started with a repeated Unicode property
336 test, for example, \p{Nd}+, there was the theoretical possibility of
337 setting up an incorrect bitmap of starting bytes, but fortunately it could
338 not have actually happened in practice until change 8 above was made (it
339 added property types that matched character-matching opcodes).
340
341 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
342 possible starting bytes for non-anchored patterns.
343
344 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
345 \R, and also a number of cases that involve Unicode properties, both
346 explicit and implicit when PCRE_UCP is set.
347
348 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
349 input, it could crash or give wrong results if characters with values
350 greater than 0xc0 were present in the subject string. (Detail: it assumed
351 UTF-8 input when processing these items.)
352
353 16. Added a lot of (int) casts to avoid compiler warnings in systems where
354 size_t is 64-bit (#991).
355
356 17. Added a check for running out of memory when PCRE is compiled with
357 --disable-stack-for-recursion (#990).
358
359 18. If the last data line in a file for pcretest does not have a newline on
360 the end, a newline was missing in the output.
361
362 19. The default pcre_chartables.c file recognizes only ASCII characters (values
363 less than 128) in its various bitmaps. However, there is a facility for
364 generating tables according to the current locale when PCRE is compiled. It
365 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
366 characters, are recognized by isspace() and therefore were getting set in
367 these tables, and indeed these tables seem to approximate to ISO 8859. This
368 caused a problem in UTF-8 mode when pcre_study() was used to create a list
369 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
370 which of course cannot start UTF-8 characters. I have changed the code so
371 that only real ASCII characters (less than 128) and the correct starting
372 bytes for UTF-8 encodings are set for characters greater than 127 when in
373 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
374 altogether.)
375
376 20. Added the /T option to pcretest so as to be able to run tests with non-
377 standard character tables, thus making it possible to include the tests
378 used for 19 above in the standard set of tests.
379
380 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
381 reference to a subpattern the other side of a comment that contains an
382 opening parenthesis caused either an internal compiling error, or a
383 reference to the wrong subpattern.
384
385
386 Version 8.02 19-Mar-2010
387 ------------------------
388
389 1. The Unicode data tables have been updated to Unicode 5.2.0.
390
391 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
392 configured.
393
394 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
395 original author of that file, following a query about its status.
396
397 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
398 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
399
400 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
401 quantifier applied to a forward-referencing subroutine call, could compile
402 incorrect code or give the error "internal error: previously-checked
403 referenced subpattern not found".
404
405 6. Both MS Visual Studio and Symbian OS have problems with initializing
406 variables to point to external functions. For these systems, therefore,
407 pcre_malloc etc. are now initialized to local functions that call the
408 relevant global functions.
409
410 7. There were two entries missing in the vectors called coptable and poptable
411 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
412 I've fixed the data, and added a kludgy way of testing at compile time that
413 the lengths are correct (equal to the number of opcodes).
414
415 8. Following on from 7, I added a similar kludge to check the length of the
416 eint vector in pcreposix.c.
417
418 9. Error texts for pcre_compile() are held as one long string to avoid too
419 much relocation at load time. To find a text, the string is searched,
420 counting zeros. There was no check for running off the end of the string,
421 which could happen if a new error number was added without updating the
422 string.
423
424 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
425
426 11. \K was not working if it appeared in an atomic group or in a group that
427 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
428 \K is "not well defined" if used in an assertion. PCRE now accepts it if
429 the assertion is positive, but not if it is negative.
430
431 12. Change 11 fortuitously reduced the size of the stack frame used in the
432 "match()" function of pcre_exec.c by one pointer. Forthcoming
433 implementation of support for (*MARK) will need an extra pointer on the
434 stack; I have reserved it now, so that the stack frame size does not
435 decrease.
436
437 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
438 item in branch that calls a recursion is a subroutine call - as in the
439 second branch in the above example - was incorrectly given the compile-
440 time error "recursive call could loop indefinitely" because pcre_compile()
441 was not correctly checking the subroutine for matching a non-empty string.
442
443 14. The checks for overrunning compiling workspace could trigger after an
444 overrun had occurred. This is a "should never occur" error, but it can be
445 triggered by pathological patterns such as hundreds of nested parentheses.
446 The checks now trigger 100 bytes before the end of the workspace.
447
448 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
449
450
451 Version 8.01 19-Jan-2010
452 ------------------------
453
454 1. If a pattern contained a conditional subpattern with only one branch (in
455 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
456 computed the wrong minimum data length (which is of course zero for such
457 subpatterns). This could cause incorrect "no match" results.
458
459 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
460 the pattern is reset in the first branch, pcre_compile() failed with
461 "internal error: code overflow at offset...". This happened only when
462 the reset was to the original external option setting. (An optimization
463 abstracts leading options settings into an external setting, which was the
464 cause of this.)
465
466 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
467 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
468 assertion pattern did not match (meaning that the assertion was true), it
469 was incorrectly treated as false if the SKIP had been reached during the
470 matching. This also applied to assertions used as conditions.
471
472 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
473 assertion subpattern, including such a pattern used as a condition,
474 unpredictable results occurred, instead of the error return
475 PCRE_ERROR_DFA_UITEM.
476
477 5. The C++ GlobalReplace function was not working like Perl for the special
478 situation when an empty string is matched. It now does the fancy magic
479 stuff that is necessary.
480
481 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
482 removed. (These were left over from very, very early versions of PCRE.)
483
484 7. Some cosmetic changes to the code to make life easier when compiling it
485 as part of something else:
486
487 (a) Change DEBUG to PCRE_DEBUG.
488
489 (b) In pcre_compile(), rename the member of the "branch_chain" structure
490 called "current" as "current_branch", to prevent a collision with the
491 Linux macro when compiled as a kernel module.
492
493 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
494 prevent a collision with the Linux macro when compiled as a kernel
495 module.
496
497 8. In pcre_compile() there are some checks for integer overflows that used to
498 cast potentially large values to (double). This has been changed to that
499 when building, a check for int64_t is made, and if it is found, it is used
500 instead, thus avoiding the use of floating point arithmetic. (There is no
501 other use of FP in PCRE.) If int64_t is not found, the fallback is to
502 double.
503
504 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
505 2005 (difference between two addresses compared to an unsigned value).
506
507 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
508 custom one, because of the following reported problem in Windows:
509
510 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
511 under Win32.
512 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
513 therefore missing the function definition.
514 - The compiler thus generates a "C" signature for the test function.
515 - The linker fails to find the "C" function.
516 - PCRE fails to configure if asked to do so against libbz2.
517
518 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
519 messages were output:
520
521 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
522 rerunning libtoolize, to keep the correct libtool macros in-tree.
523 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
524
525 I have done both of these things.
526
527 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
528 most of the time, it *can* run out if it is given a pattern that contains a
529 runaway infinite recursion. I updated the discussion in the pcrestack man
530 page.
531
532 13. Now that we have gone to the x.xx style of version numbers, the minor
533 version may start with zero. Using 08 or 09 is a bad idea because users
534 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
535 interpreted as invalid octal numbers. I've updated the previous comment in
536 configure.ac, and also added a check that gives an error if 08 or 09 are
537 used.
538
539 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
540 causing partial matching to fail when the end of the subject matched \W
541 in a UTF-8 pattern where \W was quantified with a minimum of 3.
542
543 15. There were some discrepancies between the declarations in pcre_internal.h
544 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
545 their definitions. The declarations used "const uschar *" and the
546 definitions used USPTR. Even though USPTR is normally defined as "const
547 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
548 reported that: "This difference in casting confuses some C++ compilers, for
549 example, SunCC recognizes above declarations as different functions and
550 generates broken code for hbpcre." I have changed the declarations to use
551 USPTR.
552
553 16. GNU libtool is named differently on some systems. The autogen.sh script now
554 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
555 (FreeBSD).
556
557 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
558 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
559 comment: "Figure out how to create a longlong from a string: strtoll and
560 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
561 instance, but it only takes 2 args instead of 3!"
562
563 18. A subtle bug concerned with back references has been fixed by a change of
564 specification, with a corresponding code fix. A pattern such as
565 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
566 refers, was giving matches when it shouldn't. For example, xa=xaaa would
567 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
568 same bug. Such groups have to be quantified to be useful, or contained
569 inside another quantified group. (If there's no repetition, the reference
570 can never match.) The problem arises because, having left the group and
571 moved on to the rest of the pattern, a later failure that backtracks into
572 the group uses the captured value from the final iteration of the group
573 rather than the correct earlier one. I have fixed this in PCRE by forcing
574 any group that contains a reference to itself to be an atomic group; that
575 is, there cannot be any backtracking into it once it has completed. This is
576 similar to recursive and subroutine calls.
577
578
579 Version 8.00 19-Oct-09
580 ----------------------
581
582 1. The table for translating pcre_compile() error codes into POSIX error codes
583 was out-of-date, and there was no check on the pcre_compile() error code
584 being within the table. This could lead to an OK return being given in
585 error.
586
587 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
588 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
589 in a Windows environment.
590
591 3. The pcregrep --count option prints the count for each file even when it is
592 zero, as does GNU grep. However, pcregrep was also printing all files when
593 --files-with-matches was added. Now, when both options are given, it prints
594 counts only for those files that have at least one match. (GNU grep just
595 prints the file name in this circumstance, but including the count seems
596 more useful - otherwise, why use --count?) Also ensured that the
597 combination -clh just lists non-zero counts, with no names.
598
599 4. The long form of the pcregrep -F option was incorrectly implemented as
600 --fixed_strings instead of --fixed-strings. This is an incompatible change,
601 but it seems right to fix it, and I didn't think it was worth preserving
602 the old behaviour.
603
604 5. The command line items --regex=pattern and --regexp=pattern were not
605 recognized by pcregrep, which required --regex pattern or --regexp pattern
606 (with a space rather than an '='). The man page documented the '=' forms,
607 which are compatible with GNU grep; these now work.
608
609 6. No libpcreposix.pc file was created for pkg-config; there was just
610 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
611
612 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
613 when UCP support is not needed, by modifying the Python script that
614 generates it from Unicode data files. This should not matter if the module
615 is correctly used as a library, but I received one complaint about 50K of
616 unwanted data. My guess is that the person linked everything into his
617 program rather than using a library. Anyway, it does no harm.
618
619 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
620 was a minimum greater than 1 for a wide character in a possessive
621 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
622 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
623 character. Chaos in the form of incorrect output or a compiling loop could
624 result.
625
626 9. The restrictions on what a pattern can contain when partial matching is
627 requested for pcre_exec() have been removed. All patterns can now be
628 partially matched by this function. In addition, if there are at least two
629 slots in the offset vector, the offset of the earliest inspected character
630 for the match and the offset of the end of the subject are set in them when
631 PCRE_ERROR_PARTIAL is returned.
632
633 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
634 synonymous with PCRE_PARTIAL, for backwards compatibility, and
635 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
636 and may be more useful for multi-segment matching.
637
638 11. Partial matching with pcre_exec() is now more intuitive. A partial match
639 used to be given if ever the end of the subject was reached; now it is
640 given only if matching could not proceed because another character was
641 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
642 string "Z", which now yields "no match" instead of "partial match". In the
643 case of pcre_dfa_exec(), "no match" is given if every matching path for the
644 final character ended with (*FAIL).
645
646 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
647 if the pattern had a "must contain" character that was already found in the
648 earlier partial match, unless partial matching was again requested. For
649 example, with the pattern /dog.(body)?/, the "must contain" character is
650 "g". If the first part-match was for the string "dog", restarting with
651 "sbody" failed. This bug has been fixed.
652
653 13. The string returned by pcre_dfa_exec() after a partial match has been
654 changed so that it starts at the first inspected character rather than the
655 first character of the match. This makes a difference only if the pattern
656 starts with a lookbehind assertion or \b or \B (\K is not supported by
657 pcre_dfa_exec()). It's an incompatible change, but it makes the two
658 matching functions compatible, and I think it's the right thing to do.
659
660 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
661 so that the demonstration program is easily available in environments where
662 PCRE has not been installed from source.
663
664 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
665 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
666 library.
667
668 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
669 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
670 is not the first non-POSIX option to be added. Clearly some people find
671 these options useful.
672
673 17. If a caller to the POSIX matching function regexec() passes a non-zero
674 value for nmatch with a NULL value for pmatch, the value of
675 nmatch is forced to zero.
676
677 18. RunGrepTest did not have a test for the availability of the -u option of
678 the diff command, as RunTest does. It now checks in the same way as
679 RunTest, and also checks for the -b option.
680
681 19. If an odd number of negated classes containing just a single character
682 interposed, within parentheses, between a forward reference to a named
683 subpattern and the definition of the subpattern, compilation crashed with
684 an internal error, complaining that it could not find the referenced
685 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
686 [The bug was that it was starting one character too far in when skipping
687 over the character class, thus treating the ] as data rather than
688 terminating the class. This meant it could skip too much.]
689
690 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
691 /g option in pcretest when the pattern contains \K, which makes it possible
692 to have an empty string match not at the start, even when the pattern is
693 anchored. Updated pcretest and pcredemo to use this option.
694
695 21. If the maximum number of capturing subpatterns in a recursion was greater
696 than the maximum at the outer level, the higher number was returned, but
697 with unset values at the outer level. The correct (outer level) value is
698 now given.
699
700 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
701 PCRE did not set those parentheses (unlike Perl). I have now found a way to
702 make it do so. The string so far is captured, making this feature
703 compatible with Perl.
704
705 23. The tests have been re-organized, adding tests 11 and 12, to make it
706 possible to check the Perl 5.10 features against Perl 5.10.
707
708 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
709 pattern matches a fixed length string. PCRE did not allow this; now it
710 does. Neither allows recursion.
711
712 25. I finally figured out how to implement a request to provide the minimum
713 length of subject string that was needed in order to match a given pattern.
714 (It was back references and recursion that I had previously got hung up
715 on.) This code has now been added to pcre_study(); it finds a lower bound
716 to the length of subject needed. It is not necessarily the greatest lower
717 bound, but using it to avoid searching strings that are too short does give
718 some useful speed-ups. The value is available to calling programs via
719 pcre_fullinfo().
720
721 26. While implementing 25, I discovered to my embarrassment that pcretest had
722 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
723 study optimizations had never been tested with that matching function.
724 Oops. What is worse, even when it was passed study data, there was a bug in
725 pcre_dfa_exec() that meant it never actually used it. Double oops. There
726 were also very few tests of studied patterns with pcre_dfa_exec().
727
728 27. If (?| is used to create subpatterns with duplicate numbers, they are now
729 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
730 on the other side of the coin, they are no longer allowed to have different
731 names, because these cannot be distinguished in PCRE, and this has caused
732 confusion. (This is a difference from Perl.)
733
734 28. When duplicate subpattern names are present (necessarily with different
735 numbers, as required by 27 above), and a test is made by name in a
736 conditional pattern, either for a subpattern having been matched, or for
737 recursion in such a pattern, all the associated numbered subpatterns are
738 tested, and the overall condition is true if the condition is true for any
739 one of them. This is the way Perl works, and is also more like the way
740 testing by number works.
741
742
743 Version 7.9 11-Apr-09
744 ---------------------
745
746 1. When building with support for bzlib/zlib (pcregrep) and/or readline
747 (pcretest), all targets were linked against these libraries. This included
748 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
749 libraries. This caused unwanted dependencies to be created. This problem
750 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
751 pcretest is linked with readline.
752
753 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
754 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
755 moved outside it again, because FALSE and TRUE are already defined in AIX,
756 but BOOL is not.
757
758 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
759 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
760
761 4. The pcregrep documentation said spaces were inserted as well as colons (or
762 hyphens) following file names and line numbers when outputting matching
763 lines. This is not true; no spaces are inserted. I have also clarified the
764 wording for the --colour (or --color) option.
765
766 5. In pcregrep, when --colour was used with -o, the list of matching strings
767 was not coloured; this is different to GNU grep, so I have changed it to be
768 the same.
769
770 6. When --colo(u)r was used in pcregrep, only the first matching substring in
771 each matching line was coloured. Now it goes on to look for further matches
772 of any of the test patterns, which is the same behaviour as GNU grep.
773
774 7. A pattern that could match an empty string could cause pcregrep to loop; it
775 doesn't make sense to accept an empty string match in pcregrep, so I have
776 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
777 seems to be how GNU grep behaves.
778
779 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
780 start or after a newline", because the conditional assertion was not being
781 correctly handled. The rule now is that both the assertion and what follows
782 in the first alternative must satisfy the test.
783
784 9. If auto-callout was enabled in a pattern with a conditional group whose
785 condition was an assertion, PCRE could crash during matching, both with
786 pcre_exec() and pcre_dfa_exec().
787
788 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
789 used for matching.
790
791 11. Unicode property support in character classes was not working for
792 characters (bytes) greater than 127 when not in UTF-8 mode.
793
794 12. Added the -M command line option to pcretest.
795
796 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
797
798 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
799
800 16. Added comments and documentation about mis-use of no_arg in the C++
801 wrapper.
802
803 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
804 from Martin Jerabek that uses macro names for all relevant character and
805 string constants.
806
807 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
808 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
809 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
810 these, but not everybody uses configure.
811
812 19. A conditional group that had only one branch was not being correctly
813 recognized as an item that could match an empty string. This meant that an
814 enclosing group might also not be so recognized, causing infinite looping
815 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
816 with the subject "ab", where knowledge that the repeated group can match
817 nothing is needed in order to break the loop.
818
819 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
820 exec(), but without supplying a callout function, matching went wrong.
821
822 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
823 leak if the size of the offset vector was greater than 30. When the vector
824 is smaller, the saved offsets during recursion go onto a local stack
825 vector, but for larger vectors malloc() is used. It was failing to free
826 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
827 error, in fact).
828
829 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
830 heapframe that is used only when UTF-8 support is enabled. This caused no
831 problem, but was untidy.
832
833 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
834 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
835 included within another project.
836
837 24. Steven Van Ingelgem's patches to add more options to the CMake support,
838 slightly modified by me:
839
840 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
841 not building pcregrep.
842
843 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
844 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
845
846 25. Forward references, both numeric and by name, in patterns that made use of
847 duplicate group numbers, could behave incorrectly or give incorrect errors,
848 because when scanning forward to find the reference group, PCRE was not
849 taking into account the duplicate group numbers. A pattern such as
850 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
851
852 26. Changed a few more instances of "const unsigned char *" to USPTR, making
853 the feature of a custom pointer more persuasive (as requested by a user).
854
855 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
856 pcretest.c, inside #ifndefs, because it seems they are sometimes already
857 pre-defined.
858
859 28. Added support for (*UTF8) at the start of a pattern.
860
861 29. Arrange for flags added by the "release type" setting in CMake to be shown
862 in the configuration summary.
863
864
865 Version 7.8 05-Sep-08
866 ---------------------
867
868 1. Replaced UCP searching code with optimized version as implemented for Ad
869 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
870 stage table and inline lookup instead of a function, giving speed ups of 2
871 to 5 times on some simple patterns that I tested. Permission was given to
872 distribute the MultiStage2.py script that generates the tables (it's not in
873 the tarball, but is in the Subversion repository).
874
875 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
876 scripts.
877
878 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
879 a group with a zero qualifier. The result of the study could be incorrect,
880 or the function might crash, depending on the pattern.
881
882 4. Caseless matching was not working for non-ASCII characters in back
883 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
884 It now works when Unicode Property Support is available.
885
886 5. In pcretest, an escape such as \x{de} in the data was always generating
887 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
888 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
889 truncation.
890
891 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
892
893 7. Added two (int) casts to pcregrep when printing the difference of two
894 pointers, in case they are 64-bit values.
895
896 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
897 test 2 if it fails.
898
899 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
900 and a #define of that name to empty if it is not externally set. This is to
901 allow users of MSVC to set it if necessary.
902
903 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
904 the convenience functions in the pcre_get.c source file.
905
906 11. An option change at the start of a pattern that had top-level alternatives
907 could cause overwriting and/or a crash. This command provoked a crash in
908 some environments:
909
910 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
911
912 This potential security problem was recorded as CVE-2008-2371.
913
914 12. For a pattern where the match had to start at the beginning or immediately
915 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
916 pcre_dfa_exec() could read past the end of the passed subject if there was
917 no match. To help with detecting such bugs (e.g. with valgrind), I modified
918 pcretest so that it places the subject at the end of its malloc-ed buffer.
919
920 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
921 exec() might read past the end of the data buffer in UTF-8 mode.
922
923 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
924 the data contained the byte 0x85 as part of a UTF-8 character within its
925 first line. This applied both to normal and DFA matching.
926
927 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
928 /^[^d]*?$/8 failed to match "abc".
929
930 16. Added a missing copyright notice to pcrecpp_internal.h.
931
932 17. Make it more clear in the documentation that values returned from
933 pcre_exec() in ovector are byte offsets, not character counts.
934
935 18. Tidied a few places to stop certain compilers from issuing warnings.
936
937 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
938 supplied by Stefan Weber. I made a further small update for 7.8 because
939 there is a change of source arrangements: the pcre_searchfuncs.c module is
940 replaced by pcre_ucd.c.
941
942
943 Version 7.7 07-May-08
944 ---------------------
945
946 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
947 a string to a long long, pretend we don't even have a long long." This is
948 done by checking for the strtoq, strtoll, and _strtoi64 functions.
949
950 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
951 pre-7.6 versions, which defined a global no_arg variable instead of putting
952 it in the RE class. (See also #8 below.)
953
954 3. Remove a line of dead code, identified by coverity and reported by Nuno
955 Lopes.
956
957 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
958
959 (1) The include/exclude patterns were being applied to the whole pathnames
960 of files, instead of just to the final components.
961
962 (2) If there was more than one level of directory, the subdirectories were
963 skipped unless they satisfied the include/exclude conditions. This is
964 inconsistent with GNU grep (and could even be seen as contrary to the
965 pcregrep specification - which I improved to make it absolutely clear).
966 The action now is always to scan all levels of directory, and just
967 apply the include/exclude patterns to regular files.
968
969 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
970 --exclude_dir in the tests to avoid scanning .svn directories.
971
972 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
973 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
974 doesn't support NULs in patterns.
975
976 7. Added some missing "const"s to declarations of static tables in
977 pcre_compile.c and pcre_dfa_exec.c.
978
979 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
980 caused by fix #2 above. (Subsequently also a second patch to fix the
981 first patch. And a third patch - this was a messy problem.)
982
983 9. Applied Craig's patch to remove the use of push_back().
984
985 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
986 matching function regexec().
987
988 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
989 which, however, unlike Perl's \g{...}, are subroutine calls, not back
990 references. PCRE supports relative numbers with this syntax (I don't think
991 Oniguruma does).
992
993 12. Previously, a group with a zero repeat such as (...){0} was completely
994 omitted from the compiled regex. However, this means that if the group
995 was called as a subroutine from elsewhere in the pattern, things went wrong
996 (an internal error was given). Such groups are now left in the compiled
997 pattern, with a new opcode that causes them to be skipped at execution
998 time.
999
1000 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1001 to the way PCRE behaves:
1002
1003 (a) A lone ] character is dis-allowed (Perl treats it as data).
1004
1005 (b) A back reference to an unmatched subpattern matches an empty string
1006 (Perl fails the current match path).
1007
1008 (c) A data ] in a character class must be notated as \] because if the
1009 first data character in a class is ], it defines an empty class. (In
1010 Perl it is not possible to have an empty class.) The empty class []
1011 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1012 The negative empty class [^] matches any one character, independently
1013 of the DOTALL setting.
1014
1015 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1016 non-existent subpattern following a character class starting with ']' and
1017 containing () gave an internal compiling error instead of "reference to
1018 non-existent subpattern". Fortunately, when the pattern did exist, the
1019 compiled code was correct. (When scanning forwards to check for the
1020 existencd of the subpattern, it was treating the data ']' as terminating
1021 the class, so got the count wrong. When actually compiling, the reference
1022 was subsequently set up correctly.)
1023
1024 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1025 it was being rejected as not supported by pcre_dfa_exec(), even though
1026 other assertions are supported. I have made pcre_dfa_exec() support
1027 (*FAIL).
1028
1029 16. The implementation of 13c above involved the invention of a new opcode,
1030 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1031 cannot be changed at match time, I realized I could make a small
1032 improvement to matching performance by compiling OP_ALLANY instead of
1033 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1034 on the OP_ANY path.
1035
1036 17. Compiling pcretest on Windows with readline support failed without the
1037 following two fixes: (1) Make the unistd.h include conditional on
1038 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1039
1040 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1041 ncurses library to be included for pcretest when ReadLine support is
1042 requested, but also to allow for it to be overridden. This patch came from
1043 Daniel Bergström.
1044
1045 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1046 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1047 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1048 spotting this.
1049
1050
1051 Version 7.6 28-Jan-08
1052 ---------------------
1053
1054 1. A character class containing a very large number of characters with
1055 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1056 overflow.
1057
1058 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1059 HAVE_LONG_LONG is not defined.
1060
1061 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1062 bring them up to date and include new features. This patch includes:
1063
1064 - Fixed PH's badly added libz and libbz2 support.
1065 - Fixed a problem with static linking.
1066 - Added pcredemo. [But later removed - see 7 below.]
1067 - Fixed dftables problem and added an option.
1068 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1069 HAVE_LONG_LONG.
1070 - Added readline support for pcretest.
1071 - Added an listing of the option settings after cmake has run.
1072
1073 4. A user submitted a patch to Makefile that makes it easy to create
1074 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1075 Makefile.am that cause it to include this special target, without
1076 affecting anything else. Note that the same mingw target plus all
1077 the other distribution libraries and programs are now supported
1078 when configuring with CMake (see 6 below) instead of with
1079 Configure/Make.
1080
1081 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1082 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1083 exported in the Windows port". It has not yet been confirmed that the patch
1084 solves the problem, but it does no harm.
1085
1086 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1087 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1088 with CMake, and also correct the comment about stack recursion.
1089
1090 7. Remove the automatic building of pcredemo from the ./configure system and
1091 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1092 of a program that users should build themselves after PCRE is installed, so
1093 building it automatically is not really right. What is more, it gave
1094 trouble in some build environments.
1095
1096 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1097
1098
1099 Version 7.5 10-Jan-08
1100 ---------------------
1101
1102 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1103 values in parens when parsing an RE using the C++ wrapper."
1104
1105 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1106 Characters greater than 255 were excluded from the class instead of being
1107 included.
1108
1109 3. The same bug as (2) above applied to negated POSIX classes such as
1110 [:^space:].
1111
1112 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1113 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1114 I have changed it.
1115
1116 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1117 first named subpattern) and a construct such as (?&a) would reference the
1118 first named subpattern whose name started with "a" (in other words, the
1119 length check was missing). Both these problems are fixed. "Subpattern name
1120 expected" is now given for (?&) (a zero-length name), and this patch also
1121 makes it give the same error for \k'' (previously it complained that that
1122 was a reference to a non-existent subpattern).
1123
1124 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1125 this is right because (?- can be followed by option settings as well as by
1126 digits. I have, however, made the messages clearer.
1127
1128 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1129 than the number used in the conditional) now cause a compile-time error.
1130 This is actually not compatible with Perl, which accepts such patterns, but
1131 treats the conditional as always being FALSE (as PCRE used to), but it
1132 seems to me that giving a diagnostic is better.
1133
1134 8. Change "alphameric" to the more common word "alphanumeric" in comments
1135 and messages.
1136
1137 9. Fix two occurrences of "backslash" in comments that should have been
1138 "backspace".
1139
1140 10. Remove two redundant lines of code that can never be obeyed (their function
1141 was moved elsewhere).
1142
1143 11. The program that makes PCRE's Unicode character property table had a bug
1144 which caused it to generate incorrect table entries for sequences of
1145 characters that have the same character type, but are in different scripts.
1146 It amalgamated them into a single range, with the script of the first of
1147 them. In other words, some characters were in the wrong script. There were
1148 thirteen such cases, affecting characters in the following ranges:
1149
1150 U+002b0 - U+002c1
1151 U+0060c - U+0060d
1152 U+0061e - U+00612
1153 U+0064b - U+0065e
1154 U+0074d - U+0076d
1155 U+01800 - U+01805
1156 U+01d00 - U+01d77
1157 U+01d9b - U+01dbf
1158 U+0200b - U+0200f
1159 U+030fc - U+030fe
1160 U+03260 - U+0327f
1161 U+0fb46 - U+0fbb1
1162 U+10450 - U+1049d
1163
1164 12. The -o option (show only the matching part of a line) for pcregrep was not
1165 compatible with GNU grep in that, if there was more than one match in a
1166 line, it showed only the first of them. It now behaves in the same way as
1167 GNU grep.
1168
1169 13. If the -o and -v options were combined for pcregrep, it printed a blank
1170 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1171 does the same. The return code can be used to tell if there were any
1172 non-matching lines.
1173
1174 14. Added --file-offsets and --line-offsets to pcregrep.
1175
1176 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1177 infinitely looping recursion. The bug was that positive lookaheads were not
1178 being skipped when checking for a possible empty match (negative lookaheads
1179 and both kinds of lookbehind were skipped).
1180
1181 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1182 inclusion of <windows.h> to before rather than after the definition of
1183 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1184
1185 17. Specifying a possessive quantifier with a specific limit for a Unicode
1186 character property caused pcre_compile() to compile bad code, which led at
1187 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1188 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1189 caused the error; without that there was no problem.
1190
1191 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1192
1193 19. Added --enable-pcretest-libreadline.
1194
1195 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1196 RE::GlobalReplace(). As a result, the number of replacements returned was
1197 double what it should be. I removed one of the increments, but Craig sent a
1198 later patch that removed the other one (the right fix) and added unit tests
1199 that check the return values (which was not done before).
1200
1201 21. Several CMake things:
1202
1203 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1204 the names libpcre and libpcreposix, not just pcre and pcreposix.
1205
1206 (2) The above change means that pcretest and pcregrep are now correctly
1207 linked with the newly-built libraries, not previously installed ones.
1208
1209 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1210
1211 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1212 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1213 UTF-8 newline character). The key issue is that the pattern starts .*;
1214 this means that the match must be either at the beginning, or after a
1215 newline. The bug was in the code for advancing after a failed match and
1216 checking that the new position followed a newline. It was not taking
1217 account of UTF-8 characters correctly.
1218
1219 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1220 character classes. PCRE was not treating the sequence [:...:] as a
1221 character class unless the ... were all letters. Perl, however, seems to
1222 allow any characters between [: and :], though of course it rejects as
1223 unknown any "names" that contain non-letters, because all the known class
1224 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1225 for example, whereas PCRE did not - it did not recognize a POSIX character
1226 class. This seemed a bit dangerous, so the code has been changed to be
1227 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1228 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1229 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1230 Perl does, and where it didn't before.
1231
1232 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1233 Windows environments %n is disabled by default.
1234
1235
1236 Version 7.4 21-Sep-07
1237 ---------------------
1238
1239 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1240 means that a class such as [\s] counted as "explicit reference to CR or
1241 LF". That isn't really right - the whole point of the change was to try to
1242 help when there was an actual mention of one of the two characters. So now
1243 the change happens only if \r or \n (or a literal CR or LF) character is
1244 encountered.
1245
1246 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1247 of both had grown to the point where there were only 3 bits left.
1248 Fortunately, there was spare space in the data structure, and so I have
1249 moved the internal flags into a new 16-bit field to free up more option
1250 bits.
1251
1252 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1253 but did not set the internal JCHANGED flag - either of these is enough to
1254 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1255 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1256 start sets both bits.
1257
1258 4. Added options (at build time, compile time, exec time) to change \R from
1259 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1260
1261 5. doc/pcresyntax.html was missing from the distribution.
1262
1263 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1264 compatibility, even though it is no longer used.
1265
1266 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1267 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1268 windows.h file is present (where different names are used). [This was
1269 reversed later after testing - see 16 below.]
1270
1271 8. Changed all #include <config.h> to #include "config.h". There were also
1272 some further <pcre.h> cases that I changed to "pcre.h".
1273
1274 9. When pcregrep was used with the --colour option, it missed the line ending
1275 sequence off the lines that it output.
1276
1277 10. It was pointed out to me that arrays of string pointers cause lots of
1278 relocations when a shared library is dynamically loaded. A technique of
1279 using a single long string with a table of offsets can drastically reduce
1280 these. I have refactored PCRE in four places to do this. The result is
1281 dramatic:
1282
1283 Originally: 290
1284 After changing UCP table: 187
1285 After changing error message table: 43
1286 After changing table of "verbs" 36
1287 After changing table of Posix names 22
1288
1289 Thanks to the folks working on Gregex for glib for this insight.
1290
1291 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1292 unicode-properties was also set.
1293
1294 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1295
1296 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1297 checked only for CRLF.
1298
1299 14. Added casts to pcretest.c to avoid compiler warnings.
1300
1301 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1302
1303 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1304 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1305 entirely. This removes changes made in 7 above.
1306
1307 17. The CMake files have been updated, and there is now more information about
1308 building with CMake in the NON-UNIX-USE document.
1309
1310
1311 Version 7.3 28-Aug-07
1312 ---------------------
1313
1314 1. In the rejigging of the build system that eventually resulted in 7.1, the
1315 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1316 brackets there is not right, since it causes compilers to look for an
1317 installed pcre.h, not the version that is in the source that is being
1318 compiled (which of course may be different). I have changed it back to:
1319
1320 #include "pcre.h"
1321
1322 I have a vague recollection that the change was concerned with compiling in
1323 different directories, but in the new build system, that is taken care of
1324 by the VPATH setting the Makefile.
1325
1326 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1327 when the subject happened to end in the byte 0x85 (e.g. if the last
1328 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1329 characters but of course it shouldn't be taken as a newline when it is part
1330 of another character. The bug was that, for an unlimited repeat of . in
1331 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1332 characters when looking for a newline.
1333
1334 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1335
1336 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1337 in debug output.
1338
1339 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1340 long printing in the pcrecpp unittest when running under MinGW.
1341
1342 6. ESC_K was left out of the EBCDIC table.
1343
1344 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1345 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1346 limit also applies to "virtual nesting" when a pattern is recursive, and in
1347 this case 1000 isn't so big. I have been able to remove this limit at the
1348 expense of backing off one optimization in certain circumstances. Normally,
1349 when pcre_exec() would call its internal match() function recursively and
1350 immediately return the result unconditionally, it uses a "tail recursion"
1351 feature to save stack. However, when a subpattern that can match an empty
1352 string has an unlimited repetition quantifier, it no longer makes this
1353 optimization. That gives it a stack frame in which to save the data for
1354 checking that an empty string has been matched. Previously this was taken
1355 from the 1000-entry workspace that had been reserved. So now there is no
1356 explicit limit, but more stack is used.
1357
1358 8. Applied Daniel's patches to solve problems with the import/export magic
1359 syntax that is required for Windows, and which was going wrong for the
1360 pcreposix and pcrecpp parts of the library. These were overlooked when this
1361 problem was solved for the main library.
1362
1363 9. There were some crude static tests to avoid integer overflow when computing
1364 the size of patterns that contain repeated groups with explicit upper
1365 limits. As the maximum quantifier is 65535, the maximum group length was
1366 set at 30,000 so that the product of these two numbers did not overflow a
1367 32-bit integer. However, it turns out that people want to use groups that
1368 are longer than 30,000 bytes (though not repeat them that many times).
1369 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1370 made it possible to implement the integer overflow checks in a much more
1371 dynamic way, which I have now done. The artificial limitation on group
1372 length has been removed - we now have only the limit on the total length of
1373 the compiled pattern, which depends on the LINK_SIZE setting.
1374
1375 10. Fixed a bug in the documentation for get/copy named substring when
1376 duplicate names are permitted. If none of the named substrings are set, the
1377 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1378 empty string.
1379
1380 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1381 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1382 because the ] is interpreted as the first data character and the
1383 terminating ] is not found. PCRE has been made compatible with Perl in this
1384 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1385 cause memory overwriting.
1386
1387 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1388 string has been matched (to stop an infinite loop). It was not recognizing
1389 a conditional subpattern that could match an empty string if that
1390 subpattern was within another subpattern. For example, it looped when
1391 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1392 condition was not nested. This bug has been fixed.
1393
1394 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1395 past the start of the subject in the presence of bytes with the top bit
1396 set, for example "\x8aBCD".
1397
1398 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1399 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1400
1401 14. Optimized (?!) to (*FAIL).
1402
1403 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1404 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1405 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1406 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1407 does: it's just the validity check that is more restrictive.
1408
1409 16. Inserted checks for integer overflows during escape sequence (backslash)
1410 processing, and also fixed erroneous offset values for syntax errors during
1411 backslash processing.
1412
1413 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1414 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1415
1416 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1417 caused an overrun.
1418
1419 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1420 something other than just ASCII characters) inside a group that had an
1421 unlimited repeat caused a loop at compile time (while checking to see
1422 whether the group could match an empty string).
1423
1424 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1425 [\P{Any}] did so. (Error in the code for printing property names.)
1426
1427 21. An orphan \E inside a character class could cause a crash.
1428
1429 22. A repeated capturing bracket such as (A)? could cause a wild memory
1430 reference during compilation.
1431
1432 23. There are several functions in pcre_compile() that scan along a compiled
1433 expression for various reasons (e.g. to see if it's fixed length for look
1434 behind). There were bugs in these functions when a repeated \p or \P was
1435 present in the pattern. These operators have additional parameters compared
1436 with \d, etc, and these were not being taken into account when moving along
1437 the compiled data. Specifically:
1438
1439 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1440 length.
1441
1442 (b) An item such as \pL+ within a repeated group could cause crashes or
1443 loops.
1444
1445 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1446 "reference to non-existent subpattern" error.
1447
1448 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1449
1450 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1451 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1452
1453 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1454
1455 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1456 character were causing crashes (broken optimization).
1457
1458 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1459 \p or \P) caused a compile-time loop.
1460
1461 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1462 break. For example, the unstudied pattern [\r\n]A does not match the string
1463 "\r\nA" because change 7.0/46 below moves the current point on by two
1464 characters after failing to match at the start. However, the pattern \nA
1465 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1466 the same is true. There doesn't seem any very clean way out of this, but
1467 what I have chosen to do makes the common cases work: PCRE now takes note
1468 of whether there can be an explicit match for \r or \n anywhere in the
1469 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1470 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1471 pattern has explicit CR or LF references.
1472
1473 29. Added (*CR) etc for changing newline setting at start of pattern.
1474
1475
1476 Version 7.2 19-Jun-07
1477 ---------------------
1478
1479 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1480 which is apparently normally available under Windows.
1481
1482 2. Re-jig the pcregrep tests with different newline settings in an attempt
1483 to make them independent of the local environment's newline setting.
1484
1485 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1486
1487 4. Some of the "internals" tests were previously cut out when the link size
1488 was not 2, because the output contained actual offsets. The recent new
1489 "Z" feature of pcretest means that these can be cut out, making the tests
1490 usable with all link sizes.
1491
1492 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1493 stack recursion. This gives a massive performance boost under BSD, but just
1494 a small improvement under Linux. However, it saves one field in the frame
1495 in all cases.
1496
1497 6. Added more features from the forthcoming Perl 5.10:
1498
1499 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1500 recursion call. It refers to the nth most recently opened parentheses.
1501
1502 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1503 to be opened parentheses.
1504
1505 (c) Conditions that refer to capturing parentheses can be specified
1506 relatively, for example, (?(-2)... or (?(+3)...
1507
1508 (d) \K resets the start of the current match so that everything before
1509 is not part of it.
1510
1511 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1512
1513 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1514 reference syntax.
1515
1516 (g) (?| introduces a group in which the numbering of parentheses in each
1517 alternative starts with the same number.
1518
1519 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1520
1521 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1522 PCRE_INFO_JCHANGED.
1523
1524 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1525 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1526 for detecting groups that can match an empty string.
1527
1528 9. A pattern with a very large number of alternatives (more than several
1529 hundred) was running out of internal workspace during the pre-compile
1530 phase, where pcre_compile() figures out how much memory will be needed. A
1531 bit of new cunning has reduced the workspace needed for groups with
1532 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1533 workspace instead of running out of the 4096 that are available.
1534
1535 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1536
1537 11. Applied patch from Google to remove an optimization that didn't quite work.
1538 The report of the bug said:
1539
1540 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1541 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1542 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1543
1544 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1545 it matched the wrong number of bytes.
1546
1547
1548 Version 7.1 24-Apr-07
1549 ---------------------
1550
1551 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1552 that is more "standard", making use of automake and other Autotools. There
1553 is some re-arrangement of the files and adjustment of comments consequent
1554 on this.
1555
1556 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1557 for recursive directory scanning broke on some systems because the files
1558 are not scanned in any specific order and on different systems the order
1559 was different. A call to "sort" has been inserted into RunGrepTest for the
1560 approprate test as a short-term fix. In the longer term there may be an
1561 alternative.
1562
1563 3. I had an email from Eric Raymond about problems translating some of PCRE's
1564 man pages to HTML (despite the fact that I distribute HTML pages, some
1565 people do their own conversions for various reasons). The problems
1566 concerned the use of low-level troff macros .br and .in. I have therefore
1567 removed all such uses from the man pages (some were redundant, some could
1568 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1569 HTML has been updated to handle .nf/.fi and to complain if it encounters
1570 .br or .in.
1571
1572 4. Updated comments in configure.ac that get placed in config.h.in and also
1573 arranged for config.h to be included in the distribution, with the name
1574 config.h.generic, for the benefit of those who have to compile without
1575 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1576
1577 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1578 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1579 makevp.bat for the current PCRE, using the additional files
1580 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1581
1582 6. A Windows user reported a minor discrepancy with test 2, which turned out
1583 to be caused by a trailing space on an input line that had got lost in his
1584 copy. The trailing space was an accident, so I've just removed it.
1585
1586 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1587 that is needed.
1588
1589 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1590 as "const" (a) because they are and (b) because it helps the PHP
1591 maintainers who have recently made a script to detect big data structures
1592 in the php code that should be moved to the .rodata section. I remembered
1593 to update Builducptable as well, so it won't revert if ucptable.h is ever
1594 re-created.
1595
1596 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1597 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1598 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1599 support is not required. This saves 1.5-2K of code, which is important in
1600 some applications.
1601
1602 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1603 so as not to refer to the tables, even though these functions will never be
1604 called when UTF-8 support is disabled. Otherwise there are problems with a
1605 shared library.
1606
1607 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1608
1609 (a) It was defining its arguments as char * instead of void *.
1610
1611 (b) It was assuming that all moves were upwards in memory; this was true
1612 a long time ago when I wrote it, but is no longer the case.
1613
1614 The emulated memove() is provided for those environments that have neither
1615 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1616 is clearly not the case, as these two bugs were recently reported.
1617
1618 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1619 and Detrail to create the HTML documentation, the .txt form of the man
1620 pages, and it removes trailing spaces from listed files. It also creates
1621 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1622 case, it wraps all the #defines with #ifndefs. This script should be run
1623 before "make dist".
1624
1625 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1626 with Unicode property support.
1627
1628 (a) For a maximizing quantifier, if the two different cases of the
1629 character were of different lengths in their UTF-8 codings (there are
1630 some cases like this - I found 11), and the matching function had to
1631 back up over a mixture of the two cases, it incorrectly assumed they
1632 were both the same length.
1633
1634 (b) When PCRE was configured to use the heap rather than the stack for
1635 recursion during matching, it was not correctly preserving the data for
1636 the other case of a UTF-8 character when checking ahead for a match
1637 while processing a minimizing repeat. If the check also involved
1638 matching a wide character, but failed, corruption could cause an
1639 erroneous result when trying to check for a repeat of the original
1640 character.
1641
1642 13. Some tidying changes to the testing mechanism:
1643
1644 (a) The RunTest script now detects the internal link size and whether there
1645 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1646 values substituted by "configure". (The RunGrepTest script already did
1647 this for UTF-8.) The configure.ac script no longer substitutes the
1648 relevant variables.
1649
1650 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1651 with length and offset values. This means that the output is different
1652 for different internal link sizes. Test 2 is skipped for link sizes
1653 other than 2 because of this, bypassing the problem. Unfortunately,
1654 there was also a test in test 3 (the locale tests) that used /B and
1655 failed for link sizes other than 2. Rather than cut the whole test out,
1656 I have added a new /Z option to pcretest that replaces the length and
1657 offset values with spaces. This is now used to make test 3 independent
1658 of link size. (Test 2 will be tidied up later.)
1659
1660 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1661 segmentation fault instead of returning the appropriate error message.
1662
1663 15. In multiline mode when the newline sequence was set to "any", the pattern
1664 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1665 This doesn't seem right; it now treats the CRLF combination as the line
1666 ending, and so does not match in that case. It's only a pattern such as ^$
1667 that would hit this one: something like ^ABC$ would have failed after \r
1668 and then tried again after \r\n.
1669
1670 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1671 in an attempt to make files that differ only in their line terminators
1672 compare equal. This works on Linux.
1673
1674 17. Under certain error circumstances pcregrep might try to free random memory
1675 as it exited. This is now fixed, thanks to valgrind.
1676
1677 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1678 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1679 was because its rules for how to advance for /g after matching an empty
1680 string at the end of a line did not allow for this case. They now check for
1681 it specially.
1682
1683 20. pcretest is supposed to handle patterns and data of any length, by
1684 extending its buffers when necessary. It was getting this wrong when the
1685 buffer for a data line had to be extended.
1686
1687 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1688 CRLF as a newline sequence.
1689
1690 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1691 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1692 I have nevertheless tidied it up.
1693
1694 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1695
1696 24. Added a man page for pcre-config.
1697
1698
1699 Version 7.0 19-Dec-06
1700 ---------------------
1701
1702 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1703 moving to gcc 4.1.1.
1704
1705 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1706 sys/time.h, which is documented as needed for this function. It doesn't
1707 seem to matter on Linux, but it showed up on some releases of OS X.
1708
1709 3. It seems that there are systems where bytes whose values are greater than
1710 127 match isprint() in the "C" locale. The "C" locale should be the
1711 default when a C program starts up. In most systems, only ASCII printing
1712 characters match isprint(). This difference caused the output from pcretest
1713 to vary, making some of the tests fail. I have changed pcretest so that:
1714
1715 (a) When it is outputting text in the compiled version of a pattern, bytes
1716 other than 32-126 are always shown as hex escapes.
1717
1718 (b) When it is outputting text that is a matched part of a subject string,
1719 it does the same, unless a different locale has been set for the match
1720 (using the /L modifier). In this case, it uses isprint() to decide.
1721
1722 4. Fixed a major bug that caused incorrect computation of the amount of memory
1723 required for a compiled pattern when options that changed within the
1724 pattern affected the logic of the preliminary scan that determines the
1725 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1726 that the computed length was too small. The symptoms of this bug were
1727 either the PCRE error "internal error: code overflow" from pcre_compile(),
1728 or a glibc crash with a message such as "pcretest: free(): invalid next
1729 size (fast)". Examples of patterns that provoked this bug (shown in
1730 pcretest format) are:
1731
1732 /(?-x: )/x
1733 /(?x)(?-x: \s*#\s*)/
1734 /((?i)[\x{c0}])/8
1735 /(?i:[\x{c0}])/8
1736
1737 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1738 is now done differently.
1739
1740 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1741 wrapper classes; (b) implement a new function in the C++ scanner that is
1742 more efficient than the old way of doing things because it avoids levels of
1743 recursion in the regex matching; (c) add a paragraph to the documentation
1744 for the FullMatch() function.
1745
1746 6. The escape sequence \n was being treated as whatever was defined as
1747 "newline". Not only was this contrary to the documentation, which states
1748 that \n is character 10 (hex 0A), but it also went horribly wrong when
1749 "newline" was defined as CRLF. This has been fixed.
1750
1751 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1752 was being set to -1 for the "end of line" case (supposedly a value that no
1753 character can have). Though this value is never used (the check for end of
1754 line is "zero bytes in current character"), it caused compiler complaints.
1755 I've changed it to 0xffffffff.
1756
1757 8. In pcre_version.c, the version string was being built by a sequence of
1758 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1759 string (as it is for production releases) called a macro with an empty
1760 argument. The C standard says the result of this is undefined. The gcc
1761 compiler treats it as an empty string (which was what was wanted) but it is
1762 reported that Visual C gives an error. The source has been hacked around to
1763 avoid this problem.
1764
1765 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1766 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1767 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1768 of them did).
1769
1770 10. Originally, pcretest opened its input and output without "b"; then I was
1771 told that "b" was needed in some environments, so it was added for release
1772 5.0 to both the input and output. (It makes no difference on Unix-like
1773 systems.) Later I was told that it is wrong for the input on Windows. I've
1774 now abstracted the modes into two macros, to make it easier to fiddle with
1775 them, and removed "b" from the input mode under Windows.
1776
1777 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1778
1779 12. Added -help and --help to pcretest as an official way of being reminded
1780 of the options.
1781
1782 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1783 and pcrecpp.cc because they annoy compilers at high warning levels.
1784
1785 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1786
1787 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1788 scripts are not C programs :-) and which was not noticed because it works
1789 on Linux.
1790
1791 16. pcretest is supposed to handle any length of pattern and data line (as one
1792 line or as a continued sequence of lines) by extending its input buffer if
1793 necessary. This feature was broken for very long pattern lines, leading to
1794 a string of junk being passed to pcre_compile() if the pattern was longer
1795 than about 50K.
1796
1797 17. I have done a major re-factoring of the way pcre_compile() computes the
1798 amount of memory needed for a compiled pattern. Previously, there was code
1799 that made a preliminary scan of the pattern in order to do this. That was
1800 OK when PCRE was new, but as the facilities have expanded, it has become
1801 harder and harder to keep it in step with the real compile phase, and there
1802 have been a number of bugs (see for example, 4 above). I have now found a
1803 cunning way of running the real compile function in a "fake" mode that
1804 enables it to compute how much memory it would need, while actually only
1805 ever using a few hundred bytes of working memory and without too many
1806 tests of the mode. This should make future maintenance and development
1807 easier. A side effect of this work is that the limit of 200 on the nesting
1808 depth of parentheses has been removed (though this was never a serious
1809 limitation, I suspect). However, there is a downside: pcre_compile() now
1810 runs more slowly than before (30% or more, depending on the pattern). I
1811 hope this isn't a big issue. There is no effect on runtime performance.
1812
1813 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1814 newline (only possible for the last line of a file) and it was a
1815 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1816
1817 19. Added additional timing features to pcretest. (1) The -tm option now times
1818 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1819 separate command line item, by a number that specifies the number of
1820 repeats to use when timing. The default is 50000; this gives better
1821 precision, but takes uncomfortably long for very large patterns.
1822
1823 20. Extended pcre_study() to be more clever in cases where a branch of a
1824 subpattern has no definite first character. For example, (a*|b*)[cd] would
1825 previously give no result from pcre_study(). Now it recognizes that the
1826 first character must be a, b, c, or d.
1827
1828 21. There was an incorrect error "recursive call could loop indefinitely" if
1829 a subpattern (or the entire pattern) that was being tested for matching an
1830 empty string contained only one non-empty item after a nested subpattern.
1831 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1832 incorrectly, because the \d was being skipped in the check.
1833
1834 22. The pcretest program now has a new pattern option /B and a command line
1835 option -b, which is equivalent to adding /B to every pattern. This causes
1836 it to show the compiled bytecode, without the additional information that
1837 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1838 is the same as /B/I).
1839
1840 23. A new optimization is now able automatically to treat some sequences such
1841 as a*b as a*+b. More specifically, if something simple (such as a character
1842 or a simple class like \d) has an unlimited quantifier, and is followed by
1843 something that cannot possibly match the quantified thing, the quantifier
1844 is automatically "possessified".
1845
1846 24. A recursive reference to a subpattern whose number was greater than 39
1847 went wrong under certain circumstances in UTF-8 mode. This bug could also
1848 have affected the operation of pcre_study().
1849
1850 25. Realized that a little bit of performance could be had by replacing
1851 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1852
1853 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1854
1855 27. Possessive quantifiers such as a++ were previously implemented by turning
1856 them into atomic groups such as ($>a+). Now they have their own opcodes,
1857 which improves performance. This includes the automatically created ones
1858 from 23 above.
1859
1860 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1861 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1862 the first matched character to be a colon. This applied both to named and
1863 numbered groups.
1864
1865 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1866
1867 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1868 building PCRE on Windows easier, so I have included it in the distribution.
1869
1870 31. There is now a check in pcretest against a ridiculously large number being
1871 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1872 loop, the loop is abandoned.
1873
1874 32. Forward references to subpatterns in conditions such as (?(2)...) where
1875 subpattern 2 is defined later cause pcre_compile() to search forwards in
1876 the pattern for the relevant set of parentheses. This search went wrong
1877 when there were unescaped parentheses in a character class, parentheses
1878 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1879
1880 33. "Subroutine" calls and backreferences were previously restricted to
1881 referencing subpatterns earlier in the regex. This restriction has now
1882 been removed.
1883
1884 34. Added a number of extra features that are going to be in Perl 5.10. On the
1885 whole, these are just syntactic alternatives for features that PCRE had
1886 previously implemented using the Python syntax or my own invention. The
1887 other formats are all retained for compatibility.
1888
1889 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1890 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1891 also .NET compatible.
1892
1893 (b) A recursion or subroutine call to a named group can now be defined as
1894 (?&name) as well as (?P>name).
1895
1896 (c) A backreference to a named group can now be defined as \k<name> or
1897 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1898 5.10, are also .NET compatible.
1899
1900 (d) A conditional reference to a named group can now use the syntax
1901 (?(<name>) or (?('name') as well as (?(name).
1902
1903 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1904 groups (named and numbered) that are never evaluated inline, but can be
1905 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1906 is always false. There may be only one alternative in such a group.
1907
1908 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1909 as the simple (?(R). The condition is true only if the most recent
1910 recursion is that of the given number or name. It does not search out
1911 through the entire recursion stack.
1912
1913 (g) The escape \gN or \g{N} has been added, where N is a positive or
1914 negative number, specifying an absolute or relative reference.
1915
1916 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1917 some "unreachable code" warnings.
1918
1919 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1920 things, this adds five new scripts.
1921
1922 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1923 There were also incompatibilities regarding the handling of \Q..\E inside
1924 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1925 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1926
1927 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1928 matches an empty string, and forcibly breaks the loop. There were bugs in
1929 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1930 against aaaa the result was just "a" rather than "aaaa", for example. Two
1931 separate and independent bugs (that affected different cases) have been
1932 fixed.
1933
1934 39. Refactored the code to abolish the use of different opcodes for small
1935 capturing bracket numbers. This is a tidy that I avoided doing when I
1936 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1937 The new approach is not only tidier, it makes it possible to reduce the
1938 memory needed to fix the previous bug (38).
1939
1940 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1941 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1942 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1943 mode.
1944
1945 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1946 report.
1947
1948 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1949 copy construction and assignment in the C++ wrapper.
1950
1951 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1952 couple of bugs that could have given wrong results in the "--newline=crlf"
1953 case.
1954
1955 44. Added a number of casts and did some reorganization of signed/unsigned int
1956 variables following suggestions from Dair Grant. Also renamed the variable
1957 "this" as "item" because it is a C++ keyword.
1958
1959 45. Arranged for dftables to add
1960
1961 #include "pcre_internal.h"
1962
1963 to pcre_chartables.c because without it, gcc 4.x may remove the array
1964 definition from the final binary if PCRE is built into a static library and
1965 dead code stripping is activated.
1966
1967 46. For an unanchored pattern, if a match attempt fails at the start of a
1968 newline sequence, and the newline setting is CRLF or ANY, and the next two
1969 characters are CRLF, advance by two characters instead of one.
1970
1971
1972 Version 6.7 04-Jul-06
1973 ---------------------
1974
1975 1. In order to handle tests when input lines are enormously long, pcretest has
1976 been re-factored so that it automatically extends its buffers when
1977 necessary. The code is crude, but this _is_ just a test program. The
1978 default size has been increased from 32K to 50K.
1979
1980 2. The code in pcre_study() was using the value of the re argument before
1981 testing it for NULL. (Of course, in any sensible call of the function, it
1982 won't be NULL.)
1983
1984 3. The memmove() emulation function in pcre_internal.h, which is used on
1985 systems that lack both memmove() and bcopy() - that is, hardly ever -
1986 was missing a "static" storage class specifier.
1987
1988 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1989 containing an extended class (one that cannot be represented by a bitmap
1990 because it contains high-valued characters or Unicode property items, e.g.
1991 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1992 pattern, but PCRE should not loop if you do not (it no longer does).
1993 [Detail: two cases were found: (a) a repeated subpattern containing an
1994 extended class; (b) a recursive reference to a subpattern that followed a
1995 previous extended class. It wasn't skipping over the extended class
1996 correctly when UTF-8 mode was not set.]
1997
1998 5. A negated single-character class was not being recognized as fixed-length
1999 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2000 compile error "lookbehind assertion is not fixed length".
2001
2002 6. The RunPerlTest auxiliary script was showing an unexpected difference
2003 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2004 write a Perl script that can interpret lines of an input file either as
2005 byte characters or as UTF-8, which is what "perltest" was being required to
2006 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2007 can't do is switch easily at run time between having the "use utf8;" pragma
2008 or not. In the end, I fudged it by using the RunPerlTest script to insert
2009 "use utf8;" explicitly for the UTF-8 tests.
2010
2011 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2012 the end of the subject string, contrary to the documentation and to what
2013 Perl does. This was true of both matching functions. Now it matches only at
2014 the start of the subject and immediately after *internal* newlines.
2015
2016 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2017 a pointer to an int instead of a pointer to an unsigned long int. This
2018 caused problems on 64-bit systems.
2019
2020 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2021 instance of the 'standard' template library not being so standard".
2022
2023 10. There was no check on the number of named subpatterns nor the maximum
2024 length of a subpattern name. The product of these values is used to compute
2025 the size of the memory block for a compiled pattern. By supplying a very
2026 long subpattern name and a large number of named subpatterns, the size
2027 computation could be caused to overflow. This is now prevented by limiting
2028 the length of names to 32 characters, and the number of named subpatterns
2029 to 10,000.
2030
2031 11. Subpatterns that are repeated with specific counts have to be replicated in
2032 the compiled pattern. The size of memory for this was computed from the
2033 length of the subpattern and the repeat count. The latter is limited to
2034 65535, but there was no limit on the former, meaning that integer overflow
2035 could in principle occur. The compiled length of a repeated subpattern is
2036 now limited to 30,000 bytes in order to prevent this.
2037
2038 12. Added the optional facility to have named substrings with the same name.
2039
2040 13. Added the ability to use a named substring as a condition, using the
2041 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2042 are numbers (not recommended). Forward references are permitted.
2043
2044 14. Added forward references in named backreferences (if you see what I mean).
2045
2046 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2047 pattern could run off the end of the subject. For example, the pattern
2048 "(?s)(.{1,5})"8 did this with the subject "ab".
2049
2050 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2051 PCRE_CASELESS was set when matching characters that were quantified with ?
2052 or *.
2053
2054 17. A character class other than a single negated character that had a minimum
2055 but no maximum quantifier - for example [ab]{6,} - was not handled
2056 correctly by pce_dfa_exec(). It would match only one character.
2057
2058 18. A valid (though odd) pattern that looked like a POSIX character
2059 class but used an invalid character after [ (for example [[,abc,]]) caused
2060 pcre_compile() to give the error "Failed: internal error: code overflow" or
2061 in some cases to crash with a glibc free() error. This could even happen if
2062 the pattern terminated after [[ but there just happened to be a sequence of
2063 letters, a binary zero, and a closing ] in the memory that followed.
2064
2065 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2066 over the years. Originally (before any Unicode support), just the bottom 8
2067 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2068 output from "man perlunicode" includes this:
2069
2070 The regular expression compiler produces polymorphic opcodes. That
2071 is, the pattern adapts to the data and automatically switches to
2072 the Unicode character scheme when presented with Unicode data--or
2073 instead uses a traditional byte scheme when presented with byte
2074 data.
2075
2076 Sadly, a wide octal escape does not cause a switch, and in a string with
2077 no other multibyte characters, these octal escapes are treated as before.
2078 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2079 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2080 Unicode string.
2081
2082 I have not perpetrated such confusion in PCRE. Up till now, it took just
2083 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2084 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2085 translate to the appropriate multibyte character.
2086
2087 29. Applied some refactoring to reduce the number of warnings from Microsoft
2088 and Borland compilers. This has included removing the fudge introduced
2089 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2090 a warning about an unused variable.
2091
2092 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2093 characters since release 4.0, because Perl (from release 5.004) does not.
2094 [Or at least, is documented not to: some releases seem to be in conflict
2095 with the documentation.] However, when a pattern was studied with
2096 pcre_study() and all its branches started with \s, PCRE still included VT
2097 as a possible starting character. Of course, this did no harm; it just
2098 caused an unnecessary match attempt.
2099
2100 22. Removed a now-redundant internal flag bit that recorded the fact that case
2101 dependency changed within the pattern. This was once needed for "required
2102 byte" processing, but is no longer used. This recovers a now-scarce options
2103 bit. Also moved the least significant internal flag bit to the most-
2104 significant bit of the word, which was not previously used (hangover from
2105 the days when it was an int rather than a uint) to free up another bit for
2106 the future.
2107
2108 23. Added support for CRLF line endings as well as CR and LF. As well as the
2109 default being selectable at build time, it can now be changed at runtime
2110 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2111 specify that it is scanning data with non-default line endings.
2112
2113 24. Changed the definition of CXXLINK to make it agree with the definition of
2114 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2115
2116 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2117 recursions. This makes a big different to stack usage for some patterns.
2118
2119 26. If a subpattern containing a named recursion or subroutine reference such
2120 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2121 the space required for the compiled pattern went wrong and gave too small a
2122 value. Depending on the environment, this could lead to "Failed: internal
2123 error: code overflow at offset 49" or "glibc detected double free or
2124 corruption" errors.
2125
2126 27. Applied patches from Google (a) to support the new newline modes and (b) to
2127 advance over multibyte UTF-8 characters in GlobalReplace.
2128
2129 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2130 difference for some implementation of PCRE in some Windows version.
2131
2132 29. Added some extra testing facilities to pcretest:
2133
2134 \q<number> in a data line sets the "match limit" value
2135 \Q<number> in a data line sets the "match recursion limt" value
2136 -S <number> sets the stack size, where <number> is in megabytes
2137
2138 The -S option isn't available for Windows.
2139
2140
2141 Version 6.6 06-Feb-06
2142 ---------------------
2143
2144 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2145 in pcreposix.h. I have copied the definition from pcre.h.
2146
2147 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2148 because pcre.h is no longer a built file.
2149
2150 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2151 not normally included in the compiled code.
2152
2153
2154 Version 6.5 01-Feb-06
2155 ---------------------
2156
2157 1. When using the partial match feature with pcre_dfa_exec(), it was not
2158 anchoring the second and subsequent partial matches at the new starting
2159 point. This could lead to incorrect results. For example, with the pattern
2160 /1234/, partially matching against "123" and then "a4" gave a match.
2161
2162 2. Changes to pcregrep:
2163
2164 (a) All non-match returns from pcre_exec() were being treated as failures
2165 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2166 error message is output. Some extra information is given for the
2167 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2168 probably the only errors that are likely to be caused by users (by
2169 specifying a regex that has nested indefinite repeats, for instance).
2170 If there are more than 20 of these errors, pcregrep is abandoned.
2171
2172 (b) A binary zero was treated as data while matching, but terminated the
2173 output line if it was written out. This has been fixed: binary zeroes
2174 are now no different to any other data bytes.
2175
2176 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2177 used to set a locale for matching. The --locale=xxxx long option has
2178 been added (no short equivalent) to specify a locale explicitly on the
2179 pcregrep command, overriding the environment variables.
2180
2181 (d) When -B was used with -n, some line numbers in the output were one less
2182 than they should have been.
2183
2184 (e) Added the -o (--only-matching) option.
2185
2186 (f) If -A or -C was used with -c (count only), some lines of context were
2187 accidentally printed for the final match.
2188
2189 (g) Added the -H (--with-filename) option.
2190
2191 (h) The combination of options -rh failed to suppress file names for files
2192 that were found from directory arguments.
2193
2194 (i) Added the -D (--devices) and -d (--directories) options.
2195
2196 (j) Added the -F (--fixed-strings) option.
2197
2198 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2199
2200 (l) Added the --colo(u)r option.
2201
2202 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2203 is not present by default.
2204
2205 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2206 items such as (?R) or (?1), when the recursion could match a number of
2207 alternatives. If it matched one of the alternatives, but subsequently,
2208 outside the recursion, there was a failure, the code tried to back up into
2209 the recursion. However, because of the way PCRE is implemented, this is not
2210 possible, and the result was an incorrect result from the match.
2211
2212 In order to prevent this happening, the specification of recursion has
2213 been changed so that all such subpatterns are automatically treated as
2214 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2215
2216 4. I had overlooked the fact that, in some locales, there are characters for
2217 which isalpha() is true but neither isupper() nor islower() are true. In
2218 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2219 and ordfeminine) are like this. This affected the treatment of \w and \W
2220 when they appeared in character classes, but not when they appeared outside
2221 a character class. The bit map for "word" characters is now created
2222 separately from the results of isalnum() instead of just taking it from the
2223 upper, lower, and digit maps. (Plus the underscore character, of course.)
2224
2225 5. The above bug also affected the handling of POSIX character classes such as
2226 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2227 permanent tables. Instead, the bit maps for such a class were previously
2228 created as the appropriate unions of the upper, lower, and digit bitmaps.
2229 Now they are created by subtraction from the [[:word:]] class, which has
2230 its own bitmap.
2231
2232 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2233 It is created by subtracting the vertical space characters (\x09, \x0a,
2234 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2235 subtraction was done in the overall bitmap for a character class, meaning
2236 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2237 be recognized. This bug has been fixed.
2238
2239 7. Patches from the folks at Google:
2240
2241 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2242 real life, but is still worth protecting against".
2243
2244 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2245 regular expressions".
2246
2247 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2248 have it.
2249
2250 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2251 "configure" and the latter not, in order to fix a problem somebody had
2252 with compiling the Arg class on HP-UX.
2253
2254 (e) Improve the error-handling of the C++ wrapper a little bit.
2255
2256 (f) New tests for checking recursion limiting.
2257
2258 8. The pcre_memmove() function, which is used only if the environment does not
2259 have a standard memmove() function (and is therefore rarely compiled),
2260 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2261 returning a result (though PCRE never actually uses the result).
2262
2263 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2264 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2265 returned instead of calling malloc() with an overflowing number that would
2266 most likely cause subsequent chaos.
2267
2268 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2269
2270 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2271 with this option is matched, the nmatch and pmatch options of regexec() are
2272 ignored.
2273
2274 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2275 provided in case anyone wants to the the POSIX interface with UTF-8
2276 strings.
2277
2278 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2279 C++ linking (needed for some HP-UX environments).
2280
2281 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2282 (unused parameter) and in the pcre_printint() function (omitted "default"
2283 switch label when the default is to do nothing).
2284
2285 15. Added some code to make it possible, when PCRE is compiled as a C++
2286 library, to replace subject pointers for pcre_exec() with a smart pointer
2287 class, thus making it possible to process discontinuous strings.
2288
2289 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2290 much the same function. They were added by different people who were trying
2291 to make PCRE easy to compile on non-Unix systems. It has been suggested
2292 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2293 for compiling on Windows systems. I have therefore replaced it with
2294 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2295 defaults to "extern" for C or "extern C" for C++, which works fine on
2296 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2297 SCOPE with something explicit in config.h. In addition:
2298
2299 (a) pcreposix.h still had just "extern" instead of either of these macros;
2300 I have replaced it with PCRE_DATA_SCOPE.
2301
2302 (b) Functions such as _pcre_xclass(), which are internal to the library,
2303 but external in the C sense, all had PCRE_EXPORT in their definitions.
2304 This is apparently wrong for the Windows case, so I have removed it.
2305 (It makes no difference on Unix-like systems.)
2306
2307 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2308 of recursive calls to match(). This is different to MATCH_LIMIT because
2309 that limits the total number of calls to match(), not all of which increase
2310 the depth of recursion. Limiting the recursion depth limits the amount of
2311 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2312 when PCRE is compiled, and changed at run time. A patch from Google adds
2313 this functionality to the C++ interface.
2314
2315 18. Changes to the handling of Unicode character properties:
2316
2317 (a) Updated the table to Unicode 4.1.0.
2318
2319 (b) Recognize characters that are not in the table as "Cn" (undefined).
2320
2321 (c) I revised the way the table is implemented to a much improved format
2322 which includes recognition of ranges. It now supports the ranges that
2323 are defined in UnicodeData.txt, and it also amalgamates other
2324 characters into ranges. This has reduced the number of entries in the
2325 table from around 16,000 to around 3,000, thus reducing its size
2326 considerably. I realized I did not need to use a tree structure after
2327 all - a binary chop search is just as efficient. Having reduced the
2328 number of entries, I extended their size from 6 bytes to 8 bytes to
2329 allow for more data.
2330
2331 (d) Added support for Unicode script names via properties such as \p{Han}.
2332
2333 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2334 matching that character.
2335
2336 20. When matching a repeated Unicode property with a minimum greater than zero,
2337 (for example \pL{2,}), PCRE could look past the end of the subject if it
2338 reached it while seeking the minimum number of characters. This could
2339 happen only if some of the characters were more than one byte long, because
2340 there is a check for at least the minimum number of bytes.
2341
2342 21. Refactored the implementation of \p and \P so as to be more general, to
2343 allow for more different types of property in future. This has changed the
2344 compiled form incompatibly. Anybody with saved compiled patterns that use
2345 \p or \P will have to recompile them.
2346
2347 22. Added "Any" and "L&" to the supported property types.
2348
2349 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2350 but give a compile time error if the value is greater than 0xff.
2351
2352 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2353 accidentally not being installed or uninstalled.
2354
2355 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2356 made were to insert the current release number. This seemed silly, because
2357 it made things harder for people building PCRE on systems that don't run
2358 "configure". I have turned pcre.h into a distributed file, no longer built
2359 by "configure", with the version identification directly included. There is
2360 no longer a pcre.h.in file.
2361
2362 However, this change necessitated a change to the pcre-config script as
2363 well. It is built from pcre-config.in, and one of the substitutions was the
2364 release number. I have updated configure.ac so that ./configure now finds
2365 the release number by grepping pcre.h.
2366
2367 26. Added the ability to run the tests under valgrind.
2368
2369
2370 Version 6.4 05-Sep-05
2371 ---------------------
2372
2373 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2374 "--" to be printed when multiple files were scanned, even when none of the
2375 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2376 consider it to be a bug, and have restored the previous behaviour.
2377
2378 2. A couple of code tidies to get rid of compiler warnings.
2379
2380 3. The pcretest program used to cheat by referring to symbols in the library
2381 whose names begin with _pcre_. These are internal symbols that are not
2382 really supposed to be visible externally, and in some environments it is
2383 possible to suppress them. The cheating is now confined to including
2384 certain files from the library's source, which is a bit cleaner.
2385
2386 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2387 file's purpose clearer.
2388
2389 5. Reorganized pcre_ucp_findchar().
2390
2391
2392 Version 6.3 15-Aug-05
2393 ---------------------
2394
2395 1. The file libpcre.pc.in did not have general read permission in the tarball.
2396
2397 2. There were some problems when building without C++ support:
2398
2399 (a) If C++ support was not built, "make install" and "make test" still
2400 tried to test it.
2401
2402 (b) There were problems when the value of CXX was explicitly set. Some
2403 changes have been made to try to fix these, and ...
2404
2405 (c) --disable-cpp can now be used to explicitly disable C++ support.
2406
2407 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2408 backslash in a target when C++ was disabled. This confuses some
2409 versions of "make", apparently. Using an intermediate variable solves
2410 this. (Same for CPP_LOBJ.)
2411
2412 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2413 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2414 necessary on certain architectures.
2415
2416 4. Added a setting of -export-symbols-regex to the link command to remove
2417 those symbols that are exported in the C sense, but actually are local
2418 within the library, and not documented. Their names all begin with
2419 "_pcre_". This is not a perfect job, because (a) we have to except some
2420 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2421 available (and never for static libraries). I have made a note to try to
2422 find a way round (a) in the future.
2423
2424
2425 Version 6.2 01-Aug-05
2426 ---------------------
2427
2428 1. There was no test for integer overflow of quantifier values. A construction
2429 such as {1111111111111111} would give undefined results. What is worse, if
2430 a minimum quantifier for a parenthesized subpattern overflowed and became
2431 negative, the calculation of the memory size went wrong. This could have
2432 led to memory overwriting.
2433
2434 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2435
2436 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2437 operating environments where this matters.
2438
2439 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2440 PCRE options from within the C++ wrapper.
2441
2442 5. Named capturing subpatterns were not being correctly counted when a pattern
2443 was compiled. This caused two problems: (a) If there were more than 100
2444 such subpatterns, the calculation of the memory needed for the whole
2445 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2446 back references of the form \12, where the number was greater than 9, were
2447 not recognized as back references, even though there were sufficient
2448 previous subpatterns.
2449
2450 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2451 versions of gcc, e.g. 2.95.4.
2452
2453
2454 Version 6.1 21-Jun-05
2455 ---------------------
2456
2457 1. There was one reference to the variable "posix" in pcretest.c that was not
2458 surrounded by "#if !defined NOPOSIX".
2459
2460 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2461 the cross-check on the old pcre_info() function, for the benefit of the
2462 cut-down version of PCRE that is currently imported into Exim.
2463
2464 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2465 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2466 patterns that start (?i) but I don't think that matters. The use of (?i) is
2467 just an example; this all applies to the other options as well.
2468
2469 4. Since libtool seems to echo the compile commands it is issuing, the output
2470 from "make" can be reduced a bit by putting "@" in front of each libtool
2471 compile command.
2472
2473 5. Patch from the folks at Google for configure.in to be a bit more thorough
2474 in checking for a suitable C++ installation before trying to compile the
2475 C++ stuff. This should fix a reported problem when a compiler was present,
2476 but no suitable headers.
2477
2478 6. The man pages all had just "PCRE" as their title. I have changed them to
2479 be the relevant file name. I have also arranged that these names are
2480 retained in the file doc/pcre.txt, which is a concatenation in text format
2481 of all the man pages except the little individual ones for each function.
2482
2483 7. The NON-UNIX-USE file had not been updated for the different set of source
2484 files that come with release 6. I also added a few comments about the C++
2485 wrapper.
2486
2487
2488 Version 6.0 07-Jun-05
2489 ---------------------
2490
2491 1. Some minor internal re-organization to help with my DFA experiments.
2492
2493 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2494 didn't matter for the library itself when fully configured, but did matter
2495 when compiling without UCP support, or within Exim, where the ucp files are
2496 not imported.
2497
2498 3. Refactoring of the library code to split up the various functions into
2499 different source modules. The addition of the new DFA matching code (see
2500 below) to a single monolithic source would have made it really too
2501 unwieldy, quite apart from causing all the code to be include in a
2502 statically linked application, when only some functions are used. This is
2503 relevant even without the DFA addition now that patterns can be compiled in
2504 one application and matched in another.
2505
2506 The downside of splitting up is that there have to be some external
2507 functions and data tables that are used internally in different modules of
2508 the library but which are not part of the API. These have all had their
2509 names changed to start with "_pcre_" so that they are unlikely to clash
2510 with other external names.
2511
2512 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2513 a different (DFA) algorithm. Although it is slower than the original
2514 function, it does have some advantages for certain types of matching
2515 problem.
2516
2517 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2518 including restarting after a partial match.
2519
2520 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2521 defined when compiling for Windows was sent to me. I have put it into the
2522 code, though I have no means of testing or verifying it.
2523
2524 7. Added the pcre_refcount() auxiliary function.
2525
2526 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2527 match before or at the first newline in the subject string. In pcretest,
2528 the /f option on a pattern can be used to set this.
2529
2530 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2531 would behave wrongly. This has been present in PCRE since release 4.0.
2532
2533 10. A number of changes to the pcregrep command:
2534
2535 (a) Refactored how -x works; insert ^(...)$ instead of setting
2536 PCRE_ANCHORED and checking the length, in preparation for adding
2537 something similar for -w.
2538
2539 (b) Added the -w (match as a word) option.
2540
2541 (c) Refactored the way lines are read and buffered so as to have more
2542 than one at a time available.
2543
2544 (d) Implemented a pcregrep test script.
2545
2546 (e) Added the -M (multiline match) option. This allows patterns to match
2547 over several lines of the subject. The buffering ensures that at least
2548 8K, or the rest of the document (whichever is the shorter) is available
2549 for matching (and similarly the previous 8K for lookbehind assertions).
2550
2551 (f) Changed the --help output so that it now says
2552
2553 -w, --word-regex(p)
2554
2555 instead of two lines, one with "regex" and the other with "regexp"
2556 because that confused at least one person since the short forms are the
2557 same. (This required a bit of code, as the output is generated
2558 automatically from a table. It wasn't just a text change.)
2559
2560 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2561 option but starts with a hyphen. Could be a pattern or a path name
2562 starting with a hyphen, for instance.
2563
2564 (h) "-" can be given as a file name to represent stdin.
2565
2566 (i) When file names are being printed, "(standard input)" is used for
2567 the standard input, for compatibility with GNU grep. Previously
2568 "<stdin>" was used.
2569
2570 (j) The option --label=xxx can be used to supply a name to be used for
2571 stdin when file names are being printed. There is no short form.
2572
2573 (k) Re-factored the options decoding logic because we are going to add
2574 two more options that take data. Such options can now be given in four
2575 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2576
2577 (l) Added the -A, -B, and -C options for requesting that lines of context
2578 around matches be printed.
2579
2580 (m) Added the -L option to print the names of files that do not contain
2581 any matching lines, that is, the complement of -l.
2582
2583 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2584 continue to scan other files.
2585
2586 (o) The -s option was incorrectly implemented. For compatibility with other
2587 greps, it now suppresses the error message for a non-existent or non-
2588 accessible file (but not the return code). There is a new option called
2589 -q that suppresses the output of matching lines, which was what -s was
2590 previously doing.
2591
2592 (p) Added --include and --exclude options to specify files for inclusion
2593 and exclusion when recursing.
2594
2595 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2596 Hopefully, it now does.
2597
2598 12. Missing cast in pcre_study().
2599
2600 13. Added an "uninstall" target to the makefile.
2601
2602 14. Replaced "extern" in the function prototypes in Makefile.in with
2603 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2604 world, but is set differently for Windows.
2605
2606 15. Added a second compiling function called pcre_compile2(). The only
2607 difference is that it has an extra argument, which is a pointer to an
2608 integer error code. When there is a compile-time failure, this is set
2609 non-zero, in addition to the error test pointer being set to point to an
2610 error message. The new argument may be NULL if no error number is required
2611 (but then you may as well call pcre_compile(), which is now just a
2612 wrapper). This facility is provided because some applications need a
2613 numeric error indication, but it has also enabled me to tidy up the way
2614 compile-time errors are handled in the POSIX wrapper.
2615
2616 16. Added VPATH=.libs to the makefile; this should help when building with one
2617 prefix path and installing with another. (Or so I'm told by someone who
2618 knows more about this stuff than I do.)
2619
2620 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2621 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2622 match everything, including newlines. This is not POSIX-compatible, but
2623 somebody wanted the feature. From pcretest it can be activated by using
2624 both the P and the s flags.
2625
2626 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2627
2628 19. libpcre.pc was being incorrectly installed as executable.
2629
2630 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2631 it now also looks for '\r' so that it will work unmodified on Windows.
2632
2633 21. Added Google's contributed C++ wrapper to the distribution.
2634
2635 22. Added some untidy missing memory free() calls in pcretest, to keep
2636 Electric Fence happy when testing.
2637
2638
2639
2640 Version 5.0 13-Sep-04
2641 ---------------------
2642
2643 1. Internal change: literal characters are no longer packed up into items
2644 containing multiple characters in a single byte-string. Each character
2645 is now matched using a separate opcode. However, there may be more than one
2646 byte in the character in UTF-8 mode.
2647
2648 2. The pcre_callout_block structure has two new fields: pattern_position and
2649 next_item_length. These contain the offset in the pattern to the next match
2650 item, and its length, respectively.
2651
2652 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2653 insertion of callouts before each pattern item. Added the /C option to
2654 pcretest to make use of this.
2655
2656 4. On the advice of a Windows user, the lines
2657
2658 #if defined(_WIN32) || defined(WIN32)
2659 _setmode( _fileno( stdout ), 0x8000 );
2660 #endif /* defined(_WIN32) || defined(WIN32) */
2661
2662 have been added to the source of pcretest. This apparently does useful
2663 magic in relation to line terminators.
2664
2665 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2666 for the benefit of those environments where the "b" makes a difference.
2667
2668 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2669 to know about it. I have put a hack into configure.in that adds in code
2670 to set GCC=yes if CC=icc. This seems to end up at a point in the
2671 generated configure script that is early enough to affect the setting of
2672 compiler options, which is what is needed, but I have no means of testing
2673 whether it really works. (The user who reported this had patched the
2674 generated configure script, which of course I cannot do.)
2675
2676 LATER: After change 22 below (new libtool files), the configure script
2677 seems to know about icc (and also ecc). Therefore, I have commented out
2678 this hack in configure.in.
2679
2680 7. Added support for pkg-config (2 patches were sent in).
2681
2682 8. Negated POSIX character classes that used a combination of internal tables
2683 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2684 [[:^ascii]]. Typically, they would match almost any characters. The other
2685 POSIX classes were not broken in this way.
2686
2687 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2688 to find the match, as PCRE was deluded into thinking that the match had to
2689 start at the start point or following a newline. The same bug applied to
2690 patterns with negative forward assertions or any backward assertions
2691 preceding ".*" at the start, unless the pattern required a fixed first
2692 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2693
2694 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2695 starting at the last subject character, bytes beyond the end of the subject
2696 string were read.
2697
2698 11. Renamed the variable "class" as "classbits" to make life easier for C++
2699 users. (Previously there was a macro definition, but it apparently wasn't
2700 enough.)
2701
2702 12. Added the new field "tables" to the extra data so that tables can be passed
2703 in at exec time, or the internal tables can be re-selected. This allows
2704 a compiled regex to be saved and re-used at a later time by a different
2705 program that might have everything at different addresses.
2706
2707 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2708 -R library as well as a -L library.
2709
2710 14. The debugging options of pcretest (-d on the command line or D on a
2711 pattern) showed incorrect output for anything following an extended class
2712 that contained multibyte characters and which was followed by a quantifier.
2713
2714 15. Added optional support for general category Unicode character properties
2715 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2716 support. It adds about 90K to the size of the library. The meanings of the
2717 inbuilt class escapes such as \d and \s have NOT been changed.
2718
2719 16. Updated pcredemo.c to include calls to free() to release the memory for the
2720 compiled pattern.
2721
2722 17. The generated file chartables.c was being created in the source directory
2723 instead of in the building directory. This caused the build to fail if the
2724 source directory was different from the building directory, and was
2725 read-only.
2726
2727 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2728 file. No doubt somebody will tell me if they don't make sense... Also added
2729 Dan Mooney's comments about building on OpenVMS.
2730
2731 19. Added support for partial matching via the PCRE_PARTIAL option for
2732 pcre_exec() and the \P data escape in pcretest.
2733
2734 20. Extended pcretest with 3 new pattern features:
2735
2736 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2737 write the compiled pattern to the file whose name is "rest-of-line".
2738 This is a straight binary dump of the data, with the saved pointer to
2739 the character tables forced to be NULL. The study data, if any, is
2740 written too. After writing, pcretest reads a new pattern.
2741
2742 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2743 compiled pattern from the given file. There must not be any
2744 occurrences of "<" in the file name (pretty unlikely); if there are,
2745 pcretest will instead treat the initial "<" as a pattern delimiter.
2746 After reading in the pattern, pcretest goes on to read data lines as
2747 usual.
2748
2749 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2750 and 16-bit fields in a compiled pattern, to simulate a pattern that
2751 was compiled on a host of opposite endianness.
2752
2753 21. The pcre-exec() function can now cope with patterns that were compiled on
2754 hosts of opposite endianness, with this restriction:
2755
2756 As for any compiled expression that is saved and used later, the tables
2757 pointer field cannot be preserved; the extra_data field in the arguments
2758 to pcre_exec() should be used to pass in a tables address if a value
2759 other than the default internal tables were used at compile time.
2760
2761 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2762 now diagnosed as an error. Previously, most of the time, a negative number
2763 would have been treated as zero, but if in addition "ovector" was passed as
2764 NULL, a crash could occur.
2765
2766 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2767 new versions from the libtool 1.5 distribution (the last one is a copy of
2768 a file called libtool.m4). This seems to have fixed the need to patch
2769 "configure" to support Darwin 1.3 (which I used to do). However, I still
2770 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2771 workstation).
2772
2773 24. Changed the PCRE licence to be the more standard "BSD" licence.
2774
2775
2776 Version 4.5 01-Dec-03
2777 ---------------------
2778
2779 1. There has been some re-arrangement of the code for the match() function so
2780 that it can be compiled in a version that does not call itself recursively.
2781 Instead, it keeps those local variables that need separate instances for
2782 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2783 needs to "recurse". Keeping track of where control must go is done by means
2784 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2785 hide most of the details from the main code, and operates only if
2786 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2787 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2788 operating.
2789
2790 To make it easier for callers to provide specially tailored get/free
2791 functions for this usage, two new functions, pcre_stack_malloc, and
2792 pcre_stack_free, are used. They are always called in strict stacking order,
2793 and the size of block requested is always the same.
2794
2795 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2796 PCRE has been compiled to use the stack or the heap for recursion. The
2797 -C option of pcretest uses this to show which version is compiled.
2798
2799 A new data escape \S, is added to pcretest; it causes the amounts of store
2800 obtained and freed by both kinds of malloc/free at match time to be added
2801 to the output.
2802
2803 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2804 what's available on my current Linux desktop machine.
2805
2806 3. When matching a UTF-8 string, the test for a valid string at the start has
2807 been extended. If start_offset is not zero, PCRE now checks that it points
2808 to a byte that is the start of a UTF-8 character. If not, it returns
2809 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2810 this is necessary because there may be backward assertions in the pattern.
2811 When matching the same subject several times, it may save resources to use
2812 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2813
2814 4. The code for checking the validity of UTF-8 strings has been tightened so
2815 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2816 containing "overlong sequences".
2817
2818 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2819 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2820 should have been "&", but it just so happened that all the cases this let
2821 through by mistake were picked up later in the function.
2822
2823 6. I had used a variable called "isblank" - this is a C99 function, causing
2824 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2825
2826 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2827 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2828 so the version is shown; (c) stop "make test" echoing "./RunTest".
2829
2830 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2831
2832 9. The prototype for memmove() for systems that don't have it was using
2833 size_t, but the inclusion of the header that defines size_t was later. I've
2834 moved the #includes for the C headers earlier to avoid this.
2835
2836 10. Added some adjustments to the code to make it easier to compiler on certain
2837 special systems:
2838
2839 (a) Some "const" qualifiers were missing.
2840 (b) Added the macro EXPORT before all exported functions; by default this
2841 is defined to be empty.
2842 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2843 that it reads its output file name as an argument instead of writing
2844 to the standard output and assuming this can be redirected.
2845
2846 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2847 class containing characters with values greater than 255, PCRE compilation
2848 went into a loop.
2849
2850 12. A recursive reference to a subpattern that was within another subpattern
2851 that had a minimum quantifier of zero caused PCRE to crash. For example,
2852 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2853 recursion. If the recursively-called subpattern itself had a zero repeat,
2854 that was OK.
2855
2856 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2857 buffer into which it was copied (for escape processing) was still set at
2858 1024, so long lines caused crashes.
2859
2860 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2861 "internal error: code overflow...". This applied to any character class
2862 that was followed by a possessive quantifier.
2863
2864 15. Modified the Makefile to add libpcre.la as a prerequisite for
2865 libpcreposix.la because I was told this is needed for a parallel build to
2866 work.
2867
2868 16. If a pattern that contained .* following optional items at the start was
2869 studied, the wrong optimizing data was generated, leading to matching
2870 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2871 matching string must start with a or b or c. The correct conclusion for
2872 this pattern is that a match can start with any character.
2873
2874
2875 Version 4.4 13-Aug-03
2876 ---------------------
2877
2878 1. In UTF-8 mode, a character class containing characters with values between
2879 127 and 255 was not handled correctly if the compiled pattern was studied.
2880 In fixing this, I have also improved the studying algorithm for such
2881 classes (slightly).
2882
2883 2. Three internal functions had redundant arguments passed to them. Removal
2884 might give a very teeny performance improvement.
2885
2886 3. Documentation bug: the value of the capture_top field in a callout is *one
2887 more than* the number of the hightest numbered captured substring.
2888
2889 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2890 in incorrectly linking with a previously installed version. They now link
2891 explicitly with libpcre.la.
2892
2893 5. configure.in no longer needs to recognize Cygwin specially.
2894
2895 6. A problem in pcre.in for Windows platforms is fixed.
2896
2897 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2898 pcretest, it used to include the size of the study block as part of its
2899 output. Unfortunately, the structure contains a field that has a different
2900 size on different hardware architectures. This meant that the tests that
2901 showed this size failed. As the block is currently always of a fixed size,
2902 this information isn't actually particularly useful in pcretest output, so
2903 I have just removed it.
2904
2905 8. Three pre-processor statements accidentally did not start in column 1.
2906 Sadly, there are *still* compilers around that complain, even though
2907 standard C has not required this for well over a decade. Sigh.
2908
2909 9. In pcretest, the code for checking callouts passed small integers in the
2910 callout_data field, which is a void * field. However, some picky compilers
2911 complained about the casts involved for this on 64-bit systems. Now
2912 pcretest passes the address of the small integer instead, which should get
2913 rid of the warnings.
2914
2915 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2916 both compile and run time, and gives an error if an invalid UTF-8 sequence
2917 is found. There is a option for disabling this check in cases where the
2918 string is known to be correct and/or the maximum performance is wanted.
2919
2920 11. In response to a bug report, I changed one line in Makefile.in from
2921
2922 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2923 to
2924 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2925
2926 to look similar to other lines, but I have no way of telling whether this
2927 is the right thing to do, as I do not use Windows. No doubt I'll get told
2928 if it's wrong...
2929
2930
2931 Version 4.3 21-May-03
2932 ---------------------
2933
2934 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2935 Makefile.
2936
2937 2. Some refactoring to improve the quality of the code:
2938
2939 (i) The utf8_table... variables are now declared "const".
2940
2941 (ii) The code for \cx, which used the "case flipping" table to upper case
2942 lower case letters, now just substracts 32. This is ASCII-specific,
2943 but the whole concept of \cx is ASCII-specific, so it seems
2944 reasonable.
2945
2946 (iii) PCRE was using its character types table to recognize decimal and
2947 hexadecimal digits in the pattern. This is silly, because it handles
2948 only 0-9, a-f, and A-F, but the character types table is locale-
2949 specific, which means strange things might happen. A private
2950 table is now used for this - though it costs 256 bytes, a table is
2951 much faster than multiple explicit tests. Of course, the standard
2952 character types table is still used for matching digits in subject
2953 strings against \d.
2954
2955 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2956 ending in _t are). So I've renamed it as ESC_tee.
2957
2958 3. The first argument for regexec() in the POSIX wrapper should have been
2959 defined as "const".
2960
2961 4. Changed pcretest to use malloc() for its buffers so that they can be
2962 Electric Fenced for debugging.
2963
2964 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2965 to read one or more bytes before the start of the subject string. Often this
2966 had no effect on PCRE's behaviour, but in some circumstances it could
2967 provoke a segmentation fault.
2968
2969 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2970 to try to read one or more bytes before the start of the subject string.
2971
2972 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2973 UTF-8 support could misbehave in various ways if the subject string
2974 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2975 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2976 back over UTF-8 characters.)
2977
2978
2979 Version 4.2 14-Apr-03
2980 ---------------------
2981
2982 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2983
2984 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2985 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2986 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2987 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2988 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2989 and BUILD_EXEEXT
2990 Note: automatic setting of the BUILD variables is not yet working
2991 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2992 compile-time but not at link-time
2993 [LINK]: use for linking executables only
2994 make different versions for Windows and non-Windows
2995 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2996 libraries
2997 [LINK_FOR_BUILD]: new variable
2998 [OBJEXT]: use throughout
2999 [EXEEXT]: use throughout
3000 <winshared>: new target
3001 <wininstall>: new target
3002 <dftables.o>: use native compiler
3003 <dftables>: use native linker
3004 <install>: handle Windows platform correctly
3005 <clean>: ditto
3006 <check>: ditto
3007 copy DLL to top builddir before testing
3008
3009 As part of these changes, -no-undefined was removed again. This was reported
3010 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3011 in any case.
3012
3013 3. Some tidies to get rid of compiler warnings:
3014
3015 . In the match_data structure, match_limit was an unsigned long int, whereas
3016 match_call_count was an int. I've made them both unsigned long ints.
3017
3018 . In pcretest the fact that a const uschar * doesn't automatically cast to
3019 a void * provoked a warning.
3020
3021 . Turning on some more compiler warnings threw up some "shadow" variables
3022 and a few more missing casts.
3023
3024 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3025 option, a class that contained a single character with a value between 128
3026 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3027
3028 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3029 option, a class that contained several characters, but with at least one
3030 whose value was between 128 and 255 caused PCRE to crash.
3031
3032
3033 Version 4.1 12-Mar-03
3034 ---------------------
3035
3036 1. Compiling with gcc -pedantic found a couple of places where casts were
3037 needed, and a string in dftables.c that was longer than standard compilers are
3038 required to support.
3039
3040 2. Compiling with Sun's compiler found a few more places where the code could
3041 be tidied up in order to avoid warnings.
3042
3043 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3044 first of these names is deprecated in the latest Autoconf in favour of the name
3045 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3046 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3047 analogy I have changed it to CFLAGS_FOR_BUILD.
3048
3049 4. Added -no-undefined to the linking command in the Makefile, because this is
3050 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3051 linking step for the pcreposix library.
3052
3053 5. PCRE was failing to diagnose the case of two named groups with the same
3054 name.
3055
3056 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3057 literal character that is needed in the subject for a match, and scans along to
3058 ensure that it is present before embarking on the full matching process. This
3059 saves time in cases of nested unlimited repeats that are never going to match.
3060 Problem: the scan can take a lot of time if the subject is very long (e.g.
3061 megabytes), thus penalizing straightforward matches. It is now done only if the
3062 amount of subject to be scanned is less than 1000 bytes.
3063
3064 7. A lesser problem with the same optimization is that it was recording the
3065 first character of an anchored pattern as "needed", thus provoking a search
3066 right along the subject, even when the first match of the pattern was going to
3067 fail. The "needed" character is now not set for anchored patterns, unless it
3068 follows something in the pattern that is of non-fixed length. Thus, it still
3069 fulfils its original purpose of finding quick non-matches in cases of nested
3070 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3071
3072
3073 Version 4.0 17-Feb-03
3074 ---------------------
3075
3076 1. If a comment in an extended regex that started immediately after a meta-item
3077 extended to the end of string, PCRE compiled incorrect data. This could lead to
3078 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3079
3080 2. Moved to autoconf 2.53 and libtool 1.4.2.
3081
3082 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3083 the special perltest8 script is no longer needed - all the tests can be run
3084 from a single perltest script.
3085
3086 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3087 by \s. It has now been removed in PCRE. This means it isn't recognized as
3088 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3089 class [:space:] *does* include VT, thereby creating a mess.
3090
3091 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3092 space and tab.
3093
3094 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3095 its new features into the main test script, reducing the number of scripts.
3096
3097 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3098 were backward compatible, and made the (?i) apply to the whole pattern, as if
3099 /i were given. Now it behaves more logically, and applies the option setting
3100 only to what follows. PCRE has been changed to follow suit. However, if it
3101 finds options settings right at the start of the pattern, it extracts them into
3102 the global options, as before. Thus, they show up in the info data.
3103
3104 8. Added support for the \Q...\E escape sequence. Characters in between are
3105 treated as literals. This is slightly different from Perl in that $ and @ are
3106 also handled as literals inside the quotes. In Perl, they will cause variable
3107 interpolation. Note the following examples:
3108
3109 Pattern PCRE matches Perl matches
3110
3111 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3112 \Qabc\$xyz\E abc\$xyz abc\$xyz
3113 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3114
3115 For compatibility with Perl, \Q...\E sequences are recognized inside character
3116 classes as well as outside them.
3117
3118 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3119 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3120 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3121 signed/unsigned warnings.
3122
3123 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3124 option for pcretest, so I've replaced it by a simple function that does just
3125 that job.
3126
3127 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3128 "pcregrep -".
3129
3130 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3131 Java package. This provides some syntactic sugar for simple cases of what my
3132 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3133 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3134 item, you can use this simplified notation. Note that only makes sense with
3135 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3136 greediness, whatever the setting of the PCRE_UNGREEDY option.
3137
3138 13. A change of greediness default within a pattern was not taking effect at
3139 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3140 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3141 was abstracted outside.
3142
3143 14. PCRE now supports the \G assertion. It is true when the current matching
3144 position is at the start point of the match. This differs from \A when the
3145 starting offset is non-zero. Used with the /g option of pcretest (or similar
3146 code), it works in the same way as it does for Perl's /g option. If all
3147 alternatives of a regex begin with \G, the expression is anchored to the start
3148 match position, and the "anchored" flag is set in the compiled expression.
3149
3150 15. Some bugs concerning the handling of certain option changes within patterns
3151 have been fixed. These applied to options other than (?ims). For example,
3152 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3153 been the other way round. Some of this was related to change 7 above.
3154
3155 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3156 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3157 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3158 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3159
3160 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3161 mode. Unlike ".", it always matches newline, whatever the setting of
3162 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3163 assertions. Perl allows it, but it doesn't (in general) work because it can't
3164 calculate the length of the lookbehind. At least, that's the case for Perl
3165 5.8.0 - I've been told they are going to document that it doesn't work in
3166 future.
3167
3168 18. Added an error diagnosis for escapes that PCRE does not support: these are
3169 \L, \l, \N, \P, \p, \U, \u, and \X.
3170
3171 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3172 reading past the end of the pattern in cases such as /[abcd/.
3173
3174 20. PCRE was getting more memory than necessary for patterns with classes that
3175 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3176
3177 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3178 compiling PCRE for use with Virtual Pascal.
3179
3180 22. Small fix to the Makefile to make it work properly if the build is done
3181 outside the source tree.
3182
3183 23. Added a new extension: a condition to go with recursion. If a conditional
3184 subpattern starts with (?(R) the "true" branch is used if recursion has
3185 happened, whereas the "false" branch is used only at the top level.
3186
3187 24. When there was a very long string of literal characters (over 255 bytes
3188 without UTF support, over 250 bytes with UTF support), the computation of how
3189 much memory was required could be incorrect, leading to segfaults or other
3190 strange effects.
3191
3192 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3193 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3194 there was a subsequent back reference to those brackets. This meant that, for
3195 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3196 possible to check for precisely this case. All we can do is abandon the
3197 optimization if .* occurs inside capturing brackets when there are any back
3198 references whatsoever. (See below for a better fix that came later.)
3199
3200 26. The handling of the optimization for finding the first character of a
3201 non-anchored pattern, and for finding a character that is required later in the
3202 match were failing in some cases. This didn't break the matching; it just
3203 failed to optimize when it could. The way this is done has been re-implemented.
3204
3205 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3206
3207 28. Added a new feature that provides some of the functionality that Perl
3208 provides with (?{...}). The facility is termed a "callout". The way it is done
3209 in PCRE is for the caller to provide an optional function, by setting
3210 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3211 global variable. By default it is unset, which disables all calling out. To get
3212 the function called, the regex must include (?C) at appropriate points. This
3213 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3214 This provides a means of identifying different callout points. When PCRE
3215 reaches such a point in the regex, if pcre_callout has been set, the external
3216 function is called. It is provided with data in a structure called
3217 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3218 matching continues; if it returns a non-zero value, the match at the current
3219 point fails. However, backtracking will occur if possible. [This was changed
3220 later and other features added - see item 49 below.]
3221
3222 29. pcretest is upgraded to test the callout functionality. It provides a
3223 callout function that displays information. By default, it shows the start of
3224 the match and the current position in the text. There are some new data escapes
3225 to vary what happens:
3226
3227 \C+ in addition, show current contents of captured substrings
3228 \C- do not supply a callout function
3229 \C!n return 1 when callout number n is reached
3230 \C!n!m return 1 when callout number n is reached for the mth time
3231
3232 30. If pcregrep was called with the -l option and just a single file name, it
3233 output "<stdin>" if a match was found, instead of the file name.
3234
3235 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3236 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3237 pcre_exec(). This saves a malloc/free per call. The default value of
3238 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3239 when configuring.
3240
3241 32. The default maximum size of a compiled pattern is 64K. There have been a
3242 few cases of people hitting this limit. The code now uses macros to handle the
3243 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3244 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3245 configuring. Tests 2 and 5 work only with 2-byte links because they output
3246 debugging information about compiled patterns.
3247
3248 33. Internal code re-arrangements:
3249
3250 (a) Moved the debugging function for printing out a compiled regex into
3251 its own source file (printint.c) and used #include to pull it into
3252 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3253 separate copies.
3254
3255 (b) Defined the list of op-code names for debugging as a macro in
3256 internal.h so that it is next to the definition of the opcodes.
3257
3258 (c) Defined a table of op-code lengths for simpler skipping along compiled
3259 code. This is again a macro in internal.h so that it is next to the
3260 definition of the opcodes.
3261
3262 34. Added support for recursive calls to individual subpatterns, along the
3263 lines of Robin Houston's patch (but implemented somewhat differently).
3264
3265 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3266 allow it to read and process whole directories in Win32. This code was
3267 contributed by Lionel Fourquaux; it has not been tested by me.
3268
3269 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3270 used to name a group. Names consist of alphanumerics and underscores, and must
3271 be unique. Back references use the syntax (?P=name) and recursive calls use
3272 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3273 numbers. The function pcre_fullinfo() can be used after compilation to extract
3274 a name/number map. There are three relevant calls:
3275
3276 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3277 PCRE_INFO_NAMECOUNT yields the number of entries
3278 PCRE_INFO_NAMETABLE yields a pointer to the map.
3279
3280 The map is a vector of fixed-size entries. The size of each entry depends on
3281 the length of the longest name used. The first two bytes of each entry are the
3282 group number, most significant byte first. There follows the corresponding
3283 name, zero terminated. The names are in alphabetical order.
3284
3285 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3286 case instead of 255. Making it the same both with and without UTF-8 support
3287 means that the same test output works with both.
3288
3289 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3290 calling malloc() with a zero argument.
3291
3292 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3293 optimization. I've improved things by keeping a bitmap of backreferences with
3294 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3295 fact referenced, the optimization can be applied. It is unlikely that a
3296 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3297 the match to follow \n) will appear inside brackets with a number greater than
3298 31, but if it does, any back reference > 31 suppresses the optimization.
3299
3300 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3301 of disabling numbered capturing parentheses. Any opening parenthesis that is
3302 not followed by ? behaves as if it were followed by ?: but named parentheses
3303 can still be used for capturing (and they will acquire numbers in the usual
3304 way).
3305
3306 41. Redesigned the return codes from the match() function into yes/no/error so
3307 that errors can be passed back from deep inside the nested calls. A malloc
3308 failure while inside a recursive subpattern call now causes the
3309 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3310
3311 42. It is now possible to set a limit on the number of times the match()
3312 function is called in a call to pcre_exec(). This facility makes it possible to
3313 limit the amount of recursion and backtracking, though not in a directly
3314 obvious way, because the match() function is used in a number of different
3315 circumstances. The count starts from zero for each position in the subject
3316 string (for non-anchored patterns). The default limit is, for compatibility, a
3317 large number, namely 10 000 000. You can change this in two ways:
3318
3319 (a) When configuring PCRE before making, you can use --with-match-limit=n
3320 to set a default value for the compiled library.
3321
3322 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3323 a different value is set. See 45 below.
3324
3325 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3326
3327 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3328 of things that can be changed at compile time. The first argument specifies
3329 what is wanted and the second points to where the information is to be placed.
3330 The current list of available information is:
3331
3332 PCRE_CONFIG_UTF8
3333
3334 The output is an integer that is set to one if UTF-8 support is available;
3335 otherwise it is set to zero.
3336
3337 PCRE_CONFIG_NEWLINE
3338
3339 The output is an integer that it set to the value of the code that is used for
3340 newline. It is either LF (10) or CR (13).
3341
3342 PCRE_CONFIG_LINK_SIZE
3343
3344 The output is an integer that contains the number of bytes used for internal
3345 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3346
3347 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3348
3349 The output is an integer that contains the threshold above which the POSIX
3350 interface uses malloc() for output vectors. See item 31 above.
3351
3352 PCRE_CONFIG_MATCH_LIMIT
3353
3354 The output is an unsigned integer that contains the default limit of the number
3355 of match() calls in a pcre_exec() execution. See 42 above.
3356
3357 44. pcretest has been upgraded by the addition of the -C option. This causes it
3358 to extract all the available output from the new pcre_config() function, and to
3359 output it. The program then exits immediately.
3360
3361 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3362 order to support additional features. One way would have been to define
3363 pcre_exec2() (for example) with extra arguments, but this would not have been
3364 extensible, and would also have required all calls to the original function to
3365 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3366 is used for passing in "extra" data from pcre_study().
3367
3368 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3369 contains the following fields:
3370
3371 flags a bitmap indicating which of the following fields are set
3372 study_data opaque data from pcre_study()
3373 match_limit a way of specifying a limit on match() calls for a specific
3374 call to pcre_exec()
3375 callout_data data for callouts (see 49 below)
3376
3377 The flag bits are also defined in pcre.h, and are
3378
3379 PCRE_EXTRA_STUDY_DATA
3380 PCRE_EXTRA_MATCH_LIMIT
3381 PCRE_EXTRA_CALLOUT_DATA
3382
3383 The pcre_study() function now returns one of these new pcre_extra blocks, with
3384 the actual study data pointed to by the study_data field, and the
3385 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3386 before. That is, this change is entirely upwards-compatible and requires no
3387 change to existing code.
3388
3389 If you want to pass in additional data to pcre_exec(), you can either place it
3390 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3391 block.
3392
3393 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3394 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3395 times with different match limits, until it finds the minimum value needed for
3396 pcre_exec() to complete. The value is then output. This can be instructive; for
3397 most simple matches the number is quite small, but for pathological cases it
3398 gets very large very quickly.
3399
3400 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3401 returns the size of the data block pointed to by the study_data field in a
3402 pcre_extra block, that is, the value that was passed as the argument to
3403 pcre_malloc() when PCRE was getting memory in which to place the information
3404 created by pcre_study(). The fourth argument should point to a size_t variable.
3405 pcretest has been extended so that this information is shown after a successful
3406 pcre_study() call when information about the compiled regex is being displayed.
3407
3408 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3409 because what follows is always an absolute path. (Later: it turns out that this
3410 is more than cosmetic for MinGW, because it doesn't like empty path
3411 components.)
3412
3413 49. Some changes have been made to the callout feature (see 28 above):
3414
3415 (i) A callout function now has three choices for what it returns:
3416
3417 0 => success, carry on matching
3418 > 0 => failure at this point, but backtrack if possible
3419 < 0 => serious error, return this value from pcre_exec()
3420
3421 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3422 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3423 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3424 use by callout functions. It will never be used by PCRE itself.
3425
3426 (ii) The pcre_extra structure (see 45 above) has a void * field called
3427 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3428 pcre_callout_block structure has a field of the same name. The contents of
3429 the field passed in the pcre_extra structure are passed to the callout
3430 function in the corresponding field in the callout block. This makes it
3431 easier to use the same callout-containing regex from multiple threads. For
3432 testing, the pcretest program has a new data escape
3433
3434 \C*n pass the number n (may be negative) as callout_data
3435
3436 If the callout function in pcretest receives a non-zero value as
3437 callout_data, it returns that value.
3438
3439 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3440 there were some redundant $(CFLAGS) in commands that are now specified as
3441 $(LINK), which already includes $(CFLAGS).
3442
3443 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3444 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3445 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3446 one-byte characters throughout. Note that case-insensitive matching applies
3447 only to characters whose values are less than 256. PCRE doesn't support the
3448 notion of cases for higher-valued characters.
3449
3450 (i) A character class whose characters are all within 0-255 is handled as
3451 a bit map, and the map is inverted for negative classes. Previously, a
3452 character > 255 always failed to match such a class; however it should
3453 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3454
3455 (ii) A negated character class with a single character < 255 is coded as
3456 "not this character" (OP_NOT). This wasn't working properly when the test
3457 character was multibyte, either singly or repeated.
3458
3459 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3460 mode, for example: \x{100}{2,3}.
3461
3462 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3463 singly or repeated) now correctly test multibyte characters. However,
3464 PCRE doesn't recognize any characters with values greater than 255 as
3465 digits, spaces, or word characters. Such characters always match \D, \S,
3466 and \W, and never match \d, \s, or \w.
3467
3468 (v) Classes may now contain characters and character ranges with values
3469 greater than 255. For example: [ab\x{100}-\x{400}].
3470
3471 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3472 PCRE in UTF-8 mode.
3473
3474 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3475 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3476 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3477 value.)
3478
3479 53. The single man page has become too large. I have therefore split it up into
3480 a number of separate man pages. These also give rise to individual HTML pages;
3481 these are now put in a separate directory, and there is an index.html page that
3482 lists them all. Some hyperlinking between the pages has been installed.
3483
3484 54. Added convenience functions for handling named capturing parentheses.
3485
3486 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3487 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3488 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3489 are faulted.
3490
3491 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3492 calling configure. These values are used when compiling the dftables.c program
3493 which is run to generate the source of the default character tables. They
3494 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3495 you will need to set these values.
3496
3497 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3498
3499
3500 Version 3.9 02-Jan-02
3501 ---------------------
3502
3503 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3504
3505 2. If --disable-static was given, the building process failed when trying to
3506 build pcretest and pcregrep. (For some reason it was using libtool to compile
3507 them, which is not right, as they aren't part of the library.)
3508
3509
3510 Version 3.8 18-Dec-01
3511 ---------------------
3512
3513 1. The experimental UTF-8 code was completely screwed up. It was packing the
3514 bytes in the wrong order. How dumb can you get?
3515
3516
3517 Version 3.7 29-Oct-01
3518 ---------------------
3519
3520 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3521 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3522 this didn't happen under Solaris 8, where I normally test things.
3523
3524 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3525 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3526 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3527 this fix an hour or so after the initial 3.7 release.)
3528
3529
3530 Version 3.6 23-Oct-01
3531 ---------------------
3532
3533 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3534 offsets passed as NULL with zero offset count.
3535
3536 2. The config.guess and config.sub files had not been updated when I moved to
3537 the latest autoconf.
3538
3539
3540 Version 3.5 15-Aug-01
3541 ---------------------
3542
3543 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3544 had been forgotten.
3545
3546 2. By using declared but undefined structures, we can avoid using "void"
3547 definitions in pcre.h while keeping the internal definitions of the structures
3548 private.
3549
3550 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3551 user point of view, this means that both static and shared libraries are built
3552 by default, but this can be individually controlled. More of the work of
3553 handling this static/shared cases is now inside libtool instead of PCRE's make
3554 file.
3555
3556 4. The pcretest utility is now installed along with pcregrep because it is
3557 useful for users (to test regexs) and by doing this, it automatically gets
3558 relinked by libtool. The documentation has been turned into a man page, so
3559 there are now .1, .txt, and .html versions in /doc.
3560
3561 5. Upgrades to pcregrep:
3562 (i) Added long-form option names like gnu grep.
3563 (ii) Added --help to list all options with an explanatory phrase.
3564 (iii) Added -r, --recursive to recurse into sub-directories.
3565 (iv) Added -f, --file to read patterns from a file.
3566
3567 6. pcre_exec() was referring to its "code" argument before testing that
3568 argument for NULL (and giving an error if it was NULL).
3569
3570 7. Upgraded Makefile.in to allow for compiling in a different directory from
3571 the source directory.
3572
3573 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3574 options bits, the pointer it was passed was to an int instead of to an unsigned
3575 long int. This mattered only on 64-bit systems.
3576
3577 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3578 generated) instead of pcre.in, which it its source. Also made the same change
3579 in several of the .c files.
3580
3581 10. A new release of gcc defines printf() as a macro, which broke pcretest
3582 because it had an ifdef in the middle of a string argument for printf(). Fixed
3583 by using separate calls to printf().
3584
3585 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3586 script, to force use of CR or LF instead of \n in the source. On non-Unix
3587 systems, the value can be set in config.h.
3588
3589 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3590 absolute limit. Changed the text of the error message to make this clear, and
3591 likewise updated the man page.
3592
3593 13. The limit of 99 on the number of capturing subpatterns has been removed.
3594 The new limit is 65535, which I hope will not be a "real" limit.
3595
3596
3597 Version 3.4 22-Aug-00
3598 ---------------------
3599
3600 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3601
3602 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3603
3604
3605 Version 3.3 01-Aug-00
3606 ---------------------
3607
3608 1. If an octal character was given, but the value was greater than \377, it
3609 was not getting masked to the least significant bits, as documented. This could
3610 lead to crashes in some systems.
3611
3612 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3613 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3614
3615 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3616 These just pass their arguments on to (pcre_free)(), but they are provided
3617 because some uses of PCRE bind it to non-C systems that can call its functions,
3618 but cannot call free() or pcre_free() directly.
3619
3620 4. Add "make test" as a synonym for "make check". Corrected some comments in
3621 the Makefile.
3622
3623 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3624 Makefile.
3625
3626 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3627 command called pgrep for grepping around the active processes.
3628
3629 7. Added the beginnings of support for UTF-8 character strings.
3630
3631 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3632 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3633 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3634 out for the ar command.)
3635
3636
3637 Version 3.2 12-May-00
3638 ---------------------
3639
3640 This is purely a bug fixing release.
3641
3642 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3643 of ZA. This was just one example of several cases that could provoke this bug,
3644 which was introduced by change 9 of version 2.00. The code for breaking
3645 infinite loops after an iteration that matches an empty string was't working
3646 correctly.
3647
3648 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3649 when matched against abbab (for example). After matching an empty string, it
3650 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3651 caused it to match further down the string than it should.
3652
3653 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3654 was there because it doesn't seem to be needed, and it causes trouble on some
3655 systems, as it is not a Standard C header. It has been removed.
3656
3657 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3658 were reported on the Macintosh. The changes were from
3659
3660 while ((c = *(++ptr)) != 0 && c != '\n');
3661 to
3662 while ((c = *(++ptr)) != 0 && c != '\n') ;
3663
3664 Totally extraordinary, but if that's what it takes...
3665
3666 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3667 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3668 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3669 assumes the way PCRE uses memmove() (always moving upwards).
3670
3671 6. PCRE is being used in one environment where strchr() is not available. There
3672 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3673 faster code anyway.
3674
3675
3676 Version 3.1 09-Feb-00
3677 ---------------------
3678
3679 The only change in this release is the fixing of some bugs in Makefile.in for
3680 the "install" target:
3681
3682 (1) It was failing to install pcreposix.h.
3683
3684 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3685
3686
3687 Version 3.0 01-Feb-00
3688 ---------------------
3689
3690 1. Add support for the /+ modifier to perltest (to output $` like it does in
3691 pcretest).
3692
3693 2. Add support for the /g modifier to perltest.
3694
3695 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3696 matches null strings.
3697
3698 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3699 pattern. Perl treats empty patterns specially - it reuses the most recent
3700 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3701 effect.
3702
3703 5. The POSIX interface was broken in that it was just handing over the POSIX
3704 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3705 required a bigger vector, with some working space on the end. This means that
3706 the POSIX wrapper now has to get and free some memory, and copy the results.
3707
3708 6. Added some simple autoconf support, placing the test data and the
3709 documentation in separate directories, re-organizing some of the
3710 information files, and making it build pcre-config (a GNU standard). Also added
3711 libtool support for building PCRE as a shared library, which is now the
3712 default.
3713
3714 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3715 09 are not valid octal constants. Single digits will be used for minor values
3716 less than 10.
3717
3718 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3719 existing programs that set these in the POSIX interface can use PCRE without
3720 modification.
3721
3722 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3723 return all that pcre_info() returns, plus additional data. The pcre_info()
3724 function is retained for compatibility, but is considered to be obsolete.
3725
3726 10. Added experimental recursion feature (?R) to handle one common case that
3727 Perl 5.6 will be able to do with (?p{...}).
3728
3729 11. Added support for POSIX character classes like [:alpha:], which Perl is
3730 adopting.
3731
3732
3733 Version 2.08 31-Aug-99
3734 ----------------------
3735
3736 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3737 trying to match at the startoffset position, but instead was moving forward to
3738 the next newline as if a previous match had failed.
3739
3740 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3741 and could get into a loop if a null string was matched other than at the start
3742 of the subject.
3743
3744 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3745 be distinguished at compile time, and for completeness also added PCRE_DATE.
3746
3747 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3748 in GnuWin32 environments.
3749
3750
3751 Version 2.07 29-Jul-99
3752 ----------------------
3753
3754 1. The documentation is now supplied in plain text form and HTML as well as in
3755 the form of man page sources.
3756
3757 2. C++ compilers don't like assigning (void *) values to other pointer types.
3758 In particular this affects malloc(). Although there is no problem in Standard
3759 C, I've put in casts to keep C++ compilers happy.
3760
3761 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3762 should be (const char *).
3763
3764 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3765 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3766 However, I haven't made this a standard facility. The documentation doesn't
3767 mention it, and the Makefile doesn't support it.
3768
3769 5. The Makefile now contains an "install" target, with editable destinations at
3770 the top of the file. The pcretest program is not installed.
3771
3772 6. pgrep -V now gives the PCRE version number and date.
3773
3774 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3775 causing the entire string to be ignored, instead of just the last character.
3776
3777 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3778 non-matching string, it can take a very, very long time, even for strings of
3779 quite modest length, because of the nested recursion. PCRE now does better in
3780 some of these cases. It does this by remembering the last required literal
3781 character in the pattern, and pre-searching the subject to ensure it is present
3782 before running the real match. In other words, it applies a heuristic to detect
3783 some types of certain failure quickly, and in the above example, if presented
3784 with a string that has no trailing " it gives "no match" very quickly.
3785
3786 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3787 other alternatives are tried instead.
3788
3789
3790 Version 2.06 09-Jun-99
3791 ----------------------
3792
3793 1. Change pcretest's output for amount of store used to show just the code
3794 space, because the remainder (the data block) varies in size between 32-bit and
3795 64-bit systems.
3796
3797 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3798 start matching at. This allows lookbehinds to work when searching for multiple
3799 occurrences in a string.
3800
3801 3. Added additional options to pcretest for testing multiple occurrences:
3802
3803 /+ outputs the rest of the string that follows a match
3804 /g loops for multiple occurrences, using the new startoffset argument
3805 /G loops for multiple occurrences by passing an incremented pointer
3806
3807 4. PCRE wasn't doing the "first character" optimization for patterns starting
3808 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3809 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3810 the letter 'x'. On long subject strings, this gives a significant speed-up.
3811
3812
3813 Version 2.05 21-Apr-99
3814 ----------------------
3815
3816 1. Changed the type of magic_number from int to long int so that it works
3817 properly on 16-bit systems.
3818
3819 2. Fixed a bug which caused patterns starting with .* not to work correctly
3820 when the subject string contained newline characters. PCRE was assuming
3821 anchoring for such patterns in all cases, which is not correct because .* will
3822 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3823 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3824 must be retried after every newline in the subject.
3825
3826
3827 Version 2.04 18-Feb-99
3828 ----------------------
3829
3830 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3831 computation of the store needed to hold the pattern was incorrect (too large).
3832 If such patterns were nested a few deep, this could multiply and become a real
3833 problem.
3834
3835 2. Added /M option to pcretest to show the memory requirement of a specific
3836 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3837
3838 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3839 compiled in such a way that the backtracking after subsequent failure was
3840 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3841 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3842
3843
3844 Version 2.03 02-Feb-99
3845 ----------------------
3846
3847 1. Fixed typo and small mistake in man page.
3848
3849 2. Added 4th condition (GPL supersedes if conflict) and created separate
3850 LICENCE file containing the conditions.
3851
3852 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3853 Perl, that is the internal \ allows the delimiter to be included in the
3854 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3855 the final delimiter, add \ to the end of the pattern (to test the error).
3856
3857 4. Added the convenience functions for extracting substrings after a successful
3858 match. Updated pcretest to make it able to test these functions.
3859
3860
3861 Version 2.02 14-Jan-99
3862 ----------------------
3863
3864 1. Initialized the working variables associated with each extraction so that
3865 their saving and restoring doesn't refer to uninitialized store.
3866
3867 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3868 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3869 fix the problem.
3870
3871 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3872 calls, and wasn't printing the correct value for compiling calls. Increased the
3873 default value of LOOPREPEAT, and the number of significant figures in the
3874 times.
3875
3876 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3877
3878 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3879 a building problem on Windows NT with a FAT file system.
3880
3881
3882 Version 2.01 21-Oct-98
3883 ----------------------
3884
3885 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3886 to character tables built by pcre_maketables() in the current locale. If NULL
3887 is passed, the default tables are used.
3888
3889
3890 Version 2.00 24-Sep-98
3891 ----------------------
3892
3893 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3894 it any more.
3895
3896 2. Allow quantification of (?>) groups, and make it work correctly.
3897
3898 3. The first character computation wasn't working for (?>) groups.
3899
3900 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3901 end of the subject) and add 5.005's \z, which really does match only at the
3902 very end of the subject.
3903
3904 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3905
3906 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3907 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3908 localized options. All options to pcre_study() were also removed.
3909
3910 7. Add other new features from 5.005:
3911
3912 $(?<= positive lookbehind
3913 $(?<! negative lookbehind
3914 (?imsx-imsx) added the unsetting capability
3915 such a setting is global if at outer level; local otherwise
3916 (?imsx-imsx:) non-capturing groups with option setting
3917 (?(cond)re|re) conditional pattern matching
3918
3919 A backreference to itself in a repeated group matches the previous
3920 captured string.
3921
3922 8. General tidying up of studying (both automatic and via "study")
3923 consequential on the addition of new assertions.
3924
3925 9. As in 5.005, unlimited repeated groups that could match an empty substring
3926 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3927 runtime if any iteration does actually match an empty substring.
3928
3929 10. Include the RunTest script in the distribution.
3930
3931 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3932 discrepancies, some of which were old and were also with respect to 5.004. They
3933 have now been fixed.
3934
3935
3936 Version 1.09 28-Apr-98
3937 ----------------------
3938
3939 1. A negated single character class followed by a quantifier with a minimum
3940 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3941 program crashes, or just wrong answers. This did not apply to negated classes
3942 containing more than one character, or to minima other than one.
3943
3944
3945 Version 1.08 27-Mar-98
3946 ----------------------
3947
3948 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3949
3950 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3951 latter must appear before anything that relies on it in the pattern.
3952
3953
3954 Version 1.07 16-Feb-98
3955 ----------------------
3956
3957 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3958 repeat of a potentially empty string).
3959
3960
3961 Version 1.06 23-Jan-98
3962 ----------------------
3963
3964 1. Added Markus Oberhumer's little patches for C++.
3965
3966 2. Literal strings longer than 255 characters were broken.
3967
3968
3969 Version 1.05 23-Dec-97
3970 ----------------------
3971
3972 1. Negated character classes containing more than one character were failing if
3973 PCRE_CASELESS was set at run time.
3974
3975
3976 Version 1.04 19-Dec-97
3977 ----------------------
3978
3979 1. Corrected the man page, where some "const" qualifiers had been omitted.
3980
3981 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3982 input syntax.
3983
3984 3. Fixed memory leak which occurred when a regex with back references was
3985 matched with an offsets vector that wasn't big enough. The temporary memory
3986 that is used in this case wasn't being freed if the match failed.
3987
3988 4. Tidied pcretest to ensure it frees memory that it gets.
3989
3990 5. Temporary memory was being obtained in the case where the passed offsets
3991 vector was exactly big enough.
3992
3993 6. Corrected definition of offsetof() from change 5 below.
3994
3995 7. I had screwed up change 6 below and broken the rules for the use of
3996 setjmp(). Now fixed.
3997
3998
3999 Version 1.03 18-Dec-97
4000 ----------------------
4001
4002 1. A erroneous regex with a missing opening parenthesis was correctly
4003 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4004 on some systems.
4005
4006 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4007 it was reported that one broken compiler failed on the former because "code" is
4008 also an independent variable.
4009
4010 3. The erroneous regex a[]b caused an array overrun reference.
4011
4012 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4013 fail on data ending with that character. (It was going on too far, and checking
4014 the next character, typically a binary zero.) This was specific to the
4015 optimized code for single-character negative classes.
4016
4017 5. Added a contributed patch from the TIN world which does the following:
4018
4019 + Add an undef for memmove, in case the the system defines a macro for it.
4020
4021 + Add a definition of offsetof(), in case there isn't one. (I don't know
4022 the reason behind this - offsetof() is part of the ANSI standard - but
4023 it does no harm).
4024
4025 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4026 most of the places where whitespace preceded '#'. I have given up and
4027 allowed the remaining 2 cases to be at the margin.
4028
4029 + Rename some variables in pcre to eliminate shadowing. This seems very
4030 pedantic, but does no harm, of course.
4031
4032 6. Moved the call to setjmp() into its own function, to get rid of warnings
4033 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4034
4035 7. Constructs such as \d{8,} were compiling into the equivalent of
4036 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4037 outcome, but in this particular case used more store than had been allocated,
4038 which caused the bug to be discovered because it threw up an internal error.
4039
4040 8. The debugging code in both pcre and pcretest for outputting the compiled
4041 form of a regex was going wrong in the case of back references followed by
4042 curly-bracketed repeats.
4043
4044
4045 Version 1.02 12-Dec-97
4046 ----------------------
4047
4048 1. Typos in pcre.3 and comments in the source fixed.
4049
4050 2. Applied a contributed patch to get rid of places where it used to remove
4051 'const' from variables, and fixed some signed/unsigned and uninitialized
4052 variable warnings.
4053
4054 3. Added the "runtest" target to Makefile.
4055
4056 4. Set default compiler flag to -O2 rather than just -O.
4057
4058
4059 Version 1.01 19-Nov-97
4060 ----------------------
4061
4062 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4063 like /([ab]*)*/, that is, for classes with more than one character in them.
4064
4065 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4066 as /((?>a*))*/ (a PCRE_EXTRA facility).
4067
4068
4069 Version 1.00 18-Nov-97
4070 ----------------------
4071
4072 1. Added compile-time macros to support systems such as SunOS4 which don't have
4073 memmove() or strerror() but have other things that can be used instead.
4074
4075 2. Arranged that "make clean" removes the executables.
4076
4077
4078 Version 0.99 27-Oct-97
4079 ----------------------
4080
4081 1. Fixed bug in code for optimizing classes with only one character. It was
4082 initializing a 32-byte map regardless, which could cause it to run off the end
4083 of the memory it had got.
4084
4085 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4086
4087
4088 Version 0.98 22-Oct-97
4089 ----------------------
4090
4091 1. Fixed bug in code for handling temporary memory usage when there are more
4092 back references than supplied space in the ovector. This could cause segfaults.
4093
4094
4095 Version 0.97 21-Oct-97
4096 ----------------------
4097
4098 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4099
4100 2. Optimized negated single characters not to use a bit map.
4101
4102 3. Brought error texts together as macro definitions; clarified some of them;
4103 fixed one that was wrong - it said "range out of order" when it meant "invalid
4104 escape sequence".
4105
4106 4. Changed some char * arguments to const char *.
4107
4108 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4109
4110 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4111 pcretest.
4112
4113
4114 Version 0.96 16-Oct-97
4115 ----------------------
4116
4117 1. Added a simple "pgrep" utility to the distribution.
4118
4119 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4120 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4121 where "ddd" means "one or more decimal digits".
4122
4123 3. Fixed serious bug. If a pattern had a back reference, but the call to
4124 pcre_exec() didn't supply a large enough ovector to record the related
4125 identifying subpattern, the match always failed. PCRE now remembers the number
4126 of the largest back reference, and gets some temporary memory in which to save
4127 the offsets during matching if necessary, in order to ensure that
4128 backreferences always work.
4129
4130 4. Increased the compatibility with Perl in a number of ways:
4131
4132 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4133 to request this handling. The option can be set at compile or exec time.
4134
4135 (b) $ matches before a terminating newline by default; an option
4136 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4137 mode). The option can be set at compile or exec time.
4138
4139 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4140 the same as Perl's. If the decimal number it represents is less than 10
4141 or there aren't that many previous left capturing parentheses, an octal
4142 escape is read. Inside a character class, it's always an octal escape,
4143 even if it is a single digit.
4144
4145 (d) An escaped but undefined alphabetic character is taken as a literal,
4146 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4147 escapes.
4148
4149 (e) {0} is now permitted. (The previous item is removed from the compiled
4150 pattern).
4151
4152 5. Changed all the names of code files so that the basic parts are no longer
4153 than 10 characters, and abolished the teeny "globals.c" file.
4154
4155 6. Changed the handling of character classes; they are now done with a 32-byte
4156 bit map always.
4157
4158 7. Added the -d and /D options to pcretest to make it possible to look at the
4159 internals of compilation without having to recompile pcre.
4160
4161
4162 Version 0.95 23-Sep-97
4163 ----------------------
4164
4165 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4166 \x20 at the start of a run of normal characters. These were being treated as
4167 real characters, instead of the source characters being re-checked.
4168
4169
4170 Version 0.94 18-Sep-97
4171 ----------------------
4172
4173 1. The functions are now thread-safe, with the caveat that the global variables
4174 containing pointers to malloc() and free() or alternative functions are the
4175 same for all threads.
4176
4177 2. Get pcre_study() to generate a bitmap of initial characters for non-
4178 anchored patterns when this is possible, and use it if passed to pcre_exec().
4179
4180
4181 Version 0.93 15-Sep-97
4182 ----------------------
4183
4184 1. /(b)|(:+)/ was computing an incorrect first character.
4185
4186 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4187 but not actually doing anything yet.
4188
4189 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4190 as Perl does (e.g. [-az] or [az-]).
4191
4192 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4193 all possible positions.
4194
4195 5. Split up into different modules to avoid including unneeded functions in a
4196 compiled binary. However, compile and exec are still in one module. The "study"
4197 function is split off.
4198
4199 6. The character tables are now in a separate module whose source is generated
4200 by an auxiliary program - but can then be edited by hand if required. There are
4201 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4202 toupper() in the code.
4203
4204 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4205 make them global. Abolish the function for setting them, as the caller can now
4206 set them directly.
4207
4208
4209 Version 0.92 11-Sep-97
4210 ----------------------
4211
4212 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4213 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4214
4215 2. Caseless matching was not working in character classes if the characters in
4216 the pattern were in upper case.
4217
4218 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4219
4220 4. Make PCRE_ANCHORED public and accept as a compile option.
4221
4222 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4223 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4224 pass them.
4225
4226 6. Give an error if bad option bits passed at compile or run time.
4227
4228 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4229 pcretest to cause it to pass that flag.
4230
4231 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4232 options, and the first character, if set.
4233
4234 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4235
4236
4237 Version 0.91 10-Sep-97
4238 ----------------------
4239
4240 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4241 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4242
4243 2. PCRE was looping on encountering an indefinitely repeated back reference to
4244 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4245 Perl does - treats the match as successful.
4246
4247 ****

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