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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 626 - (show annotations)
Wed Jul 20 17:51:54 2011 UTC (3 years, 10 months ago) by ph10
File size: 207087 byte(s)
Error occurred while calculating annotation data.
Add the /= modifier to pcretest so as to be able to check unset capturing 
parentheses at the ends of patterns.
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 23. Add the ++ qualifier feature to pcretest, to show the remainder of the
127 subject after a captured substring (to make it easier to tell which of a
128 number of identical substrings has been captured).
129
130 24. The way atomic groups are processed by pcre_exec() has been changed so that
131 if they are repeated, backtracking one repetition now resets captured
132 values correctly. For example, if ((?>(a+)b)+aabab) is matched against
133 "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
134 "aaa". Previously, it would have been "a". As part of this code
135 refactoring, the way recursive calls are handled has also been changed.
136
137 24. If an assertion condition captured any substrings, they were not passed
138 back unless some other capturing happened later. For example, if
139 (?(?=(a))a) was matched against "a", no capturing was returned.
140
141 25. When studying a pattern that contained subroutine calls or assertions,
142 the code for finding the minimum length of a possible match was handling
143 direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
144 group 1 called group 2 while simultaneously a separate group 2 called group
145 1). A stack overflow occurred in this case. I have fixed this by limiting
146 the recursion depth to 10.
147
148 26. Updated RunTest.bat in the distribution to the version supplied by Tom
149 Fortmann. This supports explicit test numbers on the command line, and has
150 argument validation and error reporting.
151
152 27. An instance of \X with an unlimited repeat could fail if at any point the
153 first character it looked at was a mark character.
154
155 28. Some minor code refactoring concerning Unicode properties and scripts
156 should reduce the stack requirement of match() slightly.
157
158 29. Added the '=' option to pcretest to check the setting of unused capturing
159 slots at the end of the pattern, which are documented as being -1, but are
160 not included in the return count.
161
162
163 Version 8.12 15-Jan-2011
164 ------------------------
165
166 1. Fixed some typos in the markup of the man pages, and wrote a script that
167 checks for such things as part of the documentation building process.
168
169 2. On a big-endian 64-bit system, pcregrep did not correctly process the
170 --match-limit and --recursion-limit options (added for 8.11). In
171 particular, this made one of the standard tests fail. (The integer value
172 went into the wrong half of a long int.)
173
174 3. If the --colour option was given to pcregrep with -v (invert match), it
175 did strange things, either producing crazy output, or crashing. It should,
176 of course, ignore a request for colour when reporting lines that do not
177 match.
178
179 4. Another pcregrep bug caused similar problems if --colour was specified with
180 -M (multiline) and the pattern match finished with a line ending.
181
182 5. In pcregrep, when a pattern that ended with a literal newline sequence was
183 matched in multiline mode, the following line was shown as part of the
184 match. This seems wrong, so I have changed it.
185
186 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
187 the check for further matches in the same line (so they could be coloured)
188 to overrun the end of the current line. If another match was found, it was
189 incorrectly shown (and then shown again when found in the next line).
190
191 7. If pcregrep was compiled under Windows, there was a reference to the
192 function pcregrep_exit() before it was defined. I am assuming this was
193 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
194 reported by a user. I've moved the definition above the reference.
195
196
197 Version 8.11 10-Dec-2010
198 ------------------------
199
200 1. (*THEN) was not working properly if there were untried alternatives prior
201 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
202 backtracked to try for "b" instead of moving to the next alternative branch
203 at the same level (in this case, to look for "c"). The Perl documentation
204 is clear that when (*THEN) is backtracked onto, it goes to the "next
205 alternative in the innermost enclosing group".
206
207 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
208 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
209 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
210 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
211 (*THEN).
212
213 3. If \s appeared in a character class, it removed the VT character from
214 the class, even if it had been included by some previous item, for example
215 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
216 of \s, but is part of the POSIX "space" class.)
217
218 4. A partial match never returns an empty string (because you can always
219 match an empty string at the end of the subject); however the checking for
220 an empty string was starting at the "start of match" point. This has been
221 changed to the "earliest inspected character" point, because the returned
222 data for a partial match starts at this character. This means that, for
223 example, /(?<=abc)def/ gives a partial match for the subject "abc"
224 (previously it gave "no match").
225
226 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
227 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
228 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
229 has an implication that the given string is incomplete (because a partial
230 match is preferred over a full match). For this reason, these items now
231 give a partial match in this situation. [Aside: previously, the one case
232 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
233 match rather than a full match, which was wrong by the old rules, but is
234 now correct.]
235
236 6. There was a bug in the handling of #-introduced comments, recognized when
237 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
238 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
239 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
240 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
241 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
242 places in pcre_compile().
243
244 7. Related to (6) above, when pcre_compile() was skipping #-introduced
245 comments when looking ahead for named forward references to subpatterns,
246 the only newline sequence it recognized was NL. It now handles newlines
247 according to the set newline convention.
248
249 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
250 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
251 cater for a lack of strerror(). These oversights have been fixed.
252
253 9. Added --match-limit and --recursion-limit to pcregrep.
254
255 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
256
257 11. When the -o option was used, pcregrep was setting a return code of 1, even
258 when matches were found, and --line-buffered was not being honoured.
259
260 12. Added an optional parentheses number to the -o and --only-matching options
261 of pcregrep.
262
263 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
264 can match an empty string. The code to do it in pcretest and pcredemo
265 needed fixing:
266
267 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
268 only one byte after an empty string match just before CRLF (this case
269 just got forgotten; "any" and "anycrlf" were OK).
270
271 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
272 mode when an empty string match preceded an ASCII character followed by
273 a non-ASCII character. (The code for advancing by one character rather
274 than one byte was nonsense.)
275
276 (c) The pcredemo.c sample program did not have any code at all to handle
277 the cases when CRLF is a valid newline sequence.
278
279 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
280 as a starting offset was within the subject string. There is now a new
281 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
282 negative or greater than the length of the string. In order to test this,
283 pcretest is extended to allow the setting of negative starting offsets.
284
285 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
286 starting offset points to the beginning of a UTF-8 character was
287 unnecessarily clumsy. I tidied it up.
288
289 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
290 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
291
292 17. Nobody had reported that the --include_dir option, which was added in
293 release 7.7 should have been called --include-dir (hyphen, not underscore)
294 for compatibility with GNU grep. I have changed it to --include-dir, but
295 left --include_dir as an undocumented synonym, and the same for
296 --exclude-dir, though that is not available in GNU grep, at least as of
297 release 2.5.4.
298
299 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
300 characters from a string of bytes) have been redefined so as not to use
301 loops, in order to improve performance in some environments. At the same
302 time, I abstracted some of the common code into auxiliary macros to save
303 repetition (this should not affect the compiled code).
304
305 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
306 compile-time error is now given if \c is not followed by an ASCII
307 character, that is, a byte less than 128. (In EBCDIC mode, the code is
308 different, and any byte value is allowed.)
309
310 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
311 START_OPTIMIZE option, which is now allowed at compile time - but just
312 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
313 to pcregrep and other applications that have no direct access to PCRE
314 options. The new /Y option in pcretest sets this option when calling
315 pcre_compile().
316
317 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
318 back references. Groups containing recursive back references were forced to
319 be atomic by that change, but in the case of named groups, the amount of
320 memory required was incorrectly computed, leading to "Failed: internal
321 error: code overflow". This has been fixed.
322
323 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
324 pcretest.c, to avoid build problems in some Borland environments.
325
326
327 Version 8.10 25-Jun-2010
328 ------------------------
329
330 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
331 THEN.
332
333 2. (*ACCEPT) was not working when inside an atomic group.
334
335 3. Inside a character class, \B is treated as a literal by default, but
336 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
337 causes the error). The code is unchanged, but I tidied the documentation.
338
339 4. Inside a character class, PCRE always treated \R and \X as literals,
340 whereas Perl faults them if its -w option is set. I have changed PCRE so
341 that it faults them when PCRE_EXTRA is set.
342
343 5. Added support for \N, which always matches any character other than
344 newline. (It is the same as "." when PCRE_DOTALL is not set.)
345
346 6. When compiling pcregrep with newer versions of gcc which may have
347 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
348 declared with attribute warn_unused_result" were given. Just casting the
349 result to (void) does not stop the warnings; a more elaborate fudge is
350 needed. I've used a macro to implement this.
351
352 7. Minor change to pcretest.c to avoid a compiler warning.
353
354 8. Added four artifical Unicode properties to help with an option to make
355 \s etc use properties (see next item). The new properties are: Xan
356 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
357
358 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
359 use Unicode properties. (*UCP) at the start of a pattern can be used to set
360 this option. Modified pcretest to add /W to test this facility. Added
361 REG_UCP to make it available via the POSIX interface.
362
363 10. Added --line-buffered to pcregrep.
364
365 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
366 studied, and the match started with a letter with a code point greater than
367 127 whose first byte was different to the first byte of the other case of
368 the letter, the other case of this starting letter was not recognized
369 (#976).
370
371 12. If a pattern that was studied started with a repeated Unicode property
372 test, for example, \p{Nd}+, there was the theoretical possibility of
373 setting up an incorrect bitmap of starting bytes, but fortunately it could
374 not have actually happened in practice until change 8 above was made (it
375 added property types that matched character-matching opcodes).
376
377 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
378 possible starting bytes for non-anchored patterns.
379
380 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
381 \R, and also a number of cases that involve Unicode properties, both
382 explicit and implicit when PCRE_UCP is set.
383
384 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
385 input, it could crash or give wrong results if characters with values
386 greater than 0xc0 were present in the subject string. (Detail: it assumed
387 UTF-8 input when processing these items.)
388
389 16. Added a lot of (int) casts to avoid compiler warnings in systems where
390 size_t is 64-bit (#991).
391
392 17. Added a check for running out of memory when PCRE is compiled with
393 --disable-stack-for-recursion (#990).
394
395 18. If the last data line in a file for pcretest does not have a newline on
396 the end, a newline was missing in the output.
397
398 19. The default pcre_chartables.c file recognizes only ASCII characters (values
399 less than 128) in its various bitmaps. However, there is a facility for
400 generating tables according to the current locale when PCRE is compiled. It
401 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
402 characters, are recognized by isspace() and therefore were getting set in
403 these tables, and indeed these tables seem to approximate to ISO 8859. This
404 caused a problem in UTF-8 mode when pcre_study() was used to create a list
405 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
406 which of course cannot start UTF-8 characters. I have changed the code so
407 that only real ASCII characters (less than 128) and the correct starting
408 bytes for UTF-8 encodings are set for characters greater than 127 when in
409 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
410 altogether.)
411
412 20. Added the /T option to pcretest so as to be able to run tests with non-
413 standard character tables, thus making it possible to include the tests
414 used for 19 above in the standard set of tests.
415
416 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
417 reference to a subpattern the other side of a comment that contains an
418 opening parenthesis caused either an internal compiling error, or a
419 reference to the wrong subpattern.
420
421
422 Version 8.02 19-Mar-2010
423 ------------------------
424
425 1. The Unicode data tables have been updated to Unicode 5.2.0.
426
427 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
428 configured.
429
430 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
431 original author of that file, following a query about its status.
432
433 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
434 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
435
436 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
437 quantifier applied to a forward-referencing subroutine call, could compile
438 incorrect code or give the error "internal error: previously-checked
439 referenced subpattern not found".
440
441 6. Both MS Visual Studio and Symbian OS have problems with initializing
442 variables to point to external functions. For these systems, therefore,
443 pcre_malloc etc. are now initialized to local functions that call the
444 relevant global functions.
445
446 7. There were two entries missing in the vectors called coptable and poptable
447 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
448 I've fixed the data, and added a kludgy way of testing at compile time that
449 the lengths are correct (equal to the number of opcodes).
450
451 8. Following on from 7, I added a similar kludge to check the length of the
452 eint vector in pcreposix.c.
453
454 9. Error texts for pcre_compile() are held as one long string to avoid too
455 much relocation at load time. To find a text, the string is searched,
456 counting zeros. There was no check for running off the end of the string,
457 which could happen if a new error number was added without updating the
458 string.
459
460 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
461
462 11. \K was not working if it appeared in an atomic group or in a group that
463 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
464 \K is "not well defined" if used in an assertion. PCRE now accepts it if
465 the assertion is positive, but not if it is negative.
466
467 12. Change 11 fortuitously reduced the size of the stack frame used in the
468 "match()" function of pcre_exec.c by one pointer. Forthcoming
469 implementation of support for (*MARK) will need an extra pointer on the
470 stack; I have reserved it now, so that the stack frame size does not
471 decrease.
472
473 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
474 item in branch that calls a recursion is a subroutine call - as in the
475 second branch in the above example - was incorrectly given the compile-
476 time error "recursive call could loop indefinitely" because pcre_compile()
477 was not correctly checking the subroutine for matching a non-empty string.
478
479 14. The checks for overrunning compiling workspace could trigger after an
480 overrun had occurred. This is a "should never occur" error, but it can be
481 triggered by pathological patterns such as hundreds of nested parentheses.
482 The checks now trigger 100 bytes before the end of the workspace.
483
484 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
485
486
487 Version 8.01 19-Jan-2010
488 ------------------------
489
490 1. If a pattern contained a conditional subpattern with only one branch (in
491 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
492 computed the wrong minimum data length (which is of course zero for such
493 subpatterns). This could cause incorrect "no match" results.
494
495 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
496 the pattern is reset in the first branch, pcre_compile() failed with
497 "internal error: code overflow at offset...". This happened only when
498 the reset was to the original external option setting. (An optimization
499 abstracts leading options settings into an external setting, which was the
500 cause of this.)
501
502 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
503 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
504 assertion pattern did not match (meaning that the assertion was true), it
505 was incorrectly treated as false if the SKIP had been reached during the
506 matching. This also applied to assertions used as conditions.
507
508 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
509 assertion subpattern, including such a pattern used as a condition,
510 unpredictable results occurred, instead of the error return
511 PCRE_ERROR_DFA_UITEM.
512
513 5. The C++ GlobalReplace function was not working like Perl for the special
514 situation when an empty string is matched. It now does the fancy magic
515 stuff that is necessary.
516
517 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
518 removed. (These were left over from very, very early versions of PCRE.)
519
520 7. Some cosmetic changes to the code to make life easier when compiling it
521 as part of something else:
522
523 (a) Change DEBUG to PCRE_DEBUG.
524
525 (b) In pcre_compile(), rename the member of the "branch_chain" structure
526 called "current" as "current_branch", to prevent a collision with the
527 Linux macro when compiled as a kernel module.
528
529 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
530 prevent a collision with the Linux macro when compiled as a kernel
531 module.
532
533 8. In pcre_compile() there are some checks for integer overflows that used to
534 cast potentially large values to (double). This has been changed to that
535 when building, a check for int64_t is made, and if it is found, it is used
536 instead, thus avoiding the use of floating point arithmetic. (There is no
537 other use of FP in PCRE.) If int64_t is not found, the fallback is to
538 double.
539
540 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
541 2005 (difference between two addresses compared to an unsigned value).
542
543 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
544 custom one, because of the following reported problem in Windows:
545
546 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
547 under Win32.
548 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
549 therefore missing the function definition.
550 - The compiler thus generates a "C" signature for the test function.
551 - The linker fails to find the "C" function.
552 - PCRE fails to configure if asked to do so against libbz2.
553
554 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
555 messages were output:
556
557 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
558 rerunning libtoolize, to keep the correct libtool macros in-tree.
559 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
560
561 I have done both of these things.
562
563 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
564 most of the time, it *can* run out if it is given a pattern that contains a
565 runaway infinite recursion. I updated the discussion in the pcrestack man
566 page.
567
568 13. Now that we have gone to the x.xx style of version numbers, the minor
569 version may start with zero. Using 08 or 09 is a bad idea because users
570 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
571 interpreted as invalid octal numbers. I've updated the previous comment in
572 configure.ac, and also added a check that gives an error if 08 or 09 are
573 used.
574
575 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
576 causing partial matching to fail when the end of the subject matched \W
577 in a UTF-8 pattern where \W was quantified with a minimum of 3.
578
579 15. There were some discrepancies between the declarations in pcre_internal.h
580 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
581 their definitions. The declarations used "const uschar *" and the
582 definitions used USPTR. Even though USPTR is normally defined as "const
583 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
584 reported that: "This difference in casting confuses some C++ compilers, for
585 example, SunCC recognizes above declarations as different functions and
586 generates broken code for hbpcre." I have changed the declarations to use
587 USPTR.
588
589 16. GNU libtool is named differently on some systems. The autogen.sh script now
590 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
591 (FreeBSD).
592
593 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
594 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
595 comment: "Figure out how to create a longlong from a string: strtoll and
596 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
597 instance, but it only takes 2 args instead of 3!"
598
599 18. A subtle bug concerned with back references has been fixed by a change of
600 specification, with a corresponding code fix. A pattern such as
601 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
602 refers, was giving matches when it shouldn't. For example, xa=xaaa would
603 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
604 same bug. Such groups have to be quantified to be useful, or contained
605 inside another quantified group. (If there's no repetition, the reference
606 can never match.) The problem arises because, having left the group and
607 moved on to the rest of the pattern, a later failure that backtracks into
608 the group uses the captured value from the final iteration of the group
609 rather than the correct earlier one. I have fixed this in PCRE by forcing
610 any group that contains a reference to itself to be an atomic group; that
611 is, there cannot be any backtracking into it once it has completed. This is
612 similar to recursive and subroutine calls.
613
614
615 Version 8.00 19-Oct-09
616 ----------------------
617
618 1. The table for translating pcre_compile() error codes into POSIX error codes
619 was out-of-date, and there was no check on the pcre_compile() error code
620 being within the table. This could lead to an OK return being given in
621 error.
622
623 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
624 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
625 in a Windows environment.
626
627 3. The pcregrep --count option prints the count for each file even when it is
628 zero, as does GNU grep. However, pcregrep was also printing all files when
629 --files-with-matches was added. Now, when both options are given, it prints
630 counts only for those files that have at least one match. (GNU grep just
631 prints the file name in this circumstance, but including the count seems
632 more useful - otherwise, why use --count?) Also ensured that the
633 combination -clh just lists non-zero counts, with no names.
634
635 4. The long form of the pcregrep -F option was incorrectly implemented as
636 --fixed_strings instead of --fixed-strings. This is an incompatible change,
637 but it seems right to fix it, and I didn't think it was worth preserving
638 the old behaviour.
639
640 5. The command line items --regex=pattern and --regexp=pattern were not
641 recognized by pcregrep, which required --regex pattern or --regexp pattern
642 (with a space rather than an '='). The man page documented the '=' forms,
643 which are compatible with GNU grep; these now work.
644
645 6. No libpcreposix.pc file was created for pkg-config; there was just
646 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
647
648 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
649 when UCP support is not needed, by modifying the Python script that
650 generates it from Unicode data files. This should not matter if the module
651 is correctly used as a library, but I received one complaint about 50K of
652 unwanted data. My guess is that the person linked everything into his
653 program rather than using a library. Anyway, it does no harm.
654
655 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
656 was a minimum greater than 1 for a wide character in a possessive
657 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
658 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
659 character. Chaos in the form of incorrect output or a compiling loop could
660 result.
661
662 9. The restrictions on what a pattern can contain when partial matching is
663 requested for pcre_exec() have been removed. All patterns can now be
664 partially matched by this function. In addition, if there are at least two
665 slots in the offset vector, the offset of the earliest inspected character
666 for the match and the offset of the end of the subject are set in them when
667 PCRE_ERROR_PARTIAL is returned.
668
669 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
670 synonymous with PCRE_PARTIAL, for backwards compatibility, and
671 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
672 and may be more useful for multi-segment matching.
673
674 11. Partial matching with pcre_exec() is now more intuitive. A partial match
675 used to be given if ever the end of the subject was reached; now it is
676 given only if matching could not proceed because another character was
677 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
678 string "Z", which now yields "no match" instead of "partial match". In the
679 case of pcre_dfa_exec(), "no match" is given if every matching path for the
680 final character ended with (*FAIL).
681
682 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
683 if the pattern had a "must contain" character that was already found in the
684 earlier partial match, unless partial matching was again requested. For
685 example, with the pattern /dog.(body)?/, the "must contain" character is
686 "g". If the first part-match was for the string "dog", restarting with
687 "sbody" failed. This bug has been fixed.
688
689 13. The string returned by pcre_dfa_exec() after a partial match has been
690 changed so that it starts at the first inspected character rather than the
691 first character of the match. This makes a difference only if the pattern
692 starts with a lookbehind assertion or \b or \B (\K is not supported by
693 pcre_dfa_exec()). It's an incompatible change, but it makes the two
694 matching functions compatible, and I think it's the right thing to do.
695
696 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
697 so that the demonstration program is easily available in environments where
698 PCRE has not been installed from source.
699
700 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
701 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
702 library.
703
704 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
705 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
706 is not the first non-POSIX option to be added. Clearly some people find
707 these options useful.
708
709 17. If a caller to the POSIX matching function regexec() passes a non-zero
710 value for nmatch with a NULL value for pmatch, the value of
711 nmatch is forced to zero.
712
713 18. RunGrepTest did not have a test for the availability of the -u option of
714 the diff command, as RunTest does. It now checks in the same way as
715 RunTest, and also checks for the -b option.
716
717 19. If an odd number of negated classes containing just a single character
718 interposed, within parentheses, between a forward reference to a named
719 subpattern and the definition of the subpattern, compilation crashed with
720 an internal error, complaining that it could not find the referenced
721 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
722 [The bug was that it was starting one character too far in when skipping
723 over the character class, thus treating the ] as data rather than
724 terminating the class. This meant it could skip too much.]
725
726 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
727 /g option in pcretest when the pattern contains \K, which makes it possible
728 to have an empty string match not at the start, even when the pattern is
729 anchored. Updated pcretest and pcredemo to use this option.
730
731 21. If the maximum number of capturing subpatterns in a recursion was greater
732 than the maximum at the outer level, the higher number was returned, but
733 with unset values at the outer level. The correct (outer level) value is
734 now given.
735
736 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
737 PCRE did not set those parentheses (unlike Perl). I have now found a way to
738 make it do so. The string so far is captured, making this feature
739 compatible with Perl.
740
741 23. The tests have been re-organized, adding tests 11 and 12, to make it
742 possible to check the Perl 5.10 features against Perl 5.10.
743
744 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
745 pattern matches a fixed length string. PCRE did not allow this; now it
746 does. Neither allows recursion.
747
748 25. I finally figured out how to implement a request to provide the minimum
749 length of subject string that was needed in order to match a given pattern.
750 (It was back references and recursion that I had previously got hung up
751 on.) This code has now been added to pcre_study(); it finds a lower bound
752 to the length of subject needed. It is not necessarily the greatest lower
753 bound, but using it to avoid searching strings that are too short does give
754 some useful speed-ups. The value is available to calling programs via
755 pcre_fullinfo().
756
757 26. While implementing 25, I discovered to my embarrassment that pcretest had
758 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
759 study optimizations had never been tested with that matching function.
760 Oops. What is worse, even when it was passed study data, there was a bug in
761 pcre_dfa_exec() that meant it never actually used it. Double oops. There
762 were also very few tests of studied patterns with pcre_dfa_exec().
763
764 27. If (?| is used to create subpatterns with duplicate numbers, they are now
765 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
766 on the other side of the coin, they are no longer allowed to have different
767 names, because these cannot be distinguished in PCRE, and this has caused
768 confusion. (This is a difference from Perl.)
769
770 28. When duplicate subpattern names are present (necessarily with different
771 numbers, as required by 27 above), and a test is made by name in a
772 conditional pattern, either for a subpattern having been matched, or for
773 recursion in such a pattern, all the associated numbered subpatterns are
774 tested, and the overall condition is true if the condition is true for any
775 one of them. This is the way Perl works, and is also more like the way
776 testing by number works.
777
778
779 Version 7.9 11-Apr-09
780 ---------------------
781
782 1. When building with support for bzlib/zlib (pcregrep) and/or readline
783 (pcretest), all targets were linked against these libraries. This included
784 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
785 libraries. This caused unwanted dependencies to be created. This problem
786 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
787 pcretest is linked with readline.
788
789 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
790 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
791 moved outside it again, because FALSE and TRUE are already defined in AIX,
792 but BOOL is not.
793
794 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
795 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
796
797 4. The pcregrep documentation said spaces were inserted as well as colons (or
798 hyphens) following file names and line numbers when outputting matching
799 lines. This is not true; no spaces are inserted. I have also clarified the
800 wording for the --colour (or --color) option.
801
802 5. In pcregrep, when --colour was used with -o, the list of matching strings
803 was not coloured; this is different to GNU grep, so I have changed it to be
804 the same.
805
806 6. When --colo(u)r was used in pcregrep, only the first matching substring in
807 each matching line was coloured. Now it goes on to look for further matches
808 of any of the test patterns, which is the same behaviour as GNU grep.
809
810 7. A pattern that could match an empty string could cause pcregrep to loop; it
811 doesn't make sense to accept an empty string match in pcregrep, so I have
812 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
813 seems to be how GNU grep behaves.
814
815 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
816 start or after a newline", because the conditional assertion was not being
817 correctly handled. The rule now is that both the assertion and what follows
818 in the first alternative must satisfy the test.
819
820 9. If auto-callout was enabled in a pattern with a conditional group whose
821 condition was an assertion, PCRE could crash during matching, both with
822 pcre_exec() and pcre_dfa_exec().
823
824 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
825 used for matching.
826
827 11. Unicode property support in character classes was not working for
828 characters (bytes) greater than 127 when not in UTF-8 mode.
829
830 12. Added the -M command line option to pcretest.
831
832 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
833
834 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
835
836 16. Added comments and documentation about mis-use of no_arg in the C++
837 wrapper.
838
839 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
840 from Martin Jerabek that uses macro names for all relevant character and
841 string constants.
842
843 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
844 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
845 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
846 these, but not everybody uses configure.
847
848 19. A conditional group that had only one branch was not being correctly
849 recognized as an item that could match an empty string. This meant that an
850 enclosing group might also not be so recognized, causing infinite looping
851 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
852 with the subject "ab", where knowledge that the repeated group can match
853 nothing is needed in order to break the loop.
854
855 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
856 exec(), but without supplying a callout function, matching went wrong.
857
858 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
859 leak if the size of the offset vector was greater than 30. When the vector
860 is smaller, the saved offsets during recursion go onto a local stack
861 vector, but for larger vectors malloc() is used. It was failing to free
862 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
863 error, in fact).
864
865 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
866 heapframe that is used only when UTF-8 support is enabled. This caused no
867 problem, but was untidy.
868
869 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
870 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
871 included within another project.
872
873 24. Steven Van Ingelgem's patches to add more options to the CMake support,
874 slightly modified by me:
875
876 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
877 not building pcregrep.
878
879 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
880 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
881
882 25. Forward references, both numeric and by name, in patterns that made use of
883 duplicate group numbers, could behave incorrectly or give incorrect errors,
884 because when scanning forward to find the reference group, PCRE was not
885 taking into account the duplicate group numbers. A pattern such as
886 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
887
888 26. Changed a few more instances of "const unsigned char *" to USPTR, making
889 the feature of a custom pointer more persuasive (as requested by a user).
890
891 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
892 pcretest.c, inside #ifndefs, because it seems they are sometimes already
893 pre-defined.
894
895 28. Added support for (*UTF8) at the start of a pattern.
896
897 29. Arrange for flags added by the "release type" setting in CMake to be shown
898 in the configuration summary.
899
900
901 Version 7.8 05-Sep-08
902 ---------------------
903
904 1. Replaced UCP searching code with optimized version as implemented for Ad
905 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
906 stage table and inline lookup instead of a function, giving speed ups of 2
907 to 5 times on some simple patterns that I tested. Permission was given to
908 distribute the MultiStage2.py script that generates the tables (it's not in
909 the tarball, but is in the Subversion repository).
910
911 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
912 scripts.
913
914 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
915 a group with a zero qualifier. The result of the study could be incorrect,
916 or the function might crash, depending on the pattern.
917
918 4. Caseless matching was not working for non-ASCII characters in back
919 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
920 It now works when Unicode Property Support is available.
921
922 5. In pcretest, an escape such as \x{de} in the data was always generating
923 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
924 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
925 truncation.
926
927 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
928
929 7. Added two (int) casts to pcregrep when printing the difference of two
930 pointers, in case they are 64-bit values.
931
932 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
933 test 2 if it fails.
934
935 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
936 and a #define of that name to empty if it is not externally set. This is to
937 allow users of MSVC to set it if necessary.
938
939 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
940 the convenience functions in the pcre_get.c source file.
941
942 11. An option change at the start of a pattern that had top-level alternatives
943 could cause overwriting and/or a crash. This command provoked a crash in
944 some environments:
945
946 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
947
948 This potential security problem was recorded as CVE-2008-2371.
949
950 12. For a pattern where the match had to start at the beginning or immediately
951 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
952 pcre_dfa_exec() could read past the end of the passed subject if there was
953 no match. To help with detecting such bugs (e.g. with valgrind), I modified
954 pcretest so that it places the subject at the end of its malloc-ed buffer.
955
956 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
957 exec() might read past the end of the data buffer in UTF-8 mode.
958
959 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
960 the data contained the byte 0x85 as part of a UTF-8 character within its
961 first line. This applied both to normal and DFA matching.
962
963 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
964 /^[^d]*?$/8 failed to match "abc".
965
966 16. Added a missing copyright notice to pcrecpp_internal.h.
967
968 17. Make it more clear in the documentation that values returned from
969 pcre_exec() in ovector are byte offsets, not character counts.
970
971 18. Tidied a few places to stop certain compilers from issuing warnings.
972
973 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
974 supplied by Stefan Weber. I made a further small update for 7.8 because
975 there is a change of source arrangements: the pcre_searchfuncs.c module is
976 replaced by pcre_ucd.c.
977
978
979 Version 7.7 07-May-08
980 ---------------------
981
982 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
983 a string to a long long, pretend we don't even have a long long." This is
984 done by checking for the strtoq, strtoll, and _strtoi64 functions.
985
986 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
987 pre-7.6 versions, which defined a global no_arg variable instead of putting
988 it in the RE class. (See also #8 below.)
989
990 3. Remove a line of dead code, identified by coverity and reported by Nuno
991 Lopes.
992
993 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
994
995 (1) The include/exclude patterns were being applied to the whole pathnames
996 of files, instead of just to the final components.
997
998 (2) If there was more than one level of directory, the subdirectories were
999 skipped unless they satisfied the include/exclude conditions. This is
1000 inconsistent with GNU grep (and could even be seen as contrary to the
1001 pcregrep specification - which I improved to make it absolutely clear).
1002 The action now is always to scan all levels of directory, and just
1003 apply the include/exclude patterns to regular files.
1004
1005 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1006 --exclude_dir in the tests to avoid scanning .svn directories.
1007
1008 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1009 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1010 doesn't support NULs in patterns.
1011
1012 7. Added some missing "const"s to declarations of static tables in
1013 pcre_compile.c and pcre_dfa_exec.c.
1014
1015 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1016 caused by fix #2 above. (Subsequently also a second patch to fix the
1017 first patch. And a third patch - this was a messy problem.)
1018
1019 9. Applied Craig's patch to remove the use of push_back().
1020
1021 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1022 matching function regexec().
1023
1024 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1025 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1026 references. PCRE supports relative numbers with this syntax (I don't think
1027 Oniguruma does).
1028
1029 12. Previously, a group with a zero repeat such as (...){0} was completely
1030 omitted from the compiled regex. However, this means that if the group
1031 was called as a subroutine from elsewhere in the pattern, things went wrong
1032 (an internal error was given). Such groups are now left in the compiled
1033 pattern, with a new opcode that causes them to be skipped at execution
1034 time.
1035
1036 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1037 to the way PCRE behaves:
1038
1039 (a) A lone ] character is dis-allowed (Perl treats it as data).
1040
1041 (b) A back reference to an unmatched subpattern matches an empty string
1042 (Perl fails the current match path).
1043
1044 (c) A data ] in a character class must be notated as \] because if the
1045 first data character in a class is ], it defines an empty class. (In
1046 Perl it is not possible to have an empty class.) The empty class []
1047 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1048 The negative empty class [^] matches any one character, independently
1049 of the DOTALL setting.
1050
1051 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1052 non-existent subpattern following a character class starting with ']' and
1053 containing () gave an internal compiling error instead of "reference to
1054 non-existent subpattern". Fortunately, when the pattern did exist, the
1055 compiled code was correct. (When scanning forwards to check for the
1056 existencd of the subpattern, it was treating the data ']' as terminating
1057 the class, so got the count wrong. When actually compiling, the reference
1058 was subsequently set up correctly.)
1059
1060 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1061 it was being rejected as not supported by pcre_dfa_exec(), even though
1062 other assertions are supported. I have made pcre_dfa_exec() support
1063 (*FAIL).
1064
1065 16. The implementation of 13c above involved the invention of a new opcode,
1066 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1067 cannot be changed at match time, I realized I could make a small
1068 improvement to matching performance by compiling OP_ALLANY instead of
1069 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1070 on the OP_ANY path.
1071
1072 17. Compiling pcretest on Windows with readline support failed without the
1073 following two fixes: (1) Make the unistd.h include conditional on
1074 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1075
1076 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1077 ncurses library to be included for pcretest when ReadLine support is
1078 requested, but also to allow for it to be overridden. This patch came from
1079 Daniel Bergström.
1080
1081 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1082 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1083 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1084 spotting this.
1085
1086
1087 Version 7.6 28-Jan-08
1088 ---------------------
1089
1090 1. A character class containing a very large number of characters with
1091 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1092 overflow.
1093
1094 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1095 HAVE_LONG_LONG is not defined.
1096
1097 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1098 bring them up to date and include new features. This patch includes:
1099
1100 - Fixed PH's badly added libz and libbz2 support.
1101 - Fixed a problem with static linking.
1102 - Added pcredemo. [But later removed - see 7 below.]
1103 - Fixed dftables problem and added an option.
1104 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1105 HAVE_LONG_LONG.
1106 - Added readline support for pcretest.
1107 - Added an listing of the option settings after cmake has run.
1108
1109 4. A user submitted a patch to Makefile that makes it easy to create
1110 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1111 Makefile.am that cause it to include this special target, without
1112 affecting anything else. Note that the same mingw target plus all
1113 the other distribution libraries and programs are now supported
1114 when configuring with CMake (see 6 below) instead of with
1115 Configure/Make.
1116
1117 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1118 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1119 exported in the Windows port". It has not yet been confirmed that the patch
1120 solves the problem, but it does no harm.
1121
1122 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1123 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1124 with CMake, and also correct the comment about stack recursion.
1125
1126 7. Remove the automatic building of pcredemo from the ./configure system and
1127 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1128 of a program that users should build themselves after PCRE is installed, so
1129 building it automatically is not really right. What is more, it gave
1130 trouble in some build environments.
1131
1132 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1133
1134
1135 Version 7.5 10-Jan-08
1136 ---------------------
1137
1138 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1139 values in parens when parsing an RE using the C++ wrapper."
1140
1141 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1142 Characters greater than 255 were excluded from the class instead of being
1143 included.
1144
1145 3. The same bug as (2) above applied to negated POSIX classes such as
1146 [:^space:].
1147
1148 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1149 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1150 I have changed it.
1151
1152 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1153 first named subpattern) and a construct such as (?&a) would reference the
1154 first named subpattern whose name started with "a" (in other words, the
1155 length check was missing). Both these problems are fixed. "Subpattern name
1156 expected" is now given for (?&) (a zero-length name), and this patch also
1157 makes it give the same error for \k'' (previously it complained that that
1158 was a reference to a non-existent subpattern).
1159
1160 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1161 this is right because (?- can be followed by option settings as well as by
1162 digits. I have, however, made the messages clearer.
1163
1164 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1165 than the number used in the conditional) now cause a compile-time error.
1166 This is actually not compatible with Perl, which accepts such patterns, but
1167 treats the conditional as always being FALSE (as PCRE used to), but it
1168 seems to me that giving a diagnostic is better.
1169
1170 8. Change "alphameric" to the more common word "alphanumeric" in comments
1171 and messages.
1172
1173 9. Fix two occurrences of "backslash" in comments that should have been
1174 "backspace".
1175
1176 10. Remove two redundant lines of code that can never be obeyed (their function
1177 was moved elsewhere).
1178
1179 11. The program that makes PCRE's Unicode character property table had a bug
1180 which caused it to generate incorrect table entries for sequences of
1181 characters that have the same character type, but are in different scripts.
1182 It amalgamated them into a single range, with the script of the first of
1183 them. In other words, some characters were in the wrong script. There were
1184 thirteen such cases, affecting characters in the following ranges:
1185
1186 U+002b0 - U+002c1
1187 U+0060c - U+0060d
1188 U+0061e - U+00612
1189 U+0064b - U+0065e
1190 U+0074d - U+0076d
1191 U+01800 - U+01805
1192 U+01d00 - U+01d77
1193 U+01d9b - U+01dbf
1194 U+0200b - U+0200f
1195 U+030fc - U+030fe
1196 U+03260 - U+0327f
1197 U+0fb46 - U+0fbb1
1198 U+10450 - U+1049d
1199
1200 12. The -o option (show only the matching part of a line) for pcregrep was not
1201 compatible with GNU grep in that, if there was more than one match in a
1202 line, it showed only the first of them. It now behaves in the same way as
1203 GNU grep.
1204
1205 13. If the -o and -v options were combined for pcregrep, it printed a blank
1206 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1207 does the same. The return code can be used to tell if there were any
1208 non-matching lines.
1209
1210 14. Added --file-offsets and --line-offsets to pcregrep.
1211
1212 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1213 infinitely looping recursion. The bug was that positive lookaheads were not
1214 being skipped when checking for a possible empty match (negative lookaheads
1215 and both kinds of lookbehind were skipped).
1216
1217 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1218 inclusion of <windows.h> to before rather than after the definition of
1219 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1220
1221 17. Specifying a possessive quantifier with a specific limit for a Unicode
1222 character property caused pcre_compile() to compile bad code, which led at
1223 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1224 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1225 caused the error; without that there was no problem.
1226
1227 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1228
1229 19. Added --enable-pcretest-libreadline.
1230
1231 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1232 RE::GlobalReplace(). As a result, the number of replacements returned was
1233 double what it should be. I removed one of the increments, but Craig sent a
1234 later patch that removed the other one (the right fix) and added unit tests
1235 that check the return values (which was not done before).
1236
1237 21. Several CMake things:
1238
1239 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1240 the names libpcre and libpcreposix, not just pcre and pcreposix.
1241
1242 (2) The above change means that pcretest and pcregrep are now correctly
1243 linked with the newly-built libraries, not previously installed ones.
1244
1245 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1246
1247 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1248 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1249 UTF-8 newline character). The key issue is that the pattern starts .*;
1250 this means that the match must be either at the beginning, or after a
1251 newline. The bug was in the code for advancing after a failed match and
1252 checking that the new position followed a newline. It was not taking
1253 account of UTF-8 characters correctly.
1254
1255 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1256 character classes. PCRE was not treating the sequence [:...:] as a
1257 character class unless the ... were all letters. Perl, however, seems to
1258 allow any characters between [: and :], though of course it rejects as
1259 unknown any "names" that contain non-letters, because all the known class
1260 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1261 for example, whereas PCRE did not - it did not recognize a POSIX character
1262 class. This seemed a bit dangerous, so the code has been changed to be
1263 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1264 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1265 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1266 Perl does, and where it didn't before.
1267
1268 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1269 Windows environments %n is disabled by default.
1270
1271
1272 Version 7.4 21-Sep-07
1273 ---------------------
1274
1275 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1276 means that a class such as [\s] counted as "explicit reference to CR or
1277 LF". That isn't really right - the whole point of the change was to try to
1278 help when there was an actual mention of one of the two characters. So now
1279 the change happens only if \r or \n (or a literal CR or LF) character is
1280 encountered.
1281
1282 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1283 of both had grown to the point where there were only 3 bits left.
1284 Fortunately, there was spare space in the data structure, and so I have
1285 moved the internal flags into a new 16-bit field to free up more option
1286 bits.
1287
1288 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1289 but did not set the internal JCHANGED flag - either of these is enough to
1290 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1291 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1292 start sets both bits.
1293
1294 4. Added options (at build time, compile time, exec time) to change \R from
1295 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1296
1297 5. doc/pcresyntax.html was missing from the distribution.
1298
1299 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1300 compatibility, even though it is no longer used.
1301
1302 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1303 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1304 windows.h file is present (where different names are used). [This was
1305 reversed later after testing - see 16 below.]
1306
1307 8. Changed all #include <config.h> to #include "config.h". There were also
1308 some further <pcre.h> cases that I changed to "pcre.h".
1309
1310 9. When pcregrep was used with the --colour option, it missed the line ending
1311 sequence off the lines that it output.
1312
1313 10. It was pointed out to me that arrays of string pointers cause lots of
1314 relocations when a shared library is dynamically loaded. A technique of
1315 using a single long string with a table of offsets can drastically reduce
1316 these. I have refactored PCRE in four places to do this. The result is
1317 dramatic:
1318
1319 Originally: 290
1320 After changing UCP table: 187
1321 After changing error message table: 43
1322 After changing table of "verbs" 36
1323 After changing table of Posix names 22
1324
1325 Thanks to the folks working on Gregex for glib for this insight.
1326
1327 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1328 unicode-properties was also set.
1329
1330 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1331
1332 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1333 checked only for CRLF.
1334
1335 14. Added casts to pcretest.c to avoid compiler warnings.
1336
1337 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1338
1339 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1340 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1341 entirely. This removes changes made in 7 above.
1342
1343 17. The CMake files have been updated, and there is now more information about
1344 building with CMake in the NON-UNIX-USE document.
1345
1346
1347 Version 7.3 28-Aug-07
1348 ---------------------
1349
1350 1. In the rejigging of the build system that eventually resulted in 7.1, the
1351 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1352 brackets there is not right, since it causes compilers to look for an
1353 installed pcre.h, not the version that is in the source that is being
1354 compiled (which of course may be different). I have changed it back to:
1355
1356 #include "pcre.h"
1357
1358 I have a vague recollection that the change was concerned with compiling in
1359 different directories, but in the new build system, that is taken care of
1360 by the VPATH setting the Makefile.
1361
1362 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1363 when the subject happened to end in the byte 0x85 (e.g. if the last
1364 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1365 characters but of course it shouldn't be taken as a newline when it is part
1366 of another character. The bug was that, for an unlimited repeat of . in
1367 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1368 characters when looking for a newline.
1369
1370 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1371
1372 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1373 in debug output.
1374
1375 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1376 long printing in the pcrecpp unittest when running under MinGW.
1377
1378 6. ESC_K was left out of the EBCDIC table.
1379
1380 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1381 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1382 limit also applies to "virtual nesting" when a pattern is recursive, and in
1383 this case 1000 isn't so big. I have been able to remove this limit at the
1384 expense of backing off one optimization in certain circumstances. Normally,
1385 when pcre_exec() would call its internal match() function recursively and
1386 immediately return the result unconditionally, it uses a "tail recursion"
1387 feature to save stack. However, when a subpattern that can match an empty
1388 string has an unlimited repetition quantifier, it no longer makes this
1389 optimization. That gives it a stack frame in which to save the data for
1390 checking that an empty string has been matched. Previously this was taken
1391 from the 1000-entry workspace that had been reserved. So now there is no
1392 explicit limit, but more stack is used.
1393
1394 8. Applied Daniel's patches to solve problems with the import/export magic
1395 syntax that is required for Windows, and which was going wrong for the
1396 pcreposix and pcrecpp parts of the library. These were overlooked when this
1397 problem was solved for the main library.
1398
1399 9. There were some crude static tests to avoid integer overflow when computing
1400 the size of patterns that contain repeated groups with explicit upper
1401 limits. As the maximum quantifier is 65535, the maximum group length was
1402 set at 30,000 so that the product of these two numbers did not overflow a
1403 32-bit integer. However, it turns out that people want to use groups that
1404 are longer than 30,000 bytes (though not repeat them that many times).
1405 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1406 made it possible to implement the integer overflow checks in a much more
1407 dynamic way, which I have now done. The artificial limitation on group
1408 length has been removed - we now have only the limit on the total length of
1409 the compiled pattern, which depends on the LINK_SIZE setting.
1410
1411 10. Fixed a bug in the documentation for get/copy named substring when
1412 duplicate names are permitted. If none of the named substrings are set, the
1413 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1414 empty string.
1415
1416 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1417 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1418 because the ] is interpreted as the first data character and the
1419 terminating ] is not found. PCRE has been made compatible with Perl in this
1420 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1421 cause memory overwriting.
1422
1423 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1424 string has been matched (to stop an infinite loop). It was not recognizing
1425 a conditional subpattern that could match an empty string if that
1426 subpattern was within another subpattern. For example, it looped when
1427 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1428 condition was not nested. This bug has been fixed.
1429
1430 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1431 past the start of the subject in the presence of bytes with the top bit
1432 set, for example "\x8aBCD".
1433
1434 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1435 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1436
1437 14. Optimized (?!) to (*FAIL).
1438
1439 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1440 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1441 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1442 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1443 does: it's just the validity check that is more restrictive.
1444
1445 16. Inserted checks for integer overflows during escape sequence (backslash)
1446 processing, and also fixed erroneous offset values for syntax errors during
1447 backslash processing.
1448
1449 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1450 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1451
1452 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1453 caused an overrun.
1454
1455 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1456 something other than just ASCII characters) inside a group that had an
1457 unlimited repeat caused a loop at compile time (while checking to see
1458 whether the group could match an empty string).
1459
1460 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1461 [\P{Any}] did so. (Error in the code for printing property names.)
1462
1463 21. An orphan \E inside a character class could cause a crash.
1464
1465 22. A repeated capturing bracket such as (A)? could cause a wild memory
1466 reference during compilation.
1467
1468 23. There are several functions in pcre_compile() that scan along a compiled
1469 expression for various reasons (e.g. to see if it's fixed length for look
1470 behind). There were bugs in these functions when a repeated \p or \P was
1471 present in the pattern. These operators have additional parameters compared
1472 with \d, etc, and these were not being taken into account when moving along
1473 the compiled data. Specifically:
1474
1475 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1476 length.
1477
1478 (b) An item such as \pL+ within a repeated group could cause crashes or
1479 loops.
1480
1481 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1482 "reference to non-existent subpattern" error.
1483
1484 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1485
1486 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1487 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1488
1489 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1490
1491 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1492 character were causing crashes (broken optimization).
1493
1494 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1495 \p or \P) caused a compile-time loop.
1496
1497 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1498 break. For example, the unstudied pattern [\r\n]A does not match the string
1499 "\r\nA" because change 7.0/46 below moves the current point on by two
1500 characters after failing to match at the start. However, the pattern \nA
1501 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1502 the same is true. There doesn't seem any very clean way out of this, but
1503 what I have chosen to do makes the common cases work: PCRE now takes note
1504 of whether there can be an explicit match for \r or \n anywhere in the
1505 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1506 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1507 pattern has explicit CR or LF references.
1508
1509 29. Added (*CR) etc for changing newline setting at start of pattern.
1510
1511
1512 Version 7.2 19-Jun-07
1513 ---------------------
1514
1515 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1516 which is apparently normally available under Windows.
1517
1518 2. Re-jig the pcregrep tests with different newline settings in an attempt
1519 to make them independent of the local environment's newline setting.
1520
1521 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1522
1523 4. Some of the "internals" tests were previously cut out when the link size
1524 was not 2, because the output contained actual offsets. The recent new
1525 "Z" feature of pcretest means that these can be cut out, making the tests
1526 usable with all link sizes.
1527
1528 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1529 stack recursion. This gives a massive performance boost under BSD, but just
1530 a small improvement under Linux. However, it saves one field in the frame
1531 in all cases.
1532
1533 6. Added more features from the forthcoming Perl 5.10:
1534
1535 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1536 recursion call. It refers to the nth most recently opened parentheses.
1537
1538 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1539 to be opened parentheses.
1540
1541 (c) Conditions that refer to capturing parentheses can be specified
1542 relatively, for example, (?(-2)... or (?(+3)...
1543
1544 (d) \K resets the start of the current match so that everything before
1545 is not part of it.
1546
1547 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1548
1549 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1550 reference syntax.
1551
1552 (g) (?| introduces a group in which the numbering of parentheses in each
1553 alternative starts with the same number.
1554
1555 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1556
1557 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1558 PCRE_INFO_JCHANGED.
1559
1560 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1561 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1562 for detecting groups that can match an empty string.
1563
1564 9. A pattern with a very large number of alternatives (more than several
1565 hundred) was running out of internal workspace during the pre-compile
1566 phase, where pcre_compile() figures out how much memory will be needed. A
1567 bit of new cunning has reduced the workspace needed for groups with
1568 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1569 workspace instead of running out of the 4096 that are available.
1570
1571 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1572
1573 11. Applied patch from Google to remove an optimization that didn't quite work.
1574 The report of the bug said:
1575
1576 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1577 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1578 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1579
1580 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1581 it matched the wrong number of bytes.
1582
1583
1584 Version 7.1 24-Apr-07
1585 ---------------------
1586
1587 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1588 that is more "standard", making use of automake and other Autotools. There
1589 is some re-arrangement of the files and adjustment of comments consequent
1590 on this.
1591
1592 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1593 for recursive directory scanning broke on some systems because the files
1594 are not scanned in any specific order and on different systems the order
1595 was different. A call to "sort" has been inserted into RunGrepTest for the
1596 approprate test as a short-term fix. In the longer term there may be an
1597 alternative.
1598
1599 3. I had an email from Eric Raymond about problems translating some of PCRE's
1600 man pages to HTML (despite the fact that I distribute HTML pages, some
1601 people do their own conversions for various reasons). The problems
1602 concerned the use of low-level troff macros .br and .in. I have therefore
1603 removed all such uses from the man pages (some were redundant, some could
1604 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1605 HTML has been updated to handle .nf/.fi and to complain if it encounters
1606 .br or .in.
1607
1608 4. Updated comments in configure.ac that get placed in config.h.in and also
1609 arranged for config.h to be included in the distribution, with the name
1610 config.h.generic, for the benefit of those who have to compile without
1611 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1612
1613 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1614 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1615 makevp.bat for the current PCRE, using the additional files
1616 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1617
1618 6. A Windows user reported a minor discrepancy with test 2, which turned out
1619 to be caused by a trailing space on an input line that had got lost in his
1620 copy. The trailing space was an accident, so I've just removed it.
1621
1622 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1623 that is needed.
1624
1625 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1626 as "const" (a) because they are and (b) because it helps the PHP
1627 maintainers who have recently made a script to detect big data structures
1628 in the php code that should be moved to the .rodata section. I remembered
1629 to update Builducptable as well, so it won't revert if ucptable.h is ever
1630 re-created.
1631
1632 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1633 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1634 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1635 support is not required. This saves 1.5-2K of code, which is important in
1636 some applications.
1637
1638 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1639 so as not to refer to the tables, even though these functions will never be
1640 called when UTF-8 support is disabled. Otherwise there are problems with a
1641 shared library.
1642
1643 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1644
1645 (a) It was defining its arguments as char * instead of void *.
1646
1647 (b) It was assuming that all moves were upwards in memory; this was true
1648 a long time ago when I wrote it, but is no longer the case.
1649
1650 The emulated memove() is provided for those environments that have neither
1651 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1652 is clearly not the case, as these two bugs were recently reported.
1653
1654 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1655 and Detrail to create the HTML documentation, the .txt form of the man
1656 pages, and it removes trailing spaces from listed files. It also creates
1657 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1658 case, it wraps all the #defines with #ifndefs. This script should be run
1659 before "make dist".
1660
1661 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1662 with Unicode property support.
1663
1664 (a) For a maximizing quantifier, if the two different cases of the
1665 character were of different lengths in their UTF-8 codings (there are
1666 some cases like this - I found 11), and the matching function had to
1667 back up over a mixture of the two cases, it incorrectly assumed they
1668 were both the same length.
1669
1670 (b) When PCRE was configured to use the heap rather than the stack for
1671 recursion during matching, it was not correctly preserving the data for
1672 the other case of a UTF-8 character when checking ahead for a match
1673 while processing a minimizing repeat. If the check also involved
1674 matching a wide character, but failed, corruption could cause an
1675 erroneous result when trying to check for a repeat of the original
1676 character.
1677
1678 13. Some tidying changes to the testing mechanism:
1679
1680 (a) The RunTest script now detects the internal link size and whether there
1681 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1682 values substituted by "configure". (The RunGrepTest script already did
1683 this for UTF-8.) The configure.ac script no longer substitutes the
1684 relevant variables.
1685
1686 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1687 with length and offset values. This means that the output is different
1688 for different internal link sizes. Test 2 is skipped for link sizes
1689 other than 2 because of this, bypassing the problem. Unfortunately,
1690 there was also a test in test 3 (the locale tests) that used /B and
1691 failed for link sizes other than 2. Rather than cut the whole test out,
1692 I have added a new /Z option to pcretest that replaces the length and
1693 offset values with spaces. This is now used to make test 3 independent
1694 of link size. (Test 2 will be tidied up later.)
1695
1696 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1697 segmentation fault instead of returning the appropriate error message.
1698
1699 15. In multiline mode when the newline sequence was set to "any", the pattern
1700 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1701 This doesn't seem right; it now treats the CRLF combination as the line
1702 ending, and so does not match in that case. It's only a pattern such as ^$
1703 that would hit this one: something like ^ABC$ would have failed after \r
1704 and then tried again after \r\n.
1705
1706 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1707 in an attempt to make files that differ only in their line terminators
1708 compare equal. This works on Linux.
1709
1710 17. Under certain error circumstances pcregrep might try to free random memory
1711 as it exited. This is now fixed, thanks to valgrind.
1712
1713 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1714 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1715 was because its rules for how to advance for /g after matching an empty
1716 string at the end of a line did not allow for this case. They now check for
1717 it specially.
1718
1719 20. pcretest is supposed to handle patterns and data of any length, by
1720 extending its buffers when necessary. It was getting this wrong when the
1721 buffer for a data line had to be extended.
1722
1723 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1724 CRLF as a newline sequence.
1725
1726 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1727 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1728 I have nevertheless tidied it up.
1729
1730 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1731
1732 24. Added a man page for pcre-config.
1733
1734
1735 Version 7.0 19-Dec-06
1736 ---------------------
1737
1738 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1739 moving to gcc 4.1.1.
1740
1741 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1742 sys/time.h, which is documented as needed for this function. It doesn't
1743 seem to matter on Linux, but it showed up on some releases of OS X.
1744
1745 3. It seems that there are systems where bytes whose values are greater than
1746 127 match isprint() in the "C" locale. The "C" locale should be the
1747 default when a C program starts up. In most systems, only ASCII printing
1748 characters match isprint(). This difference caused the output from pcretest
1749 to vary, making some of the tests fail. I have changed pcretest so that:
1750
1751 (a) When it is outputting text in the compiled version of a pattern, bytes
1752 other than 32-126 are always shown as hex escapes.
1753
1754 (b) When it is outputting text that is a matched part of a subject string,
1755 it does the same, unless a different locale has been set for the match
1756 (using the /L modifier). In this case, it uses isprint() to decide.
1757
1758 4. Fixed a major bug that caused incorrect computation of the amount of memory
1759 required for a compiled pattern when options that changed within the
1760 pattern affected the logic of the preliminary scan that determines the
1761 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1762 that the computed length was too small. The symptoms of this bug were
1763 either the PCRE error "internal error: code overflow" from pcre_compile(),
1764 or a glibc crash with a message such as "pcretest: free(): invalid next
1765 size (fast)". Examples of patterns that provoked this bug (shown in
1766 pcretest format) are:
1767
1768 /(?-x: )/x
1769 /(?x)(?-x: \s*#\s*)/
1770 /((?i)[\x{c0}])/8
1771 /(?i:[\x{c0}])/8
1772
1773 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1774 is now done differently.
1775
1776 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1777 wrapper classes; (b) implement a new function in the C++ scanner that is
1778 more efficient than the old way of doing things because it avoids levels of
1779 recursion in the regex matching; (c) add a paragraph to the documentation
1780 for the FullMatch() function.
1781
1782 6. The escape sequence \n was being treated as whatever was defined as
1783 "newline". Not only was this contrary to the documentation, which states
1784 that \n is character 10 (hex 0A), but it also went horribly wrong when
1785 "newline" was defined as CRLF. This has been fixed.
1786
1787 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1788 was being set to -1 for the "end of line" case (supposedly a value that no
1789 character can have). Though this value is never used (the check for end of
1790 line is "zero bytes in current character"), it caused compiler complaints.
1791 I've changed it to 0xffffffff.
1792
1793 8. In pcre_version.c, the version string was being built by a sequence of
1794 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1795 string (as it is for production releases) called a macro with an empty
1796 argument. The C standard says the result of this is undefined. The gcc
1797 compiler treats it as an empty string (which was what was wanted) but it is
1798 reported that Visual C gives an error. The source has been hacked around to
1799 avoid this problem.
1800
1801 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1802 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1803 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1804 of them did).
1805
1806 10. Originally, pcretest opened its input and output without "b"; then I was
1807 told that "b" was needed in some environments, so it was added for release
1808 5.0 to both the input and output. (It makes no difference on Unix-like
1809 systems.) Later I was told that it is wrong for the input on Windows. I've
1810 now abstracted the modes into two macros, to make it easier to fiddle with
1811 them, and removed "b" from the input mode under Windows.
1812
1813 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1814
1815 12. Added -help and --help to pcretest as an official way of being reminded
1816 of the options.
1817
1818 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1819 and pcrecpp.cc because they annoy compilers at high warning levels.
1820
1821 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1822
1823 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1824 scripts are not C programs :-) and which was not noticed because it works
1825 on Linux.
1826
1827 16. pcretest is supposed to handle any length of pattern and data line (as one
1828 line or as a continued sequence of lines) by extending its input buffer if
1829 necessary. This feature was broken for very long pattern lines, leading to
1830 a string of junk being passed to pcre_compile() if the pattern was longer
1831 than about 50K.
1832
1833 17. I have done a major re-factoring of the way pcre_compile() computes the
1834 amount of memory needed for a compiled pattern. Previously, there was code
1835 that made a preliminary scan of the pattern in order to do this. That was
1836 OK when PCRE was new, but as the facilities have expanded, it has become
1837 harder and harder to keep it in step with the real compile phase, and there
1838 have been a number of bugs (see for example, 4 above). I have now found a
1839 cunning way of running the real compile function in a "fake" mode that
1840 enables it to compute how much memory it would need, while actually only
1841 ever using a few hundred bytes of working memory and without too many
1842 tests of the mode. This should make future maintenance and development
1843 easier. A side effect of this work is that the limit of 200 on the nesting
1844 depth of parentheses has been removed (though this was never a serious
1845 limitation, I suspect). However, there is a downside: pcre_compile() now
1846 runs more slowly than before (30% or more, depending on the pattern). I
1847 hope this isn't a big issue. There is no effect on runtime performance.
1848
1849 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1850 newline (only possible for the last line of a file) and it was a
1851 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1852
1853 19. Added additional timing features to pcretest. (1) The -tm option now times
1854 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1855 separate command line item, by a number that specifies the number of
1856 repeats to use when timing. The default is 50000; this gives better
1857 precision, but takes uncomfortably long for very large patterns.
1858
1859 20. Extended pcre_study() to be more clever in cases where a branch of a
1860 subpattern has no definite first character. For example, (a*|b*)[cd] would
1861 previously give no result from pcre_study(). Now it recognizes that the
1862 first character must be a, b, c, or d.
1863
1864 21. There was an incorrect error "recursive call could loop indefinitely" if
1865 a subpattern (or the entire pattern) that was being tested for matching an
1866 empty string contained only one non-empty item after a nested subpattern.
1867 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1868 incorrectly, because the \d was being skipped in the check.
1869
1870 22. The pcretest program now has a new pattern option /B and a command line
1871 option -b, which is equivalent to adding /B to every pattern. This causes
1872 it to show the compiled bytecode, without the additional information that
1873 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1874 is the same as /B/I).
1875
1876 23. A new optimization is now able automatically to treat some sequences such
1877 as a*b as a*+b. More specifically, if something simple (such as a character
1878 or a simple class like \d) has an unlimited quantifier, and is followed by
1879 something that cannot possibly match the quantified thing, the quantifier
1880 is automatically "possessified".
1881
1882 24. A recursive reference to a subpattern whose number was greater than 39
1883 went wrong under certain circumstances in UTF-8 mode. This bug could also
1884 have affected the operation of pcre_study().
1885
1886 25. Realized that a little bit of performance could be had by replacing
1887 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1888
1889 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1890
1891 27. Possessive quantifiers such as a++ were previously implemented by turning
1892 them into atomic groups such as ($>a+). Now they have their own opcodes,
1893 which improves performance. This includes the automatically created ones
1894 from 23 above.
1895
1896 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1897 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1898 the first matched character to be a colon. This applied both to named and
1899 numbered groups.
1900
1901 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1902
1903 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1904 building PCRE on Windows easier, so I have included it in the distribution.
1905
1906 31. There is now a check in pcretest against a ridiculously large number being
1907 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1908 loop, the loop is abandoned.
1909
1910 32. Forward references to subpatterns in conditions such as (?(2)...) where
1911 subpattern 2 is defined later cause pcre_compile() to search forwards in
1912 the pattern for the relevant set of parentheses. This search went wrong
1913 when there were unescaped parentheses in a character class, parentheses
1914 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1915
1916 33. "Subroutine" calls and backreferences were previously restricted to
1917 referencing subpatterns earlier in the regex. This restriction has now
1918 been removed.
1919
1920 34. Added a number of extra features that are going to be in Perl 5.10. On the
1921 whole, these are just syntactic alternatives for features that PCRE had
1922 previously implemented using the Python syntax or my own invention. The
1923 other formats are all retained for compatibility.
1924
1925 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1926 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1927 also .NET compatible.
1928
1929 (b) A recursion or subroutine call to a named group can now be defined as
1930 (?&name) as well as (?P>name).
1931
1932 (c) A backreference to a named group can now be defined as \k<name> or
1933 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1934 5.10, are also .NET compatible.
1935
1936 (d) A conditional reference to a named group can now use the syntax
1937 (?(<name>) or (?('name') as well as (?(name).
1938
1939 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1940 groups (named and numbered) that are never evaluated inline, but can be
1941 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1942 is always false. There may be only one alternative in such a group.
1943
1944 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1945 as the simple (?(R). The condition is true only if the most recent
1946 recursion is that of the given number or name. It does not search out
1947 through the entire recursion stack.
1948
1949 (g) The escape \gN or \g{N} has been added, where N is a positive or
1950 negative number, specifying an absolute or relative reference.
1951
1952 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1953 some "unreachable code" warnings.
1954
1955 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1956 things, this adds five new scripts.
1957
1958 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1959 There were also incompatibilities regarding the handling of \Q..\E inside
1960 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1961 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1962
1963 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1964 matches an empty string, and forcibly breaks the loop. There were bugs in
1965 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1966 against aaaa the result was just "a" rather than "aaaa", for example. Two
1967 separate and independent bugs (that affected different cases) have been
1968 fixed.
1969
1970 39. Refactored the code to abolish the use of different opcodes for small
1971 capturing bracket numbers. This is a tidy that I avoided doing when I
1972 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1973 The new approach is not only tidier, it makes it possible to reduce the
1974 memory needed to fix the previous bug (38).
1975
1976 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1977 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1978 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1979 mode.
1980
1981 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1982 report.
1983
1984 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1985 copy construction and assignment in the C++ wrapper.
1986
1987 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1988 couple of bugs that could have given wrong results in the "--newline=crlf"
1989 case.
1990
1991 44. Added a number of casts and did some reorganization of signed/unsigned int
1992 variables following suggestions from Dair Grant. Also renamed the variable
1993 "this" as "item" because it is a C++ keyword.
1994
1995 45. Arranged for dftables to add
1996
1997 #include "pcre_internal.h"
1998
1999 to pcre_chartables.c because without it, gcc 4.x may remove the array
2000 definition from the final binary if PCRE is built into a static library and
2001 dead code stripping is activated.
2002
2003 46. For an unanchored pattern, if a match attempt fails at the start of a
2004 newline sequence, and the newline setting is CRLF or ANY, and the next two
2005 characters are CRLF, advance by two characters instead of one.
2006
2007
2008 Version 6.7 04-Jul-06
2009 ---------------------
2010
2011 1. In order to handle tests when input lines are enormously long, pcretest has
2012 been re-factored so that it automatically extends its buffers when
2013 necessary. The code is crude, but this _is_ just a test program. The
2014 default size has been increased from 32K to 50K.
2015
2016 2. The code in pcre_study() was using the value of the re argument before
2017 testing it for NULL. (Of course, in any sensible call of the function, it
2018 won't be NULL.)
2019
2020 3. The memmove() emulation function in pcre_internal.h, which is used on
2021 systems that lack both memmove() and bcopy() - that is, hardly ever -
2022 was missing a "static" storage class specifier.
2023
2024 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2025 containing an extended class (one that cannot be represented by a bitmap
2026 because it contains high-valued characters or Unicode property items, e.g.
2027 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2028 pattern, but PCRE should not loop if you do not (it no longer does).
2029 [Detail: two cases were found: (a) a repeated subpattern containing an
2030 extended class; (b) a recursive reference to a subpattern that followed a
2031 previous extended class. It wasn't skipping over the extended class
2032 correctly when UTF-8 mode was not set.]
2033
2034 5. A negated single-character class was not being recognized as fixed-length
2035 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2036 compile error "lookbehind assertion is not fixed length".
2037
2038 6. The RunPerlTest auxiliary script was showing an unexpected difference
2039 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2040 write a Perl script that can interpret lines of an input file either as
2041 byte characters or as UTF-8, which is what "perltest" was being required to
2042 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2043 can't do is switch easily at run time between having the "use utf8;" pragma
2044 or not. In the end, I fudged it by using the RunPerlTest script to insert
2045 "use utf8;" explicitly for the UTF-8 tests.
2046
2047 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2048 the end of the subject string, contrary to the documentation and to what
2049 Perl does. This was true of both matching functions. Now it matches only at
2050 the start of the subject and immediately after *internal* newlines.
2051
2052 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2053 a pointer to an int instead of a pointer to an unsigned long int. This
2054 caused problems on 64-bit systems.
2055
2056 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2057 instance of the 'standard' template library not being so standard".
2058
2059 10. There was no check on the number of named subpatterns nor the maximum
2060 length of a subpattern name. The product of these values is used to compute
2061 the size of the memory block for a compiled pattern. By supplying a very
2062 long subpattern name and a large number of named subpatterns, the size
2063 computation could be caused to overflow. This is now prevented by limiting
2064 the length of names to 32 characters, and the number of named subpatterns
2065 to 10,000.
2066
2067 11. Subpatterns that are repeated with specific counts have to be replicated in
2068 the compiled pattern. The size of memory for this was computed from the
2069 length of the subpattern and the repeat count. The latter is limited to
2070 65535, but there was no limit on the former, meaning that integer overflow
2071 could in principle occur. The compiled length of a repeated subpattern is
2072 now limited to 30,000 bytes in order to prevent this.
2073
2074 12. Added the optional facility to have named substrings with the same name.
2075
2076 13. Added the ability to use a named substring as a condition, using the
2077 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2078 are numbers (not recommended). Forward references are permitted.
2079
2080 14. Added forward references in named backreferences (if you see what I mean).
2081
2082 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2083 pattern could run off the end of the subject. For example, the pattern
2084 "(?s)(.{1,5})"8 did this with the subject "ab".
2085
2086 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2087 PCRE_CASELESS was set when matching characters that were quantified with ?
2088 or *.
2089
2090 17. A character class other than a single negated character that had a minimum
2091 but no maximum quantifier - for example [ab]{6,} - was not handled
2092 correctly by pce_dfa_exec(). It would match only one character.
2093
2094 18. A valid (though odd) pattern that looked like a POSIX character
2095 class but used an invalid character after [ (for example [[,abc,]]) caused
2096 pcre_compile() to give the error "Failed: internal error: code overflow" or
2097 in some cases to crash with a glibc free() error. This could even happen if
2098 the pattern terminated after [[ but there just happened to be a sequence of
2099 letters, a binary zero, and a closing ] in the memory that followed.
2100
2101 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2102 over the years. Originally (before any Unicode support), just the bottom 8
2103 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2104 output from "man perlunicode" includes this:
2105
2106 The regular expression compiler produces polymorphic opcodes. That
2107 is, the pattern adapts to the data and automatically switches to
2108 the Unicode character scheme when presented with Unicode data--or
2109 instead uses a traditional byte scheme when presented with byte
2110 data.
2111
2112 Sadly, a wide octal escape does not cause a switch, and in a string with
2113 no other multibyte characters, these octal escapes are treated as before.
2114 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2115 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2116 Unicode string.
2117
2118 I have not perpetrated such confusion in PCRE. Up till now, it took just
2119 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2120 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2121 translate to the appropriate multibyte character.
2122
2123 29. Applied some refactoring to reduce the number of warnings from Microsoft
2124 and Borland compilers. This has included removing the fudge introduced
2125 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2126 a warning about an unused variable.
2127
2128 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2129 characters since release 4.0, because Perl (from release 5.004) does not.
2130 [Or at least, is documented not to: some releases seem to be in conflict
2131 with the documentation.] However, when a pattern was studied with
2132 pcre_study() and all its branches started with \s, PCRE still included VT
2133 as a possible starting character. Of course, this did no harm; it just
2134 caused an unnecessary match attempt.
2135
2136 22. Removed a now-redundant internal flag bit that recorded the fact that case
2137 dependency changed within the pattern. This was once needed for "required
2138 byte" processing, but is no longer used. This recovers a now-scarce options
2139 bit. Also moved the least significant internal flag bit to the most-
2140 significant bit of the word, which was not previously used (hangover from
2141 the days when it was an int rather than a uint) to free up another bit for
2142 the future.
2143
2144 23. Added support for CRLF line endings as well as CR and LF. As well as the
2145 default being selectable at build time, it can now be changed at runtime
2146 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2147 specify that it is scanning data with non-default line endings.
2148
2149 24. Changed the definition of CXXLINK to make it agree with the definition of
2150 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2151
2152 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2153 recursions. This makes a big different to stack usage for some patterns.
2154
2155 26. If a subpattern containing a named recursion or subroutine reference such
2156 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2157 the space required for the compiled pattern went wrong and gave too small a
2158 value. Depending on the environment, this could lead to "Failed: internal
2159 error: code overflow at offset 49" or "glibc detected double free or
2160 corruption" errors.
2161
2162 27. Applied patches from Google (a) to support the new newline modes and (b) to
2163 advance over multibyte UTF-8 characters in GlobalReplace.
2164
2165 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2166 difference for some implementation of PCRE in some Windows version.
2167
2168 29. Added some extra testing facilities to pcretest:
2169
2170 \q<number> in a data line sets the "match limit" value
2171 \Q<number> in a data line sets the "match recursion limt" value
2172 -S <number> sets the stack size, where <number> is in megabytes
2173
2174 The -S option isn't available for Windows.
2175
2176
2177 Version 6.6 06-Feb-06
2178 ---------------------
2179
2180 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2181 in pcreposix.h. I have copied the definition from pcre.h.
2182
2183 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2184 because pcre.h is no longer a built file.
2185
2186 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2187 not normally included in the compiled code.
2188
2189
2190 Version 6.5 01-Feb-06
2191 ---------------------
2192
2193 1. When using the partial match feature with pcre_dfa_exec(), it was not
2194 anchoring the second and subsequent partial matches at the new starting
2195 point. This could lead to incorrect results. For example, with the pattern
2196 /1234/, partially matching against "123" and then "a4" gave a match.
2197
2198 2. Changes to pcregrep:
2199
2200 (a) All non-match returns from pcre_exec() were being treated as failures
2201 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2202 error message is output. Some extra information is given for the
2203 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2204 probably the only errors that are likely to be caused by users (by
2205 specifying a regex that has nested indefinite repeats, for instance).
2206 If there are more than 20 of these errors, pcregrep is abandoned.
2207
2208 (b) A binary zero was treated as data while matching, but terminated the
2209 output line if it was written out. This has been fixed: binary zeroes
2210 are now no different to any other data bytes.
2211
2212 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2213 used to set a locale for matching. The --locale=xxxx long option has
2214 been added (no short equivalent) to specify a locale explicitly on the
2215 pcregrep command, overriding the environment variables.
2216
2217 (d) When -B was used with -n, some line numbers in the output were one less
2218 than they should have been.
2219
2220 (e) Added the -o (--only-matching) option.
2221
2222 (f) If -A or -C was used with -c (count only), some lines of context were
2223 accidentally printed for the final match.
2224
2225 (g) Added the -H (--with-filename) option.
2226
2227 (h) The combination of options -rh failed to suppress file names for files
2228 that were found from directory arguments.
2229
2230 (i) Added the -D (--devices) and -d (--directories) options.
2231
2232 (j) Added the -F (--fixed-strings) option.
2233
2234 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2235
2236 (l) Added the --colo(u)r option.
2237
2238 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2239 is not present by default.
2240
2241 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2242 items such as (?R) or (?1), when the recursion could match a number of
2243 alternatives. If it matched one of the alternatives, but subsequently,
2244 outside the recursion, there was a failure, the code tried to back up into
2245 the recursion. However, because of the way PCRE is implemented, this is not
2246 possible, and the result was an incorrect result from the match.
2247
2248 In order to prevent this happening, the specification of recursion has
2249 been changed so that all such subpatterns are automatically treated as
2250 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2251
2252 4. I had overlooked the fact that, in some locales, there are characters for
2253 which isalpha() is true but neither isupper() nor islower() are true. In
2254 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2255 and ordfeminine) are like this. This affected the treatment of \w and \W
2256 when they appeared in character classes, but not when they appeared outside
2257 a character class. The bit map for "word" characters is now created
2258 separately from the results of isalnum() instead of just taking it from the
2259 upper, lower, and digit maps. (Plus the underscore character, of course.)
2260
2261 5. The above bug also affected the handling of POSIX character classes such as
2262 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2263 permanent tables. Instead, the bit maps for such a class were previously
2264 created as the appropriate unions of the upper, lower, and digit bitmaps.
2265 Now they are created by subtraction from the [[:word:]] class, which has
2266 its own bitmap.
2267
2268 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2269 It is created by subtracting the vertical space characters (\x09, \x0a,
2270 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2271 subtraction was done in the overall bitmap for a character class, meaning
2272 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2273 be recognized. This bug has been fixed.
2274
2275 7. Patches from the folks at Google:
2276
2277 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2278 real life, but is still worth protecting against".
2279
2280 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2281 regular expressions".
2282
2283 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2284 have it.
2285
2286 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2287 "configure" and the latter not, in order to fix a problem somebody had
2288 with compiling the Arg class on HP-UX.
2289
2290 (e) Improve the error-handling of the C++ wrapper a little bit.
2291
2292 (f) New tests for checking recursion limiting.
2293
2294 8. The pcre_memmove() function, which is used only if the environment does not
2295 have a standard memmove() function (and is therefore rarely compiled),
2296 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2297 returning a result (though PCRE never actually uses the result).
2298
2299 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2300 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2301 returned instead of calling malloc() with an overflowing number that would
2302 most likely cause subsequent chaos.
2303
2304 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2305
2306 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2307 with this option is matched, the nmatch and pmatch options of regexec() are
2308 ignored.
2309
2310 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2311 provided in case anyone wants to the the POSIX interface with UTF-8
2312 strings.
2313
2314 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2315 C++ linking (needed for some HP-UX environments).
2316
2317 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2318 (unused parameter) and in the pcre_printint() function (omitted "default"
2319 switch label when the default is to do nothing).
2320
2321 15. Added some code to make it possible, when PCRE is compiled as a C++
2322 library, to replace subject pointers for pcre_exec() with a smart pointer
2323 class, thus making it possible to process discontinuous strings.
2324
2325 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2326 much the same function. They were added by different people who were trying
2327 to make PCRE easy to compile on non-Unix systems. It has been suggested
2328 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2329 for compiling on Windows systems. I have therefore replaced it with
2330 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2331 defaults to "extern" for C or "extern C" for C++, which works fine on
2332 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2333 SCOPE with something explicit in config.h. In addition:
2334
2335 (a) pcreposix.h still had just "extern" instead of either of these macros;
2336 I have replaced it with PCRE_DATA_SCOPE.
2337
2338 (b) Functions such as _pcre_xclass(), which are internal to the library,
2339 but external in the C sense, all had PCRE_EXPORT in their definitions.
2340 This is apparently wrong for the Windows case, so I have removed it.
2341 (It makes no difference on Unix-like systems.)
2342
2343 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2344 of recursive calls to match(). This is different to MATCH_LIMIT because
2345 that limits the total number of calls to match(), not all of which increase
2346 the depth of recursion. Limiting the recursion depth limits the amount of
2347 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2348 when PCRE is compiled, and changed at run time. A patch from Google adds
2349 this functionality to the C++ interface.
2350
2351 18. Changes to the handling of Unicode character properties:
2352
2353 (a) Updated the table to Unicode 4.1.0.
2354
2355 (b) Recognize characters that are not in the table as "Cn" (undefined).
2356
2357 (c) I revised the way the table is implemented to a much improved format
2358 which includes recognition of ranges. It now supports the ranges that
2359 are defined in UnicodeData.txt, and it also amalgamates other
2360 characters into ranges. This has reduced the number of entries in the
2361 table from around 16,000 to around 3,000, thus reducing its size
2362 considerably. I realized I did not need to use a tree structure after
2363 all - a binary chop search is just as efficient. Having reduced the
2364 number of entries, I extended their size from 6 bytes to 8 bytes to
2365 allow for more data.
2366
2367 (d) Added support for Unicode script names via properties such as \p{Han}.
2368
2369 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2370 matching that character.
2371
2372 20. When matching a repeated Unicode property with a minimum greater than zero,
2373 (for example \pL{2,}), PCRE could look past the end of the subject if it
2374 reached it while seeking the minimum number of characters. This could
2375 happen only if some of the characters were more than one byte long, because
2376 there is a check for at least the minimum number of bytes.
2377
2378 21. Refactored the implementation of \p and \P so as to be more general, to
2379 allow for more different types of property in future. This has changed the
2380 compiled form incompatibly. Anybody with saved compiled patterns that use
2381 \p or \P will have to recompile them.
2382
2383 22. Added "Any" and "L&" to the supported property types.
2384
2385 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2386 but give a compile time error if the value is greater than 0xff.
2387
2388 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2389 accidentally not being installed or uninstalled.
2390
2391 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2392 made were to insert the current release number. This seemed silly, because
2393 it made things harder for people building PCRE on systems that don't run
2394 "configure". I have turned pcre.h into a distributed file, no longer built
2395 by "configure", with the version identification directly included. There is
2396 no longer a pcre.h.in file.
2397
2398 However, this change necessitated a change to the pcre-config script as
2399 well. It is built from pcre-config.in, and one of the substitutions was the
2400 release number. I have updated configure.ac so that ./configure now finds
2401 the release number by grepping pcre.h.
2402
2403 26. Added the ability to run the tests under valgrind.
2404
2405
2406 Version 6.4 05-Sep-05
2407 ---------------------
2408
2409 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2410 "--" to be printed when multiple files were scanned, even when none of the
2411 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2412 consider it to be a bug, and have restored the previous behaviour.
2413
2414 2. A couple of code tidies to get rid of compiler warnings.
2415
2416 3. The pcretest program used to cheat by referring to symbols in the library
2417 whose names begin with _pcre_. These are internal symbols that are not
2418 really supposed to be visible externally, and in some environments it is
2419 possible to suppress them. The cheating is now confined to including
2420 certain files from the library's source, which is a bit cleaner.
2421
2422 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2423 file's purpose clearer.
2424
2425 5. Reorganized pcre_ucp_findchar().
2426
2427
2428 Version 6.3 15-Aug-05
2429 ---------------------
2430
2431 1. The file libpcre.pc.in did not have general read permission in the tarball.
2432
2433 2. There were some problems when building without C++ support:
2434
2435 (a) If C++ support was not built, "make install" and "make test" still
2436 tried to test it.
2437
2438 (b) There were problems when the value of CXX was explicitly set. Some
2439 changes have been made to try to fix these, and ...
2440
2441 (c) --disable-cpp can now be used to explicitly disable C++ support.
2442
2443 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2444 backslash in a target when C++ was disabled. This confuses some
2445 versions of "make", apparently. Using an intermediate variable solves
2446 this. (Same for CPP_LOBJ.)
2447
2448 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2449 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2450 necessary on certain architectures.
2451
2452 4. Added a setting of -export-symbols-regex to the link command to remove
2453 those symbols that are exported in the C sense, but actually are local
2454 within the library, and not documented. Their names all begin with
2455 "_pcre_". This is not a perfect job, because (a) we have to except some
2456 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2457 available (and never for static libraries). I have made a note to try to
2458 find a way round (a) in the future.
2459
2460
2461 Version 6.2 01-Aug-05
2462 ---------------------
2463
2464 1. There was no test for integer overflow of quantifier values. A construction
2465 such as {1111111111111111} would give undefined results. What is worse, if
2466 a minimum quantifier for a parenthesized subpattern overflowed and became
2467 negative, the calculation of the memory size went wrong. This could have
2468 led to memory overwriting.
2469
2470 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2471
2472 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2473 operating environments where this matters.
2474
2475 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2476 PCRE options from within the C++ wrapper.
2477
2478 5. Named capturing subpatterns were not being correctly counted when a pattern
2479 was compiled. This caused two problems: (a) If there were more than 100
2480 such subpatterns, the calculation of the memory needed for the whole
2481 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2482 back references of the form \12, where the number was greater than 9, were
2483 not recognized as back references, even though there were sufficient
2484 previous subpatterns.
2485
2486 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2487 versions of gcc, e.g. 2.95.4.
2488
2489
2490 Version 6.1 21-Jun-05
2491 ---------------------
2492
2493 1. There was one reference to the variable "posix" in pcretest.c that was not
2494 surrounded by "#if !defined NOPOSIX".
2495
2496 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2497 the cross-check on the old pcre_info() function, for the benefit of the
2498 cut-down version of PCRE that is currently imported into Exim.
2499
2500 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2501 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2502 patterns that start (?i) but I don't think that matters. The use of (?i) is
2503 just an example; this all applies to the other options as well.
2504
2505 4. Since libtool seems to echo the compile commands it is issuing, the output
2506 from "make" can be reduced a bit by putting "@" in front of each libtool
2507 compile command.
2508
2509 5. Patch from the folks at Google for configure.in to be a bit more thorough
2510 in checking for a suitable C++ installation before trying to compile the
2511 C++ stuff. This should fix a reported problem when a compiler was present,
2512 but no suitable headers.
2513
2514 6. The man pages all had just "PCRE" as their title. I have changed them to
2515 be the relevant file name. I have also arranged that these names are
2516 retained in the file doc/pcre.txt, which is a concatenation in text format
2517 of all the man pages except the little individual ones for each function.
2518
2519 7. The NON-UNIX-USE file had not been updated for the different set of source
2520 files that come with release 6. I also added a few comments about the C++
2521 wrapper.
2522
2523
2524 Version 6.0 07-Jun-05
2525 ---------------------
2526
2527 1. Some minor internal re-organization to help with my DFA experiments.
2528
2529 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2530 didn't matter for the library itself when fully configured, but did matter
2531 when compiling without UCP support, or within Exim, where the ucp files are
2532 not imported.
2533
2534 3. Refactoring of the library code to split up the various functions into
2535 different source modules. The addition of the new DFA matching code (see
2536 below) to a single monolithic source would have made it really too
2537 unwieldy, quite apart from causing all the code to be include in a
2538 statically linked application, when only some functions are used. This is
2539 relevant even without the DFA addition now that patterns can be compiled in
2540 one application and matched in another.
2541
2542 The downside of splitting up is that there have to be some external
2543 functions and data tables that are used internally in different modules of
2544 the library but which are not part of the API. These have all had their
2545 names changed to start with "_pcre_" so that they are unlikely to clash
2546 with other external names.
2547
2548 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2549 a different (DFA) algorithm. Although it is slower than the original
2550 function, it does have some advantages for certain types of matching
2551 problem.
2552
2553 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2554 including restarting after a partial match.
2555
2556 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2557 defined when compiling for Windows was sent to me. I have put it into the
2558 code, though I have no means of testing or verifying it.
2559
2560 7. Added the pcre_refcount() auxiliary function.
2561
2562 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2563 match before or at the first newline in the subject string. In pcretest,
2564 the /f option on a pattern can be used to set this.
2565
2566 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2567 would behave wrongly. This has been present in PCRE since release 4.0.
2568
2569 10. A number of changes to the pcregrep command:
2570
2571 (a) Refactored how -x works; insert ^(...)$ instead of setting
2572 PCRE_ANCHORED and checking the length, in preparation for adding
2573 something similar for -w.
2574
2575 (b) Added the -w (match as a word) option.
2576
2577 (c) Refactored the way lines are read and buffered so as to have more
2578 than one at a time available.
2579
2580 (d) Implemented a pcregrep test script.
2581
2582 (e) Added the -M (multiline match) option. This allows patterns to match
2583 over several lines of the subject. The buffering ensures that at least
2584 8K, or the rest of the document (whichever is the shorter) is available
2585 for matching (and similarly the previous 8K for lookbehind assertions).
2586
2587 (f) Changed the --help output so that it now says
2588
2589 -w, --word-regex(p)
2590
2591 instead of two lines, one with "regex" and the other with "regexp"
2592 because that confused at least one person since the short forms are the
2593 same. (This required a bit of code, as the output is generated
2594 automatically from a table. It wasn't just a text change.)
2595
2596 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2597 option but starts with a hyphen. Could be a pattern or a path name
2598 starting with a hyphen, for instance.
2599
2600 (h) "-" can be given as a file name to represent stdin.
2601
2602 (i) When file names are being printed, "(standard input)" is used for
2603 the standard input, for compatibility with GNU grep. Previously
2604 "<stdin>" was used.
2605
2606 (j) The option --label=xxx can be used to supply a name to be used for
2607 stdin when file names are being printed. There is no short form.
2608
2609 (k) Re-factored the options decoding logic because we are going to add
2610 two more options that take data. Such options can now be given in four
2611 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2612
2613 (l) Added the -A, -B, and -C options for requesting that lines of context
2614 around matches be printed.
2615
2616 (m) Added the -L option to print the names of files that do not contain
2617 any matching lines, that is, the complement of -l.
2618
2619 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2620 continue to scan other files.
2621
2622 (o) The -s option was incorrectly implemented. For compatibility with other
2623 greps, it now suppresses the error message for a non-existent or non-
2624 accessible file (but not the return code). There is a new option called
2625 -q that suppresses the output of matching lines, which was what -s was
2626 previously doing.
2627
2628 (p) Added --include and --exclude options to specify files for inclusion
2629 and exclusion when recursing.
2630
2631 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2632 Hopefully, it now does.
2633
2634 12. Missing cast in pcre_study().
2635
2636 13. Added an "uninstall" target to the makefile.
2637
2638 14. Replaced "extern" in the function prototypes in Makefile.in with
2639 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2640 world, but is set differently for Windows.
2641
2642 15. Added a second compiling function called pcre_compile2(). The only
2643 difference is that it has an extra argument, which is a pointer to an
2644 integer error code. When there is a compile-time failure, this is set
2645 non-zero, in addition to the error test pointer being set to point to an
2646 error message. The new argument may be NULL if no error number is required
2647 (but then you may as well call pcre_compile(), which is now just a
2648 wrapper). This facility is provided because some applications need a
2649 numeric error indication, but it has also enabled me to tidy up the way
2650 compile-time errors are handled in the POSIX wrapper.
2651
2652 16. Added VPATH=.libs to the makefile; this should help when building with one
2653 prefix path and installing with another. (Or so I'm told by someone who
2654 knows more about this stuff than I do.)
2655
2656 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2657 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2658 match everything, including newlines. This is not POSIX-compatible, but
2659 somebody wanted the feature. From pcretest it can be activated by using
2660 both the P and the s flags.
2661
2662 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2663
2664 19. libpcre.pc was being incorrectly installed as executable.
2665
2666 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2667 it now also looks for '\r' so that it will work unmodified on Windows.
2668
2669 21. Added Google's contributed C++ wrapper to the distribution.
2670
2671 22. Added some untidy missing memory free() calls in pcretest, to keep
2672 Electric Fence happy when testing.
2673
2674
2675
2676 Version 5.0 13-Sep-04
2677 ---------------------
2678
2679 1. Internal change: literal characters are no longer packed up into items
2680 containing multiple characters in a single byte-string. Each character
2681 is now matched using a separate opcode. However, there may be more than one
2682 byte in the character in UTF-8 mode.
2683
2684 2. The pcre_callout_block structure has two new fields: pattern_position and
2685 next_item_length. These contain the offset in the pattern to the next match
2686 item, and its length, respectively.
2687
2688 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2689 insertion of callouts before each pattern item. Added the /C option to
2690 pcretest to make use of this.
2691
2692 4. On the advice of a Windows user, the lines
2693
2694 #if defined(_WIN32) || defined(WIN32)
2695 _setmode( _fileno( stdout ), 0x8000 );
2696 #endif /* defined(_WIN32) || defined(WIN32) */
2697
2698 have been added to the source of pcretest. This apparently does useful
2699 magic in relation to line terminators.
2700
2701 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2702 for the benefit of those environments where the "b" makes a difference.
2703
2704 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2705 to know about it. I have put a hack into configure.in that adds in code
2706 to set GCC=yes if CC=icc. This seems to end up at a point in the
2707 generated configure script that is early enough to affect the setting of
2708 compiler options, which is what is needed, but I have no means of testing
2709 whether it really works. (The user who reported this had patched the
2710 generated configure script, which of course I cannot do.)
2711
2712 LATER: After change 22 below (new libtool files), the configure script
2713 seems to know about icc (and also ecc). Therefore, I have commented out
2714 this hack in configure.in.
2715
2716 7. Added support for pkg-config (2 patches were sent in).
2717
2718 8. Negated POSIX character classes that used a combination of internal tables
2719 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2720 [[:^ascii]]. Typically, they would match almost any characters. The other
2721 POSIX classes were not broken in this way.
2722
2723 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2724 to find the match, as PCRE was deluded into thinking that the match had to
2725 start at the start point or following a newline. The same bug applied to
2726 patterns with negative forward assertions or any backward assertions
2727 preceding ".*" at the start, unless the pattern required a fixed first
2728 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2729
2730 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2731 starting at the last subject character, bytes beyond the end of the subject
2732 string were read.
2733
2734 11. Renamed the variable "class" as "classbits" to make life easier for C++
2735 users. (Previously there was a macro definition, but it apparently wasn't
2736 enough.)
2737
2738 12. Added the new field "tables" to the extra data so that tables can be passed
2739 in at exec time, or the internal tables can be re-selected. This allows
2740 a compiled regex to be saved and re-used at a later time by a different
2741 program that might have everything at different addresses.
2742
2743 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2744 -R library as well as a -L library.
2745
2746 14. The debugging options of pcretest (-d on the command line or D on a
2747 pattern) showed incorrect output for anything following an extended class
2748 that contained multibyte characters and which was followed by a quantifier.
2749
2750 15. Added optional support for general category Unicode character properties
2751 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2752 support. It adds about 90K to the size of the library. The meanings of the
2753 inbuilt class escapes such as \d and \s have NOT been changed.
2754
2755 16. Updated pcredemo.c to include calls to free() to release the memory for the
2756 compiled pattern.
2757
2758 17. The generated file chartables.c was being created in the source directory
2759 instead of in the building directory. This caused the build to fail if the
2760 source directory was different from the building directory, and was
2761 read-only.
2762
2763 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2764 file. No doubt somebody will tell me if they don't make sense... Also added
2765 Dan Mooney's comments about building on OpenVMS.
2766
2767 19. Added support for partial matching via the PCRE_PARTIAL option for
2768 pcre_exec() and the \P data escape in pcretest.
2769
2770 20. Extended pcretest with 3 new pattern features:
2771
2772 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2773 write the compiled pattern to the file whose name is "rest-of-line".
2774 This is a straight binary dump of the data, with the saved pointer to
2775 the character tables forced to be NULL. The study data, if any, is
2776 written too. After writing, pcretest reads a new pattern.
2777
2778 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2779 compiled pattern from the given file. There must not be any
2780 occurrences of "<" in the file name (pretty unlikely); if there are,
2781 pcretest will instead treat the initial "<" as a pattern delimiter.
2782 After reading in the pattern, pcretest goes on to read data lines as
2783 usual.
2784
2785 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2786 and 16-bit fields in a compiled pattern, to simulate a pattern that
2787 was compiled on a host of opposite endianness.
2788
2789 21. The pcre-exec() function can now cope with patterns that were compiled on
2790 hosts of opposite endianness, with this restriction:
2791
2792 As for any compiled expression that is saved and used later, the tables
2793 pointer field cannot be preserved; the extra_data field in the arguments
2794 to pcre_exec() should be used to pass in a tables address if a value
2795 other than the default internal tables were used at compile time.
2796
2797 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2798 now diagnosed as an error. Previously, most of the time, a negative number
2799 would have been treated as zero, but if in addition "ovector" was passed as
2800 NULL, a crash could occur.
2801
2802 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2803 new versions from the libtool 1.5 distribution (the last one is a copy of
2804 a file called libtool.m4). This seems to have fixed the need to patch
2805 "configure" to support Darwin 1.3 (which I used to do). However, I still
2806 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2807 workstation).
2808
2809 24. Changed the PCRE licence to be the more standard "BSD" licence.
2810
2811
2812 Version 4.5 01-Dec-03
2813 ---------------------
2814
2815 1. There has been some re-arrangement of the code for the match() function so
2816 that it can be compiled in a version that does not call itself recursively.
2817 Instead, it keeps those local variables that need separate instances for
2818 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2819 needs to "recurse". Keeping track of where control must go is done by means
2820 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2821 hide most of the details from the main code, and operates only if
2822 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2823 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2824 operating.
2825
2826 To make it easier for callers to provide specially tailored get/free
2827 functions for this usage, two new functions, pcre_stack_malloc, and
2828 pcre_stack_free, are used. They are always called in strict stacking order,
2829 and the size of block requested is always the same.
2830
2831 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2832 PCRE has been compiled to use the stack or the heap for recursion. The
2833 -C option of pcretest uses this to show which version is compiled.
2834
2835 A new data escape \S, is added to pcretest; it causes the amounts of store
2836 obtained and freed by both kinds of malloc/free at match time to be added
2837 to the output.
2838
2839 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2840 what's available on my current Linux desktop machine.
2841
2842 3. When matching a UTF-8 string, the test for a valid string at the start has
2843 been extended. If start_offset is not zero, PCRE now checks that it points
2844 to a byte that is the start of a UTF-8 character. If not, it returns
2845 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2846 this is necessary because there may be backward assertions in the pattern.
2847 When matching the same subject several times, it may save resources to use
2848 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2849
2850 4. The code for checking the validity of UTF-8 strings has been tightened so
2851 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2852 containing "overlong sequences".
2853
2854 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2855 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2856 should have been "&", but it just so happened that all the cases this let
2857 through by mistake were picked up later in the function.
2858
2859 6. I had used a variable called "isblank" - this is a C99 function, causing
2860 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2861
2862 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2863 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2864 so the version is shown; (c) stop "make test" echoing "./RunTest".
2865
2866 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2867
2868 9. The prototype for memmove() for systems that don't have it was using
2869 size_t, but the inclusion of the header that defines size_t was later. I've
2870 moved the #includes for the C headers earlier to avoid this.
2871
2872 10. Added some adjustments to the code to make it easier to compiler on certain
2873 special systems:
2874
2875 (a) Some "const" qualifiers were missing.
2876 (b) Added the macro EXPORT before all exported functions; by default this
2877 is defined to be empty.
2878 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2879 that it reads its output file name as an argument instead of writing
2880 to the standard output and assuming this can be redirected.
2881
2882 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2883 class containing characters with values greater than 255, PCRE compilation
2884 went into a loop.
2885
2886 12. A recursive reference to a subpattern that was within another subpattern
2887 that had a minimum quantifier of zero caused PCRE to crash. For example,
2888 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2889 recursion. If the recursively-called subpattern itself had a zero repeat,
2890 that was OK.
2891
2892 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2893 buffer into which it was copied (for escape processing) was still set at
2894 1024, so long lines caused crashes.
2895
2896 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2897 "internal error: code overflow...". This applied to any character class
2898 that was followed by a possessive quantifier.
2899
2900 15. Modified the Makefile to add libpcre.la as a prerequisite for
2901 libpcreposix.la because I was told this is needed for a parallel build to
2902 work.
2903
2904 16. If a pattern that contained .* following optional items at the start was
2905 studied, the wrong optimizing data was generated, leading to matching
2906 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2907 matching string must start with a or b or c. The correct conclusion for
2908 this pattern is that a match can start with any character.
2909
2910
2911 Version 4.4 13-Aug-03
2912 ---------------------
2913
2914 1. In UTF-8 mode, a character class containing characters with values between
2915 127 and 255 was not handled correctly if the compiled pattern was studied.
2916 In fixing this, I have also improved the studying algorithm for such
2917 classes (slightly).
2918
2919 2. Three internal functions had redundant arguments passed to them. Removal
2920 might give a very teeny performance improvement.
2921
2922 3. Documentation bug: the value of the capture_top field in a callout is *one
2923 more than* the number of the hightest numbered captured substring.
2924
2925 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2926 in incorrectly linking with a previously installed version. They now link
2927 explicitly with libpcre.la.
2928
2929 5. configure.in no longer needs to recognize Cygwin specially.
2930
2931 6. A problem in pcre.in for Windows platforms is fixed.
2932
2933 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2934 pcretest, it used to include the size of the study block as part of its
2935 output. Unfortunately, the structure contains a field that has a different
2936 size on different hardware architectures. This meant that the tests that
2937 showed this size failed. As the block is currently always of a fixed size,
2938 this information isn't actually particularly useful in pcretest output, so
2939 I have just removed it.
2940
2941 8. Three pre-processor statements accidentally did not start in column 1.
2942 Sadly, there are *still* compilers around that complain, even though
2943 standard C has not required this for well over a decade. Sigh.
2944
2945 9. In pcretest, the code for checking callouts passed small integers in the
2946 callout_data field, which is a void * field. However, some picky compilers
2947 complained about the casts involved for this on 64-bit systems. Now
2948 pcretest passes the address of the small integer instead, which should get
2949 rid of the warnings.
2950
2951 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2952 both compile and run time, and gives an error if an invalid UTF-8 sequence
2953 is found. There is a option for disabling this check in cases where the
2954 string is known to be correct and/or the maximum performance is wanted.
2955
2956 11. In response to a bug report, I changed one line in Makefile.in from
2957
2958 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2959 to
2960 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2961
2962 to look similar to other lines, but I have no way of telling whether this
2963 is the right thing to do, as I do not use Windows. No doubt I'll get told
2964 if it's wrong...
2965
2966
2967 Version 4.3 21-May-03
2968 ---------------------
2969
2970 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2971 Makefile.
2972
2973 2. Some refactoring to improve the quality of the code:
2974
2975 (i) The utf8_table... variables are now declared "const".
2976
2977 (ii) The code for \cx, which used the "case flipping" table to upper case
2978 lower case letters, now just substracts 32. This is ASCII-specific,
2979 but the whole concept of \cx is ASCII-specific, so it seems
2980 reasonable.
2981
2982 (iii) PCRE was using its character types table to recognize decimal and
2983 hexadecimal digits in the pattern. This is silly, because it handles
2984 only 0-9, a-f, and A-F, but the character types table is locale-
2985 specific, which means strange things might happen. A private
2986 table is now used for this - though it costs 256 bytes, a table is
2987 much faster than multiple explicit tests. Of course, the standard
2988 character types table is still used for matching digits in subject
2989 strings against \d.
2990
2991 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2992 ending in _t are). So I've renamed it as ESC_tee.
2993
2994 3. The first argument for regexec() in the POSIX wrapper should have been
2995 defined as "const".
2996
2997 4. Changed pcretest to use malloc() for its buffers so that they can be
2998 Electric Fenced for debugging.
2999
3000 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3001 to read one or more bytes before the start of the subject string. Often this
3002 had no effect on PCRE's behaviour, but in some circumstances it could
3003 provoke a segmentation fault.
3004
3005 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3006 to try to read one or more bytes before the start of the subject string.
3007
3008 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3009 UTF-8 support could misbehave in various ways if the subject string
3010 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3011 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3012 back over UTF-8 characters.)
3013
3014
3015 Version 4.2 14-Apr-03
3016 ---------------------
3017
3018 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3019
3020 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3021 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3022 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3023 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3024 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3025 and BUILD_EXEEXT
3026 Note: automatic setting of the BUILD variables is not yet working
3027 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3028 compile-time but not at link-time
3029 [LINK]: use for linking executables only
3030 make different versions for Windows and non-Windows
3031 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3032 libraries
3033 [LINK_FOR_BUILD]: new variable
3034 [OBJEXT]: use throughout
3035 [EXEEXT]: use throughout
3036 <winshared>: new target
3037 <wininstall>: new target
3038 <dftables.o>: use native compiler
3039 <dftables>: use native linker
3040 <install>: handle Windows platform correctly
3041 <clean>: ditto
3042 <check>: ditto
3043 copy DLL to top builddir before testing
3044
3045 As part of these changes, -no-undefined was removed again. This was reported
3046 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3047 in any case.
3048
3049 3. Some tidies to get rid of compiler warnings:
3050
3051 . In the match_data structure, match_limit was an unsigned long int, whereas
3052 match_call_count was an int. I've made them both unsigned long ints.
3053
3054 . In pcretest the fact that a const uschar * doesn't automatically cast to
3055 a void * provoked a warning.
3056
3057 . Turning on some more compiler warnings threw up some "shadow" variables
3058 and a few more missing casts.
3059
3060 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3061 option, a class that contained a single character with a value between 128
3062 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3063
3064 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3065 option, a class that contained several characters, but with at least one
3066 whose value was between 128 and 255 caused PCRE to crash.
3067
3068
3069 Version 4.1 12-Mar-03
3070 ---------------------
3071
3072 1. Compiling with gcc -pedantic found a couple of places where casts were
3073 needed, and a string in dftables.c that was longer than standard compilers are
3074 required to support.
3075
3076 2. Compiling with Sun's compiler found a few more places where the code could
3077 be tidied up in order to avoid warnings.
3078
3079 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3080 first of these names is deprecated in the latest Autoconf in favour of the name
3081 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3082 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3083 analogy I have changed it to CFLAGS_FOR_BUILD.
3084
3085 4. Added -no-undefined to the linking command in the Makefile, because this is
3086 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3087 linking step for the pcreposix library.
3088
3089 5. PCRE was failing to diagnose the case of two named groups with the same
3090 name.
3091
3092 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3093 literal character that is needed in the subject for a match, and scans along to
3094 ensure that it is present before embarking on the full matching process. This
3095 saves time in cases of nested unlimited repeats that are never going to match.
3096 Problem: the scan can take a lot of time if the subject is very long (e.g.
3097 megabytes), thus penalizing straightforward matches. It is now done only if the
3098 amount of subject to be scanned is less than 1000 bytes.
3099
3100 7. A lesser problem with the same optimization is that it was recording the
3101 first character of an anchored pattern as "needed", thus provoking a search
3102 right along the subject, even when the first match of the pattern was going to
3103 fail. The "needed" character is now not set for anchored patterns, unless it
3104 follows something in the pattern that is of non-fixed length. Thus, it still
3105 fulfils its original purpose of finding quick non-matches in cases of nested
3106 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3107
3108
3109 Version 4.0 17-Feb-03
3110 ---------------------
3111
3112 1. If a comment in an extended regex that started immediately after a meta-item
3113 extended to the end of string, PCRE compiled incorrect data. This could lead to
3114 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3115
3116 2. Moved to autoconf 2.53 and libtool 1.4.2.
3117
3118 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3119 the special perltest8 script is no longer needed - all the tests can be run
3120 from a single perltest script.
3121
3122 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3123 by \s. It has now been removed in PCRE. This means it isn't recognized as
3124 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3125 class [:space:] *does* include VT, thereby creating a mess.
3126
3127 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3128 space and tab.
3129
3130 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3131 its new features into the main test script, reducing the number of scripts.
3132
3133 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3134 were backward compatible, and made the (?i) apply to the whole pattern, as if
3135 /i were given. Now it behaves more logically, and applies the option setting
3136 only to what follows. PCRE has been changed to follow suit. However, if it
3137 finds options settings right at the start of the pattern, it extracts them into
3138 the global options, as before. Thus, they show up in the info data.
3139
3140 8. Added support for the \Q...\E escape sequence. Characters in between are
3141 treated as literals. This is slightly different from Perl in that $ and @ are
3142 also handled as literals inside the quotes. In Perl, they will cause variable
3143 interpolation. Note the following examples:
3144
3145 Pattern PCRE matches Perl matches
3146
3147 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3148 \Qabc\$xyz\E abc\$xyz abc\$xyz
3149 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3150
3151 For compatibility with Perl, \Q...\E sequences are recognized inside character
3152 classes as well as outside them.
3153
3154 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3155 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3156 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3157 signed/unsigned warnings.
3158
3159 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3160 option for pcretest, so I've replaced it by a simple function that does just
3161 that job.
3162
3163 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3164 "pcregrep -".
3165
3166 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3167 Java package. This provides some syntactic sugar for simple cases of what my
3168 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3169 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3170 item, you can use this simplified notation. Note that only makes sense with
3171 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3172 greediness, whatever the setting of the PCRE_UNGREEDY option.
3173
3174 13. A change of greediness default within a pattern was not taking effect at
3175 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3176 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3177 was abstracted outside.
3178
3179 14. PCRE now supports the \G assertion. It is true when the current matching
3180 position is at the start point of the match. This differs from \A when the
3181 starting offset is non-zero. Used with the /g option of pcretest (or similar
3182 code), it works in the same way as it does for Perl's /g option. If all
3183 alternatives of a regex begin with \G, the expression is anchored to the start
3184 match position, and the "anchored" flag is set in the compiled expression.
3185
3186 15. Some bugs concerning the handling of certain option changes within patterns
3187 have been fixed. These applied to options other than (?ims). For example,
3188 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3189 been the other way round. Some of this was related to change 7 above.
3190
3191 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3192 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3193 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3194 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3195
3196 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3197 mode. Unlike ".", it always matches newline, whatever the setting of
3198 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3199 assertions. Perl allows it, but it doesn't (in general) work because it can't
3200 calculate the length of the lookbehind. At least, that's the case for Perl
3201 5.8.0 - I've been told they are going to document that it doesn't work in
3202 future.
3203
3204 18. Added an error diagnosis for escapes that PCRE does not support: these are
3205 \L, \l, \N, \P, \p, \U, \u, and \X.
3206
3207 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3208 reading past the end of the pattern in cases such as /[abcd/.
3209
3210 20. PCRE was getting more memory than necessary for patterns with classes that
3211 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3212
3213 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3214 compiling PCRE for use with Virtual Pascal.
3215
3216 22. Small fix to the Makefile to make it work properly if the build is done
3217 outside the source tree.
3218
3219 23. Added a new extension: a condition to go with recursion. If a conditional
3220 subpattern starts with (?(R) the "true" branch is used if recursion has
3221 happened, whereas the "false" branch is used only at the top level.
3222
3223 24. When there was a very long string of literal characters (over 255 bytes
3224 without UTF support, over 250 bytes with UTF support), the computation of how
3225 much memory was required could be incorrect, leading to segfaults or other
3226 strange effects.
3227
3228 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3229 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3230 there was a subsequent back reference to those brackets. This meant that, for
3231 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3232 possible to check for precisely this case. All we can do is abandon the
3233 optimization if .* occurs inside capturing brackets when there are any back
3234 references whatsoever. (See below for a better fix that came later.)
3235
3236 26. The handling of the optimization for finding the first character of a
3237 non-anchored pattern, and for finding a character that is required later in the
3238 match were failing in some cases. This didn't break the matching; it just
3239 failed to optimize when it could. The way this is done has been re-implemented.
3240
3241 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3242
3243 28. Added a new feature that provides some of the functionality that Perl
3244 provides with (?{...}). The facility is termed a "callout". The way it is done
3245 in PCRE is for the caller to provide an optional function, by setting
3246 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3247 global variable. By default it is unset, which disables all calling out. To get
3248 the function called, the regex must include (?C) at appropriate points. This
3249 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3250 This provides a means of identifying different callout points. When PCRE
3251 reaches such a point in the regex, if pcre_callout has been set, the external
3252 function is called. It is provided with data in a structure called
3253 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3254 matching continues; if it returns a non-zero value, the match at the current
3255 point fails. However, backtracking will occur if possible. [This was changed
3256 later and other features added - see item 49 below.]
3257
3258 29. pcretest is upgraded to test the callout functionality. It provides a
3259 callout function that displays information. By default, it shows the start of
3260 the match and the current position in the text. There are some new data escapes
3261 to vary what happens:
3262
3263 \C+ in addition, show current contents of captured substrings
3264 \C- do not supply a callout function
3265 \C!n return 1 when callout number n is reached
3266 \C!n!m return 1 when callout number n is reached for the mth time
3267
3268 30. If pcregrep was called with the -l option and just a single file name, it
3269 output "<stdin>" if a match was found, instead of the file name.
3270
3271 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3272 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3273 pcre_exec(). This saves a malloc/free per call. The default value of
3274 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3275 when configuring.
3276
3277 32. The default maximum size of a compiled pattern is 64K. There have been a
3278 few cases of people hitting this limit. The code now uses macros to handle the
3279 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3280 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3281 configuring. Tests 2 and 5 work only with 2-byte links because they output
3282 debugging information about compiled patterns.
3283
3284 33. Internal code re-arrangements:
3285
3286 (a) Moved the debugging function for printing out a compiled regex into
3287 its own source file (printint.c) and used #include to pull it into
3288 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3289 separate copies.
3290
3291 (b) Defined the list of op-code names for debugging as a macro in
3292 internal.h so that it is next to the definition of the opcodes.
3293
3294 (c) Defined a table of op-code lengths for simpler skipping along compiled
3295 code. This is again a macro in internal.h so that it is next to the
3296 definition of the opcodes.
3297
3298 34. Added support for recursive calls to individual subpatterns, along the
3299 lines of Robin Houston's patch (but implemented somewhat differently).
3300
3301 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3302 allow it to read and process whole directories in Win32. This code was
3303 contributed by Lionel Fourquaux; it has not been tested by me.
3304
3305 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3306 used to name a group. Names consist of alphanumerics and underscores, and must
3307 be unique. Back references use the syntax (?P=name) and recursive calls use
3308 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3309 numbers. The function pcre_fullinfo() can be used after compilation to extract
3310 a name/number map. There are three relevant calls:
3311
3312 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3313 PCRE_INFO_NAMECOUNT yields the number of entries
3314 PCRE_INFO_NAMETABLE yields a pointer to the map.
3315
3316 The map is a vector of fixed-size entries. The size of each entry depends on
3317 the length of the longest name used. The first two bytes of each entry are the
3318 group number, most significant byte first. There follows the corresponding
3319 name, zero terminated. The names are in alphabetical order.
3320
3321 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3322 case instead of 255. Making it the same both with and without UTF-8 support
3323 means that the same test output works with both.
3324
3325 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3326 calling malloc() with a zero argument.
3327
3328 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3329 optimization. I've improved things by keeping a bitmap of backreferences with
3330 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3331 fact referenced, the optimization can be applied. It is unlikely that a
3332 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3333 the match to follow \n) will appear inside brackets with a number greater than
3334 31, but if it does, any back reference > 31 suppresses the optimization.
3335
3336 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3337 of disabling numbered capturing parentheses. Any opening parenthesis that is
3338 not followed by ? behaves as if it were followed by ?: but named parentheses
3339 can still be used for capturing (and they will acquire numbers in the usual
3340 way).
3341
3342 41. Redesigned the return codes from the match() function into yes/no/error so
3343 that errors can be passed back from deep inside the nested calls. A malloc
3344 failure while inside a recursive subpattern call now causes the
3345 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3346
3347 42. It is now possible to set a limit on the number of times the match()
3348 function is called in a call to pcre_exec(). This facility makes it possible to
3349 limit the amount of recursion and backtracking, though not in a directly
3350 obvious way, because the match() function is used in a number of different
3351 circumstances. The count starts from zero for each position in the subject
3352 string (for non-anchored patterns). The default limit is, for compatibility, a
3353 large number, namely 10 000 000. You can change this in two ways:
3354
3355 (a) When configuring PCRE before making, you can use --with-match-limit=n
3356 to set a default value for the compiled library.
3357
3358 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3359 a different value is set. See 45 below.
3360
3361 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3362
3363 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3364 of things that can be changed at compile time. The first argument specifies
3365 what is wanted and the second points to where the information is to be placed.
3366 The current list of available information is:
3367
3368 PCRE_CONFIG_UTF8
3369
3370 The output is an integer that is set to one if UTF-8 support is available;
3371 otherwise it is set to zero.
3372
3373 PCRE_CONFIG_NEWLINE
3374
3375 The output is an integer that it set to the value of the code that is used for
3376 newline. It is either LF (10) or CR (13).
3377
3378 PCRE_CONFIG_LINK_SIZE
3379
3380 The output is an integer that contains the number of bytes used for internal
3381 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3382
3383 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3384
3385 The output is an integer that contains the threshold above which the POSIX
3386 interface uses malloc() for output vectors. See item 31 above.
3387
3388 PCRE_CONFIG_MATCH_LIMIT
3389
3390 The output is an unsigned integer that contains the default limit of the number
3391 of match() calls in a pcre_exec() execution. See 42 above.
3392
3393 44. pcretest has been upgraded by the addition of the -C option. This causes it
3394 to extract all the available output from the new pcre_config() function, and to
3395 output it. The program then exits immediately.
3396
3397 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3398 order to support additional features. One way would have been to define
3399 pcre_exec2() (for example) with extra arguments, but this would not have been
3400 extensible, and would also have required all calls to the original function to
3401 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3402 is used for passing in "extra" data from pcre_study().
3403
3404 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3405 contains the following fields:
3406
3407 flags a bitmap indicating which of the following fields are set
3408 study_data opaque data from pcre_study()
3409 match_limit a way of specifying a limit on match() calls for a specific
3410 call to pcre_exec()
3411 callout_data data for callouts (see 49 below)
3412
3413 The flag bits are also defined in pcre.h, and are
3414
3415 PCRE_EXTRA_STUDY_DATA
3416 PCRE_EXTRA_MATCH_LIMIT
3417 PCRE_EXTRA_CALLOUT_DATA
3418
3419 The pcre_study() function now returns one of these new pcre_extra blocks, with
3420 the actual study data pointed to by the study_data field, and the
3421 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3422 before. That is, this change is entirely upwards-compatible and requires no
3423 change to existing code.
3424
3425 If you want to pass in additional data to pcre_exec(), you can either place it
3426 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3427 block.
3428
3429 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3430 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3431 times with different match limits, until it finds the minimum value needed for
3432 pcre_exec() to complete. The value is then output. This can be instructive; for
3433 most simple matches the number is quite small, but for pathological cases it
3434 gets very large very quickly.
3435
3436 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3437 returns the size of the data block pointed to by the study_data field in a
3438 pcre_extra block, that is, the value that was passed as the argument to
3439 pcre_malloc() when PCRE was getting memory in which to place the information
3440 created by pcre_study(). The fourth argument should point to a size_t variable.
3441 pcretest has been extended so that this information is shown after a successful
3442 pcre_study() call when information about the compiled regex is being displayed.
3443
3444 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3445 because what follows is always an absolute path. (Later: it turns out that this
3446 is more than cosmetic for MinGW, because it doesn't like empty path
3447 components.)
3448
3449 49. Some changes have been made to the callout feature (see 28 above):
3450
3451 (i) A callout function now has three choices for what it returns:
3452
3453 0 => success, carry on matching
3454 > 0 => failure at this point, but backtrack if possible
3455 < 0 => serious error, return this value from pcre_exec()
3456
3457 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3458 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3459 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3460 use by callout functions. It will never be used by PCRE itself.
3461
3462 (ii) The pcre_extra structure (see 45 above) has a void * field called
3463 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3464 pcre_callout_block structure has a field of the same name. The contents of
3465 the field passed in the pcre_extra structure are passed to the callout
3466 function in the corresponding field in the callout block. This makes it
3467 easier to use the same callout-containing regex from multiple threads. For
3468 testing, the pcretest program has a new data escape
3469
3470 \C*n pass the number n (may be negative) as callout_data
3471
3472 If the callout function in pcretest receives a non-zero value as
3473 callout_data, it returns that value.
3474
3475 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3476 there were some redundant $(CFLAGS) in commands that are now specified as
3477 $(LINK), which already includes $(CFLAGS).
3478
3479 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3480 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3481 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3482 one-byte characters throughout. Note that case-insensitive matching applies
3483 only to characters whose values are less than 256. PCRE doesn't support the
3484 notion of cases for higher-valued characters.
3485
3486 (i) A character class whose characters are all within 0-255 is handled as
3487 a bit map, and the map is inverted for negative classes. Previously, a
3488 character > 255 always failed to match such a class; however it should
3489 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3490
3491 (ii) A negated character class with a single character < 255 is coded as
3492 "not this character" (OP_NOT). This wasn't working properly when the test
3493 character was multibyte, either singly or repeated.
3494
3495 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3496 mode, for example: \x{100}{2,3}.
3497
3498 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3499 singly or repeated) now correctly test multibyte characters. However,
3500 PCRE doesn't recognize any characters with values greater than 255 as
3501 digits, spaces, or word characters. Such characters always match \D, \S,
3502 and \W, and never match \d, \s, or \w.
3503
3504 (v) Classes may now contain characters and character ranges with values
3505 greater than 255. For example: [ab\x{100}-\x{400}].
3506
3507 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3508 PCRE in UTF-8 mode.
3509
3510 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3511 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3512 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3513 value.)
3514
3515 53. The single man page has become too large. I have therefore split it up into
3516 a number of separate man pages. These also give rise to individual HTML pages;
3517 these are now put in a separate directory, and there is an index.html page that
3518 lists them all. Some hyperlinking between the pages has been installed.
3519
3520 54. Added convenience functions for handling named capturing parentheses.
3521
3522 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3523 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3524 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3525 are faulted.
3526
3527 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3528 calling configure. These values are used when compiling the dftables.c program
3529 which is run to generate the source of the default character tables. They
3530 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3531 you will need to set these values.
3532
3533 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3534
3535
3536 Version 3.9 02-Jan-02
3537 ---------------------
3538
3539 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3540
3541 2. If --disable-static was given, the building process failed when trying to
3542 build pcretest and pcregrep. (For some reason it was using libtool to compile
3543 them, which is not right, as they aren't part of the library.)
3544
3545
3546 Version 3.8 18-Dec-01
3547 ---------------------
3548
3549 1. The experimental UTF-8 code was completely screwed up. It was packing the
3550 bytes in the wrong order. How dumb can you get?
3551
3552
3553 Version 3.7 29-Oct-01
3554 ---------------------
3555
3556 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3557 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3558 this didn't happen under Solaris 8, where I normally test things.
3559
3560 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3561 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3562 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3563 this fix an hour or so after the initial 3.7 release.)
3564
3565
3566 Version 3.6 23-Oct-01
3567 ---------------------
3568
3569 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3570 offsets passed as NULL with zero offset count.
3571
3572 2. The config.guess and config.sub files had not been updated when I moved to
3573 the latest autoconf.
3574
3575
3576 Version 3.5 15-Aug-01
3577 ---------------------
3578
3579 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3580 had been forgotten.
3581
3582 2. By using declared but undefined structures, we can avoid using "void"
3583 definitions in pcre.h while keeping the internal definitions of the structures
3584 private.
3585
3586 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3587 user point of view, this means that both static and shared libraries are built
3588 by default, but this can be individually controlled. More of the work of
3589 handling this static/shared cases is now inside libtool instead of PCRE's make
3590 file.
3591
3592 4. The pcretest utility is now installed along with pcregrep because it is
3593 useful for users (to test regexs) and by doing this, it automatically gets
3594 relinked by libtool. The documentation has been turned into a man page, so
3595 there are now .1, .txt, and .html versions in /doc.
3596
3597 5. Upgrades to pcregrep:
3598 (i) Added long-form option names like gnu grep.
3599 (ii) Added --help to list all options with an explanatory phrase.
3600 (iii) Added -r, --recursive to recurse into sub-directories.
3601 (iv) Added -f, --file to read patterns from a file.
3602
3603 6. pcre_exec() was referring to its "code" argument before testing that
3604 argument for NULL (and giving an error if it was NULL).
3605
3606 7. Upgraded Makefile.in to allow for compiling in a different directory from
3607 the source directory.
3608
3609 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3610 options bits, the pointer it was passed was to an int instead of to an unsigned
3611 long int. This mattered only on 64-bit systems.
3612
3613 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3614 generated) instead of pcre.in, which it its source. Also made the same change
3615 in several of the .c files.
3616
3617 10. A new release of gcc defines printf() as a macro, which broke pcretest
3618 because it had an ifdef in the middle of a string argument for printf(). Fixed
3619 by using separate calls to printf().
3620
3621 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3622 script, to force use of CR or LF instead of \n in the source. On non-Unix
3623 systems, the value can be set in config.h.
3624
3625 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3626 absolute limit. Changed the text of the error message to make this clear, and
3627 likewise updated the man page.
3628
3629 13. The limit of 99 on the number of capturing subpatterns has been removed.
3630 The new limit is 65535, which I hope will not be a "real" limit.
3631
3632
3633 Version 3.4 22-Aug-00
3634 ---------------------
3635
3636 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3637
3638 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3639
3640
3641 Version 3.3 01-Aug-00
3642 ---------------------
3643
3644 1. If an octal character was given, but the value was greater than \377, it
3645 was not getting masked to the least significant bits, as documented. This could
3646 lead to crashes in some systems.
3647
3648 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3649 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3650
3651 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3652 These just pass their arguments on to (pcre_free)(), but they are provided
3653 because some uses of PCRE bind it to non-C systems that can call its functions,
3654 but cannot call free() or pcre_free() directly.
3655
3656 4. Add "make test" as a synonym for "make check". Corrected some comments in
3657 the Makefile.
3658
3659 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3660 Makefile.
3661
3662 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3663 command called pgrep for grepping around the active processes.
3664
3665 7. Added the beginnings of support for UTF-8 character strings.
3666
3667 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3668 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3669 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3670 out for the ar command.)
3671
3672
3673 Version 3.2 12-May-00
3674 ---------------------
3675
3676 This is purely a bug fixing release.
3677
3678 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3679 of ZA. This was just one example of several cases that could provoke this bug,
3680 which was introduced by change 9 of version 2.00. The code for breaking
3681 infinite loops after an iteration that matches an empty string was't working
3682 correctly.
3683
3684 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3685 when matched against abbab (for example). After matching an empty string, it
3686 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3687 caused it to match further down the string than it should.
3688
3689 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3690 was there because it doesn't seem to be needed, and it causes trouble on some
3691 systems, as it is not a Standard C header. It has been removed.
3692
3693 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3694 were reported on the Macintosh. The changes were from
3695
3696 while ((c = *(++ptr)) != 0 && c != '\n');
3697 to
3698 while ((c = *(++ptr)) != 0 && c != '\n') ;
3699
3700 Totally extraordinary, but if that's what it takes...
3701
3702 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3703 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3704 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3705 assumes the way PCRE uses memmove() (always moving upwards).
3706
3707 6. PCRE is being used in one environment where strchr() is not available. There
3708 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3709 faster code anyway.
3710
3711
3712 Version 3.1 09-Feb-00
3713 ---------------------
3714
3715 The only change in this release is the fixing of some bugs in Makefile.in for
3716 the "install" target:
3717
3718 (1) It was failing to install pcreposix.h.
3719
3720 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3721
3722
3723 Version 3.0 01-Feb-00
3724 ---------------------
3725
3726 1. Add support for the /+ modifier to perltest (to output $` like it does in
3727 pcretest).
3728
3729 2. Add support for the /g modifier to perltest.
3730
3731 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3732 matches null strings.
3733
3734 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3735 pattern. Perl treats empty patterns specially - it reuses the most recent
3736 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3737 effect.
3738
3739 5. The POSIX interface was broken in that it was just handing over the POSIX
3740 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3741 required a bigger vector, with some working space on the end. This means that
3742 the POSIX wrapper now has to get and free some memory, and copy the results.
3743
3744 6. Added some simple autoconf support, placing the test data and the
3745 documentation in separate directories, re-organizing some of the
3746 information files, and making it build pcre-config (a GNU standard). Also added
3747 libtool support for building PCRE as a shared library, which is now the
3748 default.
3749
3750 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3751 09 are not valid octal constants. Single digits will be used for minor values
3752 less than 10.
3753
3754 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3755 existing programs that set these in the POSIX interface can use PCRE without
3756 modification.
3757
3758 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3759 return all that pcre_info() returns, plus additional data. The pcre_info()
3760 function is retained for compatibility, but is considered to be obsolete.
3761
3762 10. Added experimental recursion feature (?R) to handle one common case that
3763 Perl 5.6 will be able to do with (?p{...}).
3764
3765 11. Added support for POSIX character classes like [:alpha:], which Perl is
3766 adopting.
3767
3768
3769 Version 2.08 31-Aug-99
3770 ----------------------
3771
3772 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3773 trying to match at the startoffset position, but instead was moving forward to
3774 the next newline as if a previous match had failed.
3775
3776 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3777 and could get into a loop if a null string was matched other than at the start
3778 of the subject.
3779
3780 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3781 be distinguished at compile time, and for completeness also added PCRE_DATE.
3782
3783 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3784 in GnuWin32 environments.
3785
3786
3787 Version 2.07 29-Jul-99
3788 ----------------------
3789
3790 1. The documentation is now supplied in plain text form and HTML as well as in
3791 the form of man page sources.
3792
3793 2. C++ compilers don't like assigning (void *) values to other pointer types.
3794 In particular this affects malloc(). Although there is no problem in Standard
3795 C, I've put in casts to keep C++ compilers happy.
3796
3797 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3798 should be (const char *).
3799
3800 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3801 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3802 However, I haven't made this a standard facility. The documentation doesn't
3803 mention it, and the Makefile doesn't support it.
3804
3805 5. The Makefile now contains an "install" target, with editable destinations at
3806 the top of the file. The pcretest program is not installed.
3807
3808 6. pgrep -V now gives the PCRE version number and date.
3809
3810 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3811 causing the entire string to be ignored, instead of just the last character.
3812
3813 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3814 non-matching string, it can take a very, very long time, even for strings of
3815 quite modest length, because of the nested recursion. PCRE now does better in
3816 some of these cases. It does this by remembering the last required literal
3817 character in the pattern, and pre-searching the subject to ensure it is present
3818 before running the real match. In other words, it applies a heuristic to detect
3819 some types of certain failure quickly, and in the above example, if presented
3820 with a string that has no trailing " it gives "no match" very quickly.
3821
3822 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3823 other alternatives are tried instead.
3824
3825
3826 Version 2.06 09-Jun-99
3827 ----------------------
3828
3829 1. Change pcretest's output for amount of store used to show just the code
3830 space, because the remainder (the data block) varies in size between 32-bit and
3831 64-bit systems.
3832
3833 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3834 start matching at. This allows lookbehinds to work when searching for multiple
3835 occurrences in a string.
3836
3837 3. Added additional options to pcretest for testing multiple occurrences:
3838
3839 /+ outputs the rest of the string that follows a match
3840 /g loops for multiple occurrences, using the new startoffset argument
3841 /G loops for multiple occurrences by passing an incremented pointer
3842
3843 4. PCRE wasn't doing the "first character" optimization for patterns starting
3844 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3845 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3846 the letter 'x'. On long subject strings, this gives a significant speed-up.
3847
3848
3849 Version 2.05 21-Apr-99
3850 ----------------------
3851
3852 1. Changed the type of magic_number from int to long int so that it works
3853 properly on 16-bit systems.
3854
3855 2. Fixed a bug which caused patterns starting with .* not to work correctly
3856 when the subject string contained newline characters. PCRE was assuming
3857 anchoring for such patterns in all cases, which is not correct because .* will
3858 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3859 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3860 must be retried after every newline in the subject.
3861
3862
3863 Version 2.04 18-Feb-99
3864 ----------------------
3865
3866 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3867 computation of the store needed to hold the pattern was incorrect (too large).
3868 If such patterns were nested a few deep, this could multiply and become a real
3869 problem.
3870
3871 2. Added /M option to pcretest to show the memory requirement of a specific
3872 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3873
3874 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3875 compiled in such a way that the backtracking after subsequent failure was
3876 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3877 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3878
3879
3880 Version 2.03 02-Feb-99
3881 ----------------------
3882
3883 1. Fixed typo and small mistake in man page.
3884
3885 2. Added 4th condition (GPL supersedes if conflict) and created separate
3886 LICENCE file containing the conditions.
3887
3888 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3889 Perl, that is the internal \ allows the delimiter to be included in the
3890 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3891 the final delimiter, add \ to the end of the pattern (to test the error).
3892
3893 4. Added the convenience functions for extracting substrings after a successful
3894 match. Updated pcretest to make it able to test these functions.
3895
3896
3897 Version 2.02 14-Jan-99
3898 ----------------------
3899
3900 1. Initialized the working variables associated with each extraction so that
3901 their saving and restoring doesn't refer to uninitialized store.
3902
3903 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3904 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3905 fix the problem.
3906
3907 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3908 calls, and wasn't printing the correct value for compiling calls. Increased the
3909 default value of LOOPREPEAT, and the number of significant figures in the
3910 times.
3911
3912 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3913
3914 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3915 a building problem on Windows NT with a FAT file system.
3916
3917
3918 Version 2.01 21-Oct-98
3919 ----------------------
3920
3921 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3922 to character tables built by pcre_maketables() in the current locale. If NULL
3923 is passed, the default tables are used.
3924
3925
3926 Version 2.00 24-Sep-98
3927 ----------------------
3928
3929 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3930 it any more.
3931
3932 2. Allow quantification of (?>) groups, and make it work correctly.
3933
3934 3. The first character computation wasn't working for (?>) groups.
3935
3936 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3937 end of the subject) and add 5.005's \z, which really does match only at the
3938 very end of the subject.
3939
3940 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3941
3942 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3943 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3944 localized options. All options to pcre_study() were also removed.
3945
3946 7. Add other new features from 5.005:
3947
3948 $(?<= positive lookbehind
3949 $(?<! negative lookbehind
3950 (?imsx-imsx) added the unsetting capability
3951 such a setting is global if at outer level; local otherwise
3952 (?imsx-imsx:) non-capturing groups with option setting
3953 (?(cond)re|re) conditional pattern matching
3954
3955 A backreference to itself in a repeated group matches the previous
3956 captured string.
3957
3958 8. General tidying up of studying (both automatic and via "study")
3959 consequential on the addition of new assertions.
3960
3961 9. As in 5.005, unlimited repeated groups that could match an empty substring
3962 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3963 runtime if any iteration does actually match an empty substring.
3964
3965 10. Include the RunTest script in the distribution.
3966
3967 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3968 discrepancies, some of which were old and were also with respect to 5.004. They
3969 have now been fixed.
3970
3971
3972 Version 1.09 28-Apr-98
3973 ----------------------
3974
3975 1. A negated single character class followed by a quantifier with a minimum
3976 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3977 program crashes, or just wrong answers. This did not apply to negated classes
3978 containing more than one character, or to minima other than one.
3979
3980
3981 Version 1.08 27-Mar-98
3982 ----------------------
3983
3984 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3985
3986 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3987 latter must appear before anything that relies on it in the pattern.
3988
3989
3990 Version 1.07 16-Feb-98
3991 ----------------------
3992
3993 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3994 repeat of a potentially empty string).
3995
3996
3997 Version 1.06 23-Jan-98
3998 ----------------------
3999
4000 1. Added Markus Oberhumer's little patches for C++.
4001
4002 2. Literal strings longer than 255 characters were broken.
4003
4004
4005 Version 1.05 23-Dec-97
4006 ----------------------
4007
4008 1. Negated character classes containing more than one character were failing if
4009 PCRE_CASELESS was set at run time.
4010
4011
4012 Version 1.04 19-Dec-97
4013 ----------------------
4014
4015 1. Corrected the man page, where some "const" qualifiers had been omitted.
4016
4017 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4018 input syntax.
4019
4020 3. Fixed memory leak which occurred when a regex with back references was
4021 matched with an offsets vector that wasn't big enough. The temporary memory
4022 that is used in this case wasn't being freed if the match failed.
4023
4024 4. Tidied pcretest to ensure it frees memory that it gets.
4025
4026 5. Temporary memory was being obtained in the case where the passed offsets
4027 vector was exactly big enough.
4028
4029 6. Corrected definition of offsetof() from change 5 below.
4030
4031 7. I had screwed up change 6 below and broken the rules for the use of
4032 setjmp(). Now fixed.
4033
4034
4035 Version 1.03 18-Dec-97
4036 ----------------------
4037
4038 1. A erroneous regex with a missing opening parenthesis was correctly
4039 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4040 on some systems.
4041
4042 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4043 it was reported that one broken compiler failed on the former because "code" is
4044 also an independent variable.
4045
4046 3. The erroneous regex a[]b caused an array overrun reference.
4047
4048 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4049 fail on data ending with that character. (It was going on too far, and checking
4050 the next character, typically a binary zero.) This was specific to the
4051 optimized code for single-character negative classes.
4052
4053 5. Added a contributed patch from the TIN world which does the following:
4054
4055 + Add an undef for memmove, in case the the system defines a macro for it.
4056
4057 + Add a definition of offsetof(), in case there isn't one. (I don't know
4058 the reason behind this - offsetof() is part of the ANSI standard - but
4059 it does no harm).
4060
4061 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4062 most of the places where whitespace preceded '#'. I have given up and
4063 allowed the remaining 2 cases to be at the margin.
4064
4065 + Rename some variables in pcre to eliminate shadowing. This seems very
4066 pedantic, but does no harm, of course.
4067
4068 6. Moved the call to setjmp() into its own function, to get rid of warnings
4069 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4070
4071 7. Constructs such as \d{8,} were compiling into the equivalent of
4072 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4073 outcome, but in this particular case used more store than had been allocated,
4074 which caused the bug to be discovered because it threw up an internal error.
4075
4076 8. The debugging code in both pcre and pcretest for outputting the compiled
4077 form of a regex was going wrong in the case of back references followed by
4078 curly-bracketed repeats.
4079
4080
4081 Version 1.02 12-Dec-97
4082 ----------------------
4083
4084 1. Typos in pcre.3 and comments in the source fixed.
4085
4086 2. Applied a contributed patch to get rid of places where it used to remove
4087 'const' from variables, and fixed some signed/unsigned and uninitialized
4088 variable warnings.
4089
4090 3. Added the "runtest" target to Makefile.
4091
4092 4. Set default compiler flag to -O2 rather than just -O.
4093
4094
4095 Version 1.01 19-Nov-97
4096 ----------------------
4097
4098 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4099 like /([ab]*)*/, that is, for classes with more than one character in them.
4100
4101 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4102 as /((?>a*))*/ (a PCRE_EXTRA facility).
4103
4104
4105 Version 1.00 18-Nov-97
4106 ----------------------
4107
4108 1. Added compile-time macros to support systems such as SunOS4 which don't have
4109 memmove() or strerror() but have other things that can be used instead.
4110
4111 2. Arranged that "make clean" removes the executables.
4112
4113
4114 Version 0.99 27-Oct-97
4115 ----------------------
4116
4117 1. Fixed bug in code for optimizing classes with only one character. It was
4118 initializing a 32-byte map regardless, which could cause it to run off the end
4119 of the memory it had got.
4120
4121 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4122
4123
4124 Version 0.98 22-Oct-97
4125 ----------------------
4126
4127 1. Fixed bug in code for handling temporary memory usage when there are more
4128 back references than supplied space in the ovector. This could cause segfaults.
4129
4130
4131 Version 0.97 21-Oct-97
4132 ----------------------
4133
4134 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4135
4136 2. Optimized negated single characters not to use a bit map.
4137
4138 3. Brought error texts together as macro definitions; clarified some of them;
4139 fixed one that was wrong - it said "range out of order" when it meant "invalid
4140 escape sequence".
4141
4142 4. Changed some char * arguments to const char *.
4143
4144 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4145
4146 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4147 pcretest.
4148
4149
4150 Version 0.96 16-Oct-97
4151 ----------------------
4152
4153 1. Added a simple "pgrep" utility to the distribution.
4154
4155 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4156 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4157 where "ddd" means "one or more decimal digits".
4158
4159 3. Fixed serious bug. If a pattern had a back reference, but the call to
4160 pcre_exec() didn't supply a large enough ovector to record the related
4161 identifying subpattern, the match always failed. PCRE now remembers the number
4162 of the largest back reference, and gets some temporary memory in which to save
4163 the offsets during matching if necessary, in order to ensure that
4164 backreferences always work.
4165
4166 4. Increased the compatibility with Perl in a number of ways:
4167
4168 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4169 to request this handling. The option can be set at compile or exec time.
4170
4171 (b) $ matches before a terminating newline by default; an option
4172 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4173 mode). The option can be set at compile or exec time.
4174
4175 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4176 the same as Perl's. If the decimal number it represents is less than 10
4177 or there aren't that many previous left capturing parentheses, an octal
4178 escape is read. Inside a character class, it's always an octal escape,
4179 even if it is a single digit.
4180
4181 (d) An escaped but undefined alphabetic character is taken as a literal,
4182 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4183 escapes.
4184
4185 (e) {0} is now permitted. (The previous item is removed from the compiled
4186 pattern).
4187
4188 5. Changed all the names of code files so that the basic parts are no longer
4189 than 10 characters, and abolished the teeny "globals.c" file.
4190
4191 6. Changed the handling of character classes; they are now done with a 32-byte
4192 bit map always.
4193
4194 7. Added the -d and /D options to pcretest to make it possible to look at the
4195 internals of compilation without having to recompile pcre.
4196
4197
4198 Version 0.95 23-Sep-97
4199 ----------------------
4200
4201 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4202 \x20 at the start of a run of normal characters. These were being treated as
4203 real characters, instead of the source characters being re-checked.
4204
4205
4206 Version 0.94 18-Sep-97
4207 ----------------------
4208
4209 1. The functions are now thread-safe, with the caveat that the global variables
4210 containing pointers to malloc() and free() or alternative functions are the
4211 same for all threads.
4212
4213 2. Get pcre_study() to generate a bitmap of initial characters for non-
4214 anchored patterns when this is possible, and use it if passed to pcre_exec().
4215
4216
4217 Version 0.93 15-Sep-97
4218 ----------------------
4219
4220 1. /(b)|(:+)/ was computing an incorrect first character.
4221
4222 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4223 but not actually doing anything yet.
4224
4225 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4226 as Perl does (e.g. [-az] or [az-]).
4227
4228 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4229 all possible positions.
4230
4231 5. Split up into different modules to avoid including unneeded functions in a
4232 compiled binary. However, compile and exec are still in one module. The "study"
4233 function is split off.
4234
4235 6. The character tables are now in a separate module whose source is generated
4236 by an auxiliary program - but can then be edited by hand if required. There are
4237 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4238 toupper() in the code.
4239
4240 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4241 make them global. Abolish the function for setting them, as the caller can now
4242 set them directly.
4243
4244
4245 Version 0.92 11-Sep-97
4246 ----------------------
4247
4248 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4249 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4250
4251 2. Caseless matching was not working in character classes if the characters in
4252 the pattern were in upper case.
4253
4254 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4255
4256 4. Make PCRE_ANCHORED public and accept as a compile option.
4257
4258 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4259 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4260 pass them.
4261
4262 6. Give an error if bad option bits passed at compile or run time.
4263
4264 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4265 pcretest to cause it to pass that flag.
4266
4267 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4268 options, and the first character, if set.
4269
4270 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4271
4272
4273 Version 0.91 10-Sep-97
4274 ----------------------
4275
4276 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4277 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4278
4279 2. PCRE was looping on encountering an indefinitely repeated back reference to
4280 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4281 Perl does - treats the match as successful.
4282
4283 ****

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5