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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 641 - (show annotations) (download)
Mon Jul 25 16:56:54 2011 UTC (3 years ago) by ph10
File size: 208879 byte(s)
Add minix to list of OS not supporting -S in pcretest.

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 30. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
163 compiled something random. Now it gives a compile-time error (as does
164 Perl).
165
166 31. A *MARK encountered during the processing of a positive assertion is now
167 recorded and passed back (compatible with Perl).
168
169 32. If --only-matching or --colour was set on a pcregrep call whose pattern
170 had alternative anchored branches, the search for a second match in a line
171 was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
172 matched the line "0102" twice. The same bug affected patterns that started
173 with a backwards assertion. For example /\b01|\b02/ also matched "0102"
174 twice.
175
176 33. Previously, PCRE did not allow quantification of assertions. However, Perl
177 does, and because of capturing effects, quantifying parenthesized
178 assertions may at times be useful. Quantifiers are now allowed for
179 parenthesized assertions.
180
181 34. A minor code tidy in pcre_compile() when checking options for \R usage.
182
183 35. \g was being checked for fancy things in a character class, when it should
184 just be a literal "g".
185
186 36. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
187 appearance of a nested POSIX class supersedes an apparent external class.
188 For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
189 unescaped square brackets may also appear as part of class names. For
190 example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
191 more like Perl.
192
193 37. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
194 was because it thought it was \N{name}, which is not supported).
195
196 38. Add minix to OS list not supporting the -S option in pcretest.
197
198
199 Version 8.12 15-Jan-2011
200 ------------------------
201
202 1. Fixed some typos in the markup of the man pages, and wrote a script that
203 checks for such things as part of the documentation building process.
204
205 2. On a big-endian 64-bit system, pcregrep did not correctly process the
206 --match-limit and --recursion-limit options (added for 8.11). In
207 particular, this made one of the standard tests fail. (The integer value
208 went into the wrong half of a long int.)
209
210 3. If the --colour option was given to pcregrep with -v (invert match), it
211 did strange things, either producing crazy output, or crashing. It should,
212 of course, ignore a request for colour when reporting lines that do not
213 match.
214
215 4. Another pcregrep bug caused similar problems if --colour was specified with
216 -M (multiline) and the pattern match finished with a line ending.
217
218 5. In pcregrep, when a pattern that ended with a literal newline sequence was
219 matched in multiline mode, the following line was shown as part of the
220 match. This seems wrong, so I have changed it.
221
222 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
223 the check for further matches in the same line (so they could be coloured)
224 to overrun the end of the current line. If another match was found, it was
225 incorrectly shown (and then shown again when found in the next line).
226
227 7. If pcregrep was compiled under Windows, there was a reference to the
228 function pcregrep_exit() before it was defined. I am assuming this was
229 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
230 reported by a user. I've moved the definition above the reference.
231
232
233 Version 8.11 10-Dec-2010
234 ------------------------
235
236 1. (*THEN) was not working properly if there were untried alternatives prior
237 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
238 backtracked to try for "b" instead of moving to the next alternative branch
239 at the same level (in this case, to look for "c"). The Perl documentation
240 is clear that when (*THEN) is backtracked onto, it goes to the "next
241 alternative in the innermost enclosing group".
242
243 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
244 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
245 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
246 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
247 (*THEN).
248
249 3. If \s appeared in a character class, it removed the VT character from
250 the class, even if it had been included by some previous item, for example
251 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
252 of \s, but is part of the POSIX "space" class.)
253
254 4. A partial match never returns an empty string (because you can always
255 match an empty string at the end of the subject); however the checking for
256 an empty string was starting at the "start of match" point. This has been
257 changed to the "earliest inspected character" point, because the returned
258 data for a partial match starts at this character. This means that, for
259 example, /(?<=abc)def/ gives a partial match for the subject "abc"
260 (previously it gave "no match").
261
262 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
263 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
264 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
265 has an implication that the given string is incomplete (because a partial
266 match is preferred over a full match). For this reason, these items now
267 give a partial match in this situation. [Aside: previously, the one case
268 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
269 match rather than a full match, which was wrong by the old rules, but is
270 now correct.]
271
272 6. There was a bug in the handling of #-introduced comments, recognized when
273 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
274 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
275 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
276 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
277 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
278 places in pcre_compile().
279
280 7. Related to (6) above, when pcre_compile() was skipping #-introduced
281 comments when looking ahead for named forward references to subpatterns,
282 the only newline sequence it recognized was NL. It now handles newlines
283 according to the set newline convention.
284
285 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
286 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
287 cater for a lack of strerror(). These oversights have been fixed.
288
289 9. Added --match-limit and --recursion-limit to pcregrep.
290
291 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
292
293 11. When the -o option was used, pcregrep was setting a return code of 1, even
294 when matches were found, and --line-buffered was not being honoured.
295
296 12. Added an optional parentheses number to the -o and --only-matching options
297 of pcregrep.
298
299 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
300 can match an empty string. The code to do it in pcretest and pcredemo
301 needed fixing:
302
303 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
304 only one byte after an empty string match just before CRLF (this case
305 just got forgotten; "any" and "anycrlf" were OK).
306
307 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
308 mode when an empty string match preceded an ASCII character followed by
309 a non-ASCII character. (The code for advancing by one character rather
310 than one byte was nonsense.)
311
312 (c) The pcredemo.c sample program did not have any code at all to handle
313 the cases when CRLF is a valid newline sequence.
314
315 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
316 as a starting offset was within the subject string. There is now a new
317 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
318 negative or greater than the length of the string. In order to test this,
319 pcretest is extended to allow the setting of negative starting offsets.
320
321 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
322 starting offset points to the beginning of a UTF-8 character was
323 unnecessarily clumsy. I tidied it up.
324
325 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
326 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
327
328 17. Nobody had reported that the --include_dir option, which was added in
329 release 7.7 should have been called --include-dir (hyphen, not underscore)
330 for compatibility with GNU grep. I have changed it to --include-dir, but
331 left --include_dir as an undocumented synonym, and the same for
332 --exclude-dir, though that is not available in GNU grep, at least as of
333 release 2.5.4.
334
335 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
336 characters from a string of bytes) have been redefined so as not to use
337 loops, in order to improve performance in some environments. At the same
338 time, I abstracted some of the common code into auxiliary macros to save
339 repetition (this should not affect the compiled code).
340
341 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
342 compile-time error is now given if \c is not followed by an ASCII
343 character, that is, a byte less than 128. (In EBCDIC mode, the code is
344 different, and any byte value is allowed.)
345
346 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
347 START_OPTIMIZE option, which is now allowed at compile time - but just
348 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
349 to pcregrep and other applications that have no direct access to PCRE
350 options. The new /Y option in pcretest sets this option when calling
351 pcre_compile().
352
353 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
354 back references. Groups containing recursive back references were forced to
355 be atomic by that change, but in the case of named groups, the amount of
356 memory required was incorrectly computed, leading to "Failed: internal
357 error: code overflow". This has been fixed.
358
359 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
360 pcretest.c, to avoid build problems in some Borland environments.
361
362
363 Version 8.10 25-Jun-2010
364 ------------------------
365
366 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
367 THEN.
368
369 2. (*ACCEPT) was not working when inside an atomic group.
370
371 3. Inside a character class, \B is treated as a literal by default, but
372 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
373 causes the error). The code is unchanged, but I tidied the documentation.
374
375 4. Inside a character class, PCRE always treated \R and \X as literals,
376 whereas Perl faults them if its -w option is set. I have changed PCRE so
377 that it faults them when PCRE_EXTRA is set.
378
379 5. Added support for \N, which always matches any character other than
380 newline. (It is the same as "." when PCRE_DOTALL is not set.)
381
382 6. When compiling pcregrep with newer versions of gcc which may have
383 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
384 declared with attribute warn_unused_result" were given. Just casting the
385 result to (void) does not stop the warnings; a more elaborate fudge is
386 needed. I've used a macro to implement this.
387
388 7. Minor change to pcretest.c to avoid a compiler warning.
389
390 8. Added four artifical Unicode properties to help with an option to make
391 \s etc use properties (see next item). The new properties are: Xan
392 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
393
394 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
395 use Unicode properties. (*UCP) at the start of a pattern can be used to set
396 this option. Modified pcretest to add /W to test this facility. Added
397 REG_UCP to make it available via the POSIX interface.
398
399 10. Added --line-buffered to pcregrep.
400
401 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
402 studied, and the match started with a letter with a code point greater than
403 127 whose first byte was different to the first byte of the other case of
404 the letter, the other case of this starting letter was not recognized
405 (#976).
406
407 12. If a pattern that was studied started with a repeated Unicode property
408 test, for example, \p{Nd}+, there was the theoretical possibility of
409 setting up an incorrect bitmap of starting bytes, but fortunately it could
410 not have actually happened in practice until change 8 above was made (it
411 added property types that matched character-matching opcodes).
412
413 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
414 possible starting bytes for non-anchored patterns.
415
416 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
417 \R, and also a number of cases that involve Unicode properties, both
418 explicit and implicit when PCRE_UCP is set.
419
420 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
421 input, it could crash or give wrong results if characters with values
422 greater than 0xc0 were present in the subject string. (Detail: it assumed
423 UTF-8 input when processing these items.)
424
425 16. Added a lot of (int) casts to avoid compiler warnings in systems where
426 size_t is 64-bit (#991).
427
428 17. Added a check for running out of memory when PCRE is compiled with
429 --disable-stack-for-recursion (#990).
430
431 18. If the last data line in a file for pcretest does not have a newline on
432 the end, a newline was missing in the output.
433
434 19. The default pcre_chartables.c file recognizes only ASCII characters (values
435 less than 128) in its various bitmaps. However, there is a facility for
436 generating tables according to the current locale when PCRE is compiled. It
437 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
438 characters, are recognized by isspace() and therefore were getting set in
439 these tables, and indeed these tables seem to approximate to ISO 8859. This
440 caused a problem in UTF-8 mode when pcre_study() was used to create a list
441 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
442 which of course cannot start UTF-8 characters. I have changed the code so
443 that only real ASCII characters (less than 128) and the correct starting
444 bytes for UTF-8 encodings are set for characters greater than 127 when in
445 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
446 altogether.)
447
448 20. Added the /T option to pcretest so as to be able to run tests with non-
449 standard character tables, thus making it possible to include the tests
450 used for 19 above in the standard set of tests.
451
452 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
453 reference to a subpattern the other side of a comment that contains an
454 opening parenthesis caused either an internal compiling error, or a
455 reference to the wrong subpattern.
456
457
458 Version 8.02 19-Mar-2010
459 ------------------------
460
461 1. The Unicode data tables have been updated to Unicode 5.2.0.
462
463 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
464 configured.
465
466 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
467 original author of that file, following a query about its status.
468
469 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
470 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
471
472 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
473 quantifier applied to a forward-referencing subroutine call, could compile
474 incorrect code or give the error "internal error: previously-checked
475 referenced subpattern not found".
476
477 6. Both MS Visual Studio and Symbian OS have problems with initializing
478 variables to point to external functions. For these systems, therefore,
479 pcre_malloc etc. are now initialized to local functions that call the
480 relevant global functions.
481
482 7. There were two entries missing in the vectors called coptable and poptable
483 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
484 I've fixed the data, and added a kludgy way of testing at compile time that
485 the lengths are correct (equal to the number of opcodes).
486
487 8. Following on from 7, I added a similar kludge to check the length of the
488 eint vector in pcreposix.c.
489
490 9. Error texts for pcre_compile() are held as one long string to avoid too
491 much relocation at load time. To find a text, the string is searched,
492 counting zeros. There was no check for running off the end of the string,
493 which could happen if a new error number was added without updating the
494 string.
495
496 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
497
498 11. \K was not working if it appeared in an atomic group or in a group that
499 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
500 \K is "not well defined" if used in an assertion. PCRE now accepts it if
501 the assertion is positive, but not if it is negative.
502
503 12. Change 11 fortuitously reduced the size of the stack frame used in the
504 "match()" function of pcre_exec.c by one pointer. Forthcoming
505 implementation of support for (*MARK) will need an extra pointer on the
506 stack; I have reserved it now, so that the stack frame size does not
507 decrease.
508
509 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
510 item in branch that calls a recursion is a subroutine call - as in the
511 second branch in the above example - was incorrectly given the compile-
512 time error "recursive call could loop indefinitely" because pcre_compile()
513 was not correctly checking the subroutine for matching a non-empty string.
514
515 14. The checks for overrunning compiling workspace could trigger after an
516 overrun had occurred. This is a "should never occur" error, but it can be
517 triggered by pathological patterns such as hundreds of nested parentheses.
518 The checks now trigger 100 bytes before the end of the workspace.
519
520 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
521
522
523 Version 8.01 19-Jan-2010
524 ------------------------
525
526 1. If a pattern contained a conditional subpattern with only one branch (in
527 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
528 computed the wrong minimum data length (which is of course zero for such
529 subpatterns). This could cause incorrect "no match" results.
530
531 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
532 the pattern is reset in the first branch, pcre_compile() failed with
533 "internal error: code overflow at offset...". This happened only when
534 the reset was to the original external option setting. (An optimization
535 abstracts leading options settings into an external setting, which was the
536 cause of this.)
537
538 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
539 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
540 assertion pattern did not match (meaning that the assertion was true), it
541 was incorrectly treated as false if the SKIP had been reached during the
542 matching. This also applied to assertions used as conditions.
543
544 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
545 assertion subpattern, including such a pattern used as a condition,
546 unpredictable results occurred, instead of the error return
547 PCRE_ERROR_DFA_UITEM.
548
549 5. The C++ GlobalReplace function was not working like Perl for the special
550 situation when an empty string is matched. It now does the fancy magic
551 stuff that is necessary.
552
553 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
554 removed. (These were left over from very, very early versions of PCRE.)
555
556 7. Some cosmetic changes to the code to make life easier when compiling it
557 as part of something else:
558
559 (a) Change DEBUG to PCRE_DEBUG.
560
561 (b) In pcre_compile(), rename the member of the "branch_chain" structure
562 called "current" as "current_branch", to prevent a collision with the
563 Linux macro when compiled as a kernel module.
564
565 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
566 prevent a collision with the Linux macro when compiled as a kernel
567 module.
568
569 8. In pcre_compile() there are some checks for integer overflows that used to
570 cast potentially large values to (double). This has been changed to that
571 when building, a check for int64_t is made, and if it is found, it is used
572 instead, thus avoiding the use of floating point arithmetic. (There is no
573 other use of FP in PCRE.) If int64_t is not found, the fallback is to
574 double.
575
576 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
577 2005 (difference between two addresses compared to an unsigned value).
578
579 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
580 custom one, because of the following reported problem in Windows:
581
582 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
583 under Win32.
584 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
585 therefore missing the function definition.
586 - The compiler thus generates a "C" signature for the test function.
587 - The linker fails to find the "C" function.
588 - PCRE fails to configure if asked to do so against libbz2.
589
590 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
591 messages were output:
592
593 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
594 rerunning libtoolize, to keep the correct libtool macros in-tree.
595 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
596
597 I have done both of these things.
598
599 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
600 most of the time, it *can* run out if it is given a pattern that contains a
601 runaway infinite recursion. I updated the discussion in the pcrestack man
602 page.
603
604 13. Now that we have gone to the x.xx style of version numbers, the minor
605 version may start with zero. Using 08 or 09 is a bad idea because users
606 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
607 interpreted as invalid octal numbers. I've updated the previous comment in
608 configure.ac, and also added a check that gives an error if 08 or 09 are
609 used.
610
611 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
612 causing partial matching to fail when the end of the subject matched \W
613 in a UTF-8 pattern where \W was quantified with a minimum of 3.
614
615 15. There were some discrepancies between the declarations in pcre_internal.h
616 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
617 their definitions. The declarations used "const uschar *" and the
618 definitions used USPTR. Even though USPTR is normally defined as "const
619 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
620 reported that: "This difference in casting confuses some C++ compilers, for
621 example, SunCC recognizes above declarations as different functions and
622 generates broken code for hbpcre." I have changed the declarations to use
623 USPTR.
624
625 16. GNU libtool is named differently on some systems. The autogen.sh script now
626 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
627 (FreeBSD).
628
629 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
630 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
631 comment: "Figure out how to create a longlong from a string: strtoll and
632 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
633 instance, but it only takes 2 args instead of 3!"
634
635 18. A subtle bug concerned with back references has been fixed by a change of
636 specification, with a corresponding code fix. A pattern such as
637 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
638 refers, was giving matches when it shouldn't. For example, xa=xaaa would
639 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
640 same bug. Such groups have to be quantified to be useful, or contained
641 inside another quantified group. (If there's no repetition, the reference
642 can never match.) The problem arises because, having left the group and
643 moved on to the rest of the pattern, a later failure that backtracks into
644 the group uses the captured value from the final iteration of the group
645 rather than the correct earlier one. I have fixed this in PCRE by forcing
646 any group that contains a reference to itself to be an atomic group; that
647 is, there cannot be any backtracking into it once it has completed. This is
648 similar to recursive and subroutine calls.
649
650
651 Version 8.00 19-Oct-09
652 ----------------------
653
654 1. The table for translating pcre_compile() error codes into POSIX error codes
655 was out-of-date, and there was no check on the pcre_compile() error code
656 being within the table. This could lead to an OK return being given in
657 error.
658
659 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
660 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
661 in a Windows environment.
662
663 3. The pcregrep --count option prints the count for each file even when it is
664 zero, as does GNU grep. However, pcregrep was also printing all files when
665 --files-with-matches was added. Now, when both options are given, it prints
666 counts only for those files that have at least one match. (GNU grep just
667 prints the file name in this circumstance, but including the count seems
668 more useful - otherwise, why use --count?) Also ensured that the
669 combination -clh just lists non-zero counts, with no names.
670
671 4. The long form of the pcregrep -F option was incorrectly implemented as
672 --fixed_strings instead of --fixed-strings. This is an incompatible change,
673 but it seems right to fix it, and I didn't think it was worth preserving
674 the old behaviour.
675
676 5. The command line items --regex=pattern and --regexp=pattern were not
677 recognized by pcregrep, which required --regex pattern or --regexp pattern
678 (with a space rather than an '='). The man page documented the '=' forms,
679 which are compatible with GNU grep; these now work.
680
681 6. No libpcreposix.pc file was created for pkg-config; there was just
682 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
683
684 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
685 when UCP support is not needed, by modifying the Python script that
686 generates it from Unicode data files. This should not matter if the module
687 is correctly used as a library, but I received one complaint about 50K of
688 unwanted data. My guess is that the person linked everything into his
689 program rather than using a library. Anyway, it does no harm.
690
691 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
692 was a minimum greater than 1 for a wide character in a possessive
693 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
694 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
695 character. Chaos in the form of incorrect output or a compiling loop could
696 result.
697
698 9. The restrictions on what a pattern can contain when partial matching is
699 requested for pcre_exec() have been removed. All patterns can now be
700 partially matched by this function. In addition, if there are at least two
701 slots in the offset vector, the offset of the earliest inspected character
702 for the match and the offset of the end of the subject are set in them when
703 PCRE_ERROR_PARTIAL is returned.
704
705 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
706 synonymous with PCRE_PARTIAL, for backwards compatibility, and
707 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
708 and may be more useful for multi-segment matching.
709
710 11. Partial matching with pcre_exec() is now more intuitive. A partial match
711 used to be given if ever the end of the subject was reached; now it is
712 given only if matching could not proceed because another character was
713 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
714 string "Z", which now yields "no match" instead of "partial match". In the
715 case of pcre_dfa_exec(), "no match" is given if every matching path for the
716 final character ended with (*FAIL).
717
718 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
719 if the pattern had a "must contain" character that was already found in the
720 earlier partial match, unless partial matching was again requested. For
721 example, with the pattern /dog.(body)?/, the "must contain" character is
722 "g". If the first part-match was for the string "dog", restarting with
723 "sbody" failed. This bug has been fixed.
724
725 13. The string returned by pcre_dfa_exec() after a partial match has been
726 changed so that it starts at the first inspected character rather than the
727 first character of the match. This makes a difference only if the pattern
728 starts with a lookbehind assertion or \b or \B (\K is not supported by
729 pcre_dfa_exec()). It's an incompatible change, but it makes the two
730 matching functions compatible, and I think it's the right thing to do.
731
732 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
733 so that the demonstration program is easily available in environments where
734 PCRE has not been installed from source.
735
736 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
737 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
738 library.
739
740 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
741 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
742 is not the first non-POSIX option to be added. Clearly some people find
743 these options useful.
744
745 17. If a caller to the POSIX matching function regexec() passes a non-zero
746 value for nmatch with a NULL value for pmatch, the value of
747 nmatch is forced to zero.
748
749 18. RunGrepTest did not have a test for the availability of the -u option of
750 the diff command, as RunTest does. It now checks in the same way as
751 RunTest, and also checks for the -b option.
752
753 19. If an odd number of negated classes containing just a single character
754 interposed, within parentheses, between a forward reference to a named
755 subpattern and the definition of the subpattern, compilation crashed with
756 an internal error, complaining that it could not find the referenced
757 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
758 [The bug was that it was starting one character too far in when skipping
759 over the character class, thus treating the ] as data rather than
760 terminating the class. This meant it could skip too much.]
761
762 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
763 /g option in pcretest when the pattern contains \K, which makes it possible
764 to have an empty string match not at the start, even when the pattern is
765 anchored. Updated pcretest and pcredemo to use this option.
766
767 21. If the maximum number of capturing subpatterns in a recursion was greater
768 than the maximum at the outer level, the higher number was returned, but
769 with unset values at the outer level. The correct (outer level) value is
770 now given.
771
772 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
773 PCRE did not set those parentheses (unlike Perl). I have now found a way to
774 make it do so. The string so far is captured, making this feature
775 compatible with Perl.
776
777 23. The tests have been re-organized, adding tests 11 and 12, to make it
778 possible to check the Perl 5.10 features against Perl 5.10.
779
780 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
781 pattern matches a fixed length string. PCRE did not allow this; now it
782 does. Neither allows recursion.
783
784 25. I finally figured out how to implement a request to provide the minimum
785 length of subject string that was needed in order to match a given pattern.
786 (It was back references and recursion that I had previously got hung up
787 on.) This code has now been added to pcre_study(); it finds a lower bound
788 to the length of subject needed. It is not necessarily the greatest lower
789 bound, but using it to avoid searching strings that are too short does give
790 some useful speed-ups. The value is available to calling programs via
791 pcre_fullinfo().
792
793 26. While implementing 25, I discovered to my embarrassment that pcretest had
794 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
795 study optimizations had never been tested with that matching function.
796 Oops. What is worse, even when it was passed study data, there was a bug in
797 pcre_dfa_exec() that meant it never actually used it. Double oops. There
798 were also very few tests of studied patterns with pcre_dfa_exec().
799
800 27. If (?| is used to create subpatterns with duplicate numbers, they are now
801 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
802 on the other side of the coin, they are no longer allowed to have different
803 names, because these cannot be distinguished in PCRE, and this has caused
804 confusion. (This is a difference from Perl.)
805
806 28. When duplicate subpattern names are present (necessarily with different
807 numbers, as required by 27 above), and a test is made by name in a
808 conditional pattern, either for a subpattern having been matched, or for
809 recursion in such a pattern, all the associated numbered subpatterns are
810 tested, and the overall condition is true if the condition is true for any
811 one of them. This is the way Perl works, and is also more like the way
812 testing by number works.
813
814
815 Version 7.9 11-Apr-09
816 ---------------------
817
818 1. When building with support for bzlib/zlib (pcregrep) and/or readline
819 (pcretest), all targets were linked against these libraries. This included
820 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
821 libraries. This caused unwanted dependencies to be created. This problem
822 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
823 pcretest is linked with readline.
824
825 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
826 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
827 moved outside it again, because FALSE and TRUE are already defined in AIX,
828 but BOOL is not.
829
830 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
831 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
832
833 4. The pcregrep documentation said spaces were inserted as well as colons (or
834 hyphens) following file names and line numbers when outputting matching
835 lines. This is not true; no spaces are inserted. I have also clarified the
836 wording for the --colour (or --color) option.
837
838 5. In pcregrep, when --colour was used with -o, the list of matching strings
839 was not coloured; this is different to GNU grep, so I have changed it to be
840 the same.
841
842 6. When --colo(u)r was used in pcregrep, only the first matching substring in
843 each matching line was coloured. Now it goes on to look for further matches
844 of any of the test patterns, which is the same behaviour as GNU grep.
845
846 7. A pattern that could match an empty string could cause pcregrep to loop; it
847 doesn't make sense to accept an empty string match in pcregrep, so I have
848 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
849 seems to be how GNU grep behaves.
850
851 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
852 start or after a newline", because the conditional assertion was not being
853 correctly handled. The rule now is that both the assertion and what follows
854 in the first alternative must satisfy the test.
855
856 9. If auto-callout was enabled in a pattern with a conditional group whose
857 condition was an assertion, PCRE could crash during matching, both with
858 pcre_exec() and pcre_dfa_exec().
859
860 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
861 used for matching.
862
863 11. Unicode property support in character classes was not working for
864 characters (bytes) greater than 127 when not in UTF-8 mode.
865
866 12. Added the -M command line option to pcretest.
867
868 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
869
870 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
871
872 16. Added comments and documentation about mis-use of no_arg in the C++
873 wrapper.
874
875 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
876 from Martin Jerabek that uses macro names for all relevant character and
877 string constants.
878
879 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
880 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
881 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
882 these, but not everybody uses configure.
883
884 19. A conditional group that had only one branch was not being correctly
885 recognized as an item that could match an empty string. This meant that an
886 enclosing group might also not be so recognized, causing infinite looping
887 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
888 with the subject "ab", where knowledge that the repeated group can match
889 nothing is needed in order to break the loop.
890
891 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
892 exec(), but without supplying a callout function, matching went wrong.
893
894 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
895 leak if the size of the offset vector was greater than 30. When the vector
896 is smaller, the saved offsets during recursion go onto a local stack
897 vector, but for larger vectors malloc() is used. It was failing to free
898 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
899 error, in fact).
900
901 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
902 heapframe that is used only when UTF-8 support is enabled. This caused no
903 problem, but was untidy.
904
905 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
906 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
907 included within another project.
908
909 24. Steven Van Ingelgem's patches to add more options to the CMake support,
910 slightly modified by me:
911
912 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
913 not building pcregrep.
914
915 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
916 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
917
918 25. Forward references, both numeric and by name, in patterns that made use of
919 duplicate group numbers, could behave incorrectly or give incorrect errors,
920 because when scanning forward to find the reference group, PCRE was not
921 taking into account the duplicate group numbers. A pattern such as
922 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
923
924 26. Changed a few more instances of "const unsigned char *" to USPTR, making
925 the feature of a custom pointer more persuasive (as requested by a user).
926
927 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
928 pcretest.c, inside #ifndefs, because it seems they are sometimes already
929 pre-defined.
930
931 28. Added support for (*UTF8) at the start of a pattern.
932
933 29. Arrange for flags added by the "release type" setting in CMake to be shown
934 in the configuration summary.
935
936
937 Version 7.8 05-Sep-08
938 ---------------------
939
940 1. Replaced UCP searching code with optimized version as implemented for Ad
941 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
942 stage table and inline lookup instead of a function, giving speed ups of 2
943 to 5 times on some simple patterns that I tested. Permission was given to
944 distribute the MultiStage2.py script that generates the tables (it's not in
945 the tarball, but is in the Subversion repository).
946
947 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
948 scripts.
949
950 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
951 a group with a zero qualifier. The result of the study could be incorrect,
952 or the function might crash, depending on the pattern.
953
954 4. Caseless matching was not working for non-ASCII characters in back
955 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
956 It now works when Unicode Property Support is available.
957
958 5. In pcretest, an escape such as \x{de} in the data was always generating
959 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
960 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
961 truncation.
962
963 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
964
965 7. Added two (int) casts to pcregrep when printing the difference of two
966 pointers, in case they are 64-bit values.
967
968 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
969 test 2 if it fails.
970
971 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
972 and a #define of that name to empty if it is not externally set. This is to
973 allow users of MSVC to set it if necessary.
974
975 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
976 the convenience functions in the pcre_get.c source file.
977
978 11. An option change at the start of a pattern that had top-level alternatives
979 could cause overwriting and/or a crash. This command provoked a crash in
980 some environments:
981
982 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
983
984 This potential security problem was recorded as CVE-2008-2371.
985
986 12. For a pattern where the match had to start at the beginning or immediately
987 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
988 pcre_dfa_exec() could read past the end of the passed subject if there was
989 no match. To help with detecting such bugs (e.g. with valgrind), I modified
990 pcretest so that it places the subject at the end of its malloc-ed buffer.
991
992 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
993 exec() might read past the end of the data buffer in UTF-8 mode.
994
995 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
996 the data contained the byte 0x85 as part of a UTF-8 character within its
997 first line. This applied both to normal and DFA matching.
998
999 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1000 /^[^d]*?$/8 failed to match "abc".
1001
1002 16. Added a missing copyright notice to pcrecpp_internal.h.
1003
1004 17. Make it more clear in the documentation that values returned from
1005 pcre_exec() in ovector are byte offsets, not character counts.
1006
1007 18. Tidied a few places to stop certain compilers from issuing warnings.
1008
1009 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1010 supplied by Stefan Weber. I made a further small update for 7.8 because
1011 there is a change of source arrangements: the pcre_searchfuncs.c module is
1012 replaced by pcre_ucd.c.
1013
1014
1015 Version 7.7 07-May-08
1016 ---------------------
1017
1018 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
1019 a string to a long long, pretend we don't even have a long long." This is
1020 done by checking for the strtoq, strtoll, and _strtoi64 functions.
1021
1022 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1023 pre-7.6 versions, which defined a global no_arg variable instead of putting
1024 it in the RE class. (See also #8 below.)
1025
1026 3. Remove a line of dead code, identified by coverity and reported by Nuno
1027 Lopes.
1028
1029 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
1030
1031 (1) The include/exclude patterns were being applied to the whole pathnames
1032 of files, instead of just to the final components.
1033
1034 (2) If there was more than one level of directory, the subdirectories were
1035 skipped unless they satisfied the include/exclude conditions. This is
1036 inconsistent with GNU grep (and could even be seen as contrary to the
1037 pcregrep specification - which I improved to make it absolutely clear).
1038 The action now is always to scan all levels of directory, and just
1039 apply the include/exclude patterns to regular files.
1040
1041 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1042 --exclude_dir in the tests to avoid scanning .svn directories.
1043
1044 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1045 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1046 doesn't support NULs in patterns.
1047
1048 7. Added some missing "const"s to declarations of static tables in
1049 pcre_compile.c and pcre_dfa_exec.c.
1050
1051 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1052 caused by fix #2 above. (Subsequently also a second patch to fix the
1053 first patch. And a third patch - this was a messy problem.)
1054
1055 9. Applied Craig's patch to remove the use of push_back().
1056
1057 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1058 matching function regexec().
1059
1060 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1061 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1062 references. PCRE supports relative numbers with this syntax (I don't think
1063 Oniguruma does).
1064
1065 12. Previously, a group with a zero repeat such as (...){0} was completely
1066 omitted from the compiled regex. However, this means that if the group
1067 was called as a subroutine from elsewhere in the pattern, things went wrong
1068 (an internal error was given). Such groups are now left in the compiled
1069 pattern, with a new opcode that causes them to be skipped at execution
1070 time.
1071
1072 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1073 to the way PCRE behaves:
1074
1075 (a) A lone ] character is dis-allowed (Perl treats it as data).
1076
1077 (b) A back reference to an unmatched subpattern matches an empty string
1078 (Perl fails the current match path).
1079
1080 (c) A data ] in a character class must be notated as \] because if the
1081 first data character in a class is ], it defines an empty class. (In
1082 Perl it is not possible to have an empty class.) The empty class []
1083 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1084 The negative empty class [^] matches any one character, independently
1085 of the DOTALL setting.
1086
1087 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1088 non-existent subpattern following a character class starting with ']' and
1089 containing () gave an internal compiling error instead of "reference to
1090 non-existent subpattern". Fortunately, when the pattern did exist, the
1091 compiled code was correct. (When scanning forwards to check for the
1092 existencd of the subpattern, it was treating the data ']' as terminating
1093 the class, so got the count wrong. When actually compiling, the reference
1094 was subsequently set up correctly.)
1095
1096 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1097 it was being rejected as not supported by pcre_dfa_exec(), even though
1098 other assertions are supported. I have made pcre_dfa_exec() support
1099 (*FAIL).
1100
1101 16. The implementation of 13c above involved the invention of a new opcode,
1102 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1103 cannot be changed at match time, I realized I could make a small
1104 improvement to matching performance by compiling OP_ALLANY instead of
1105 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1106 on the OP_ANY path.
1107
1108 17. Compiling pcretest on Windows with readline support failed without the
1109 following two fixes: (1) Make the unistd.h include conditional on
1110 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1111
1112 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1113 ncurses library to be included for pcretest when ReadLine support is
1114 requested, but also to allow for it to be overridden. This patch came from
1115 Daniel Bergström.
1116
1117 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1118 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1119 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1120 spotting this.
1121
1122
1123 Version 7.6 28-Jan-08
1124 ---------------------
1125
1126 1. A character class containing a very large number of characters with
1127 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1128 overflow.
1129
1130 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1131 HAVE_LONG_LONG is not defined.
1132
1133 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1134 bring them up to date and include new features. This patch includes:
1135
1136 - Fixed PH's badly added libz and libbz2 support.
1137 - Fixed a problem with static linking.
1138 - Added pcredemo. [But later removed - see 7 below.]
1139 - Fixed dftables problem and added an option.
1140 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1141 HAVE_LONG_LONG.
1142 - Added readline support for pcretest.
1143 - Added an listing of the option settings after cmake has run.
1144
1145 4. A user submitted a patch to Makefile that makes it easy to create
1146 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1147 Makefile.am that cause it to include this special target, without
1148 affecting anything else. Note that the same mingw target plus all
1149 the other distribution libraries and programs are now supported
1150 when configuring with CMake (see 6 below) instead of with
1151 Configure/Make.
1152
1153 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1154 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1155 exported in the Windows port". It has not yet been confirmed that the patch
1156 solves the problem, but it does no harm.
1157
1158 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1159 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1160 with CMake, and also correct the comment about stack recursion.
1161
1162 7. Remove the automatic building of pcredemo from the ./configure system and
1163 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1164 of a program that users should build themselves after PCRE is installed, so
1165 building it automatically is not really right. What is more, it gave
1166 trouble in some build environments.
1167
1168 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1169
1170
1171 Version 7.5 10-Jan-08
1172 ---------------------
1173
1174 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1175 values in parens when parsing an RE using the C++ wrapper."
1176
1177 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1178 Characters greater than 255 were excluded from the class instead of being
1179 included.
1180
1181 3. The same bug as (2) above applied to negated POSIX classes such as
1182 [:^space:].
1183
1184 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1185 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1186 I have changed it.
1187
1188 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1189 first named subpattern) and a construct such as (?&a) would reference the
1190 first named subpattern whose name started with "a" (in other words, the
1191 length check was missing). Both these problems are fixed. "Subpattern name
1192 expected" is now given for (?&) (a zero-length name), and this patch also
1193 makes it give the same error for \k'' (previously it complained that that
1194 was a reference to a non-existent subpattern).
1195
1196 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1197 this is right because (?- can be followed by option settings as well as by
1198 digits. I have, however, made the messages clearer.
1199
1200 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1201 than the number used in the conditional) now cause a compile-time error.
1202 This is actually not compatible with Perl, which accepts such patterns, but
1203 treats the conditional as always being FALSE (as PCRE used to), but it
1204 seems to me that giving a diagnostic is better.
1205
1206 8. Change "alphameric" to the more common word "alphanumeric" in comments
1207 and messages.
1208
1209 9. Fix two occurrences of "backslash" in comments that should have been
1210 "backspace".
1211
1212 10. Remove two redundant lines of code that can never be obeyed (their function
1213 was moved elsewhere).
1214
1215 11. The program that makes PCRE's Unicode character property table had a bug
1216 which caused it to generate incorrect table entries for sequences of
1217 characters that have the same character type, but are in different scripts.
1218 It amalgamated them into a single range, with the script of the first of
1219 them. In other words, some characters were in the wrong script. There were
1220 thirteen such cases, affecting characters in the following ranges:
1221
1222 U+002b0 - U+002c1
1223 U+0060c - U+0060d
1224 U+0061e - U+00612
1225 U+0064b - U+0065e
1226 U+0074d - U+0076d
1227 U+01800 - U+01805
1228 U+01d00 - U+01d77
1229 U+01d9b - U+01dbf
1230 U+0200b - U+0200f
1231 U+030fc - U+030fe
1232 U+03260 - U+0327f
1233 U+0fb46 - U+0fbb1
1234 U+10450 - U+1049d
1235
1236 12. The -o option (show only the matching part of a line) for pcregrep was not
1237 compatible with GNU grep in that, if there was more than one match in a
1238 line, it showed only the first of them. It now behaves in the same way as
1239 GNU grep.
1240
1241 13. If the -o and -v options were combined for pcregrep, it printed a blank
1242 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1243 does the same. The return code can be used to tell if there were any
1244 non-matching lines.
1245
1246 14. Added --file-offsets and --line-offsets to pcregrep.
1247
1248 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1249 infinitely looping recursion. The bug was that positive lookaheads were not
1250 being skipped when checking for a possible empty match (negative lookaheads
1251 and both kinds of lookbehind were skipped).
1252
1253 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1254 inclusion of <windows.h> to before rather than after the definition of
1255 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1256
1257 17. Specifying a possessive quantifier with a specific limit for a Unicode
1258 character property caused pcre_compile() to compile bad code, which led at
1259 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1260 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1261 caused the error; without that there was no problem.
1262
1263 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1264
1265 19. Added --enable-pcretest-libreadline.
1266
1267 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1268 RE::GlobalReplace(). As a result, the number of replacements returned was
1269 double what it should be. I removed one of the increments, but Craig sent a
1270 later patch that removed the other one (the right fix) and added unit tests
1271 that check the return values (which was not done before).
1272
1273 21. Several CMake things:
1274
1275 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1276 the names libpcre and libpcreposix, not just pcre and pcreposix.
1277
1278 (2) The above change means that pcretest and pcregrep are now correctly
1279 linked with the newly-built libraries, not previously installed ones.
1280
1281 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1282
1283 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1284 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1285 UTF-8 newline character). The key issue is that the pattern starts .*;
1286 this means that the match must be either at the beginning, or after a
1287 newline. The bug was in the code for advancing after a failed match and
1288 checking that the new position followed a newline. It was not taking
1289 account of UTF-8 characters correctly.
1290
1291 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1292 character classes. PCRE was not treating the sequence [:...:] as a
1293 character class unless the ... were all letters. Perl, however, seems to
1294 allow any characters between [: and :], though of course it rejects as
1295 unknown any "names" that contain non-letters, because all the known class
1296 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1297 for example, whereas PCRE did not - it did not recognize a POSIX character
1298 class. This seemed a bit dangerous, so the code has been changed to be
1299 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1300 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1301 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1302 Perl does, and where it didn't before.
1303
1304 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1305 Windows environments %n is disabled by default.
1306
1307
1308 Version 7.4 21-Sep-07
1309 ---------------------
1310
1311 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1312 means that a class such as [\s] counted as "explicit reference to CR or
1313 LF". That isn't really right - the whole point of the change was to try to
1314 help when there was an actual mention of one of the two characters. So now
1315 the change happens only if \r or \n (or a literal CR or LF) character is
1316 encountered.
1317
1318 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1319 of both had grown to the point where there were only 3 bits left.
1320 Fortunately, there was spare space in the data structure, and so I have
1321 moved the internal flags into a new 16-bit field to free up more option
1322 bits.
1323
1324 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1325 but did not set the internal JCHANGED flag - either of these is enough to
1326 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1327 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1328 start sets both bits.
1329
1330 4. Added options (at build time, compile time, exec time) to change \R from
1331 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1332
1333 5. doc/pcresyntax.html was missing from the distribution.
1334
1335 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1336 compatibility, even though it is no longer used.
1337
1338 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1339 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1340 windows.h file is present (where different names are used). [This was
1341 reversed later after testing - see 16 below.]
1342
1343 8. Changed all #include <config.h> to #include "config.h". There were also
1344 some further <pcre.h> cases that I changed to "pcre.h".
1345
1346 9. When pcregrep was used with the --colour option, it missed the line ending
1347 sequence off the lines that it output.
1348
1349 10. It was pointed out to me that arrays of string pointers cause lots of
1350 relocations when a shared library is dynamically loaded. A technique of
1351 using a single long string with a table of offsets can drastically reduce
1352 these. I have refactored PCRE in four places to do this. The result is
1353 dramatic:
1354
1355 Originally: 290
1356 After changing UCP table: 187
1357 After changing error message table: 43
1358 After changing table of "verbs" 36
1359 After changing table of Posix names 22
1360
1361 Thanks to the folks working on Gregex for glib for this insight.
1362
1363 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1364 unicode-properties was also set.
1365
1366 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1367
1368 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1369 checked only for CRLF.
1370
1371 14. Added casts to pcretest.c to avoid compiler warnings.
1372
1373 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1374
1375 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1376 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1377 entirely. This removes changes made in 7 above.
1378
1379 17. The CMake files have been updated, and there is now more information about
1380 building with CMake in the NON-UNIX-USE document.
1381
1382
1383 Version 7.3 28-Aug-07
1384 ---------------------
1385
1386 1. In the rejigging of the build system that eventually resulted in 7.1, the
1387 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1388 brackets there is not right, since it causes compilers to look for an
1389 installed pcre.h, not the version that is in the source that is being
1390 compiled (which of course may be different). I have changed it back to:
1391
1392 #include "pcre.h"
1393
1394 I have a vague recollection that the change was concerned with compiling in
1395 different directories, but in the new build system, that is taken care of
1396 by the VPATH setting the Makefile.
1397
1398 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1399 when the subject happened to end in the byte 0x85 (e.g. if the last
1400 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1401 characters but of course it shouldn't be taken as a newline when it is part
1402 of another character. The bug was that, for an unlimited repeat of . in
1403 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1404 characters when looking for a newline.
1405
1406 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1407
1408 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1409 in debug output.
1410
1411 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1412 long printing in the pcrecpp unittest when running under MinGW.
1413
1414 6. ESC_K was left out of the EBCDIC table.
1415
1416 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1417 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1418 limit also applies to "virtual nesting" when a pattern is recursive, and in
1419 this case 1000 isn't so big. I have been able to remove this limit at the
1420 expense of backing off one optimization in certain circumstances. Normally,
1421 when pcre_exec() would call its internal match() function recursively and
1422 immediately return the result unconditionally, it uses a "tail recursion"
1423 feature to save stack. However, when a subpattern that can match an empty
1424 string has an unlimited repetition quantifier, it no longer makes this
1425 optimization. That gives it a stack frame in which to save the data for
1426 checking that an empty string has been matched. Previously this was taken
1427 from the 1000-entry workspace that had been reserved. So now there is no
1428 explicit limit, but more stack is used.
1429
1430 8. Applied Daniel's patches to solve problems with the import/export magic
1431 syntax that is required for Windows, and which was going wrong for the
1432 pcreposix and pcrecpp parts of the library. These were overlooked when this
1433 problem was solved for the main library.
1434
1435 9. There were some crude static tests to avoid integer overflow when computing
1436 the size of patterns that contain repeated groups with explicit upper
1437 limits. As the maximum quantifier is 65535, the maximum group length was
1438 set at 30,000 so that the product of these two numbers did not overflow a
1439 32-bit integer. However, it turns out that people want to use groups that
1440 are longer than 30,000 bytes (though not repeat them that many times).
1441 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1442 made it possible to implement the integer overflow checks in a much more
1443 dynamic way, which I have now done. The artificial limitation on group
1444 length has been removed - we now have only the limit on the total length of
1445 the compiled pattern, which depends on the LINK_SIZE setting.
1446
1447 10. Fixed a bug in the documentation for get/copy named substring when
1448 duplicate names are permitted. If none of the named substrings are set, the
1449 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1450 empty string.
1451
1452 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1453 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1454 because the ] is interpreted as the first data character and the
1455 terminating ] is not found. PCRE has been made compatible with Perl in this
1456 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1457 cause memory overwriting.
1458
1459 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1460 string has been matched (to stop an infinite loop). It was not recognizing
1461 a conditional subpattern that could match an empty string if that
1462 subpattern was within another subpattern. For example, it looped when
1463 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1464 condition was not nested. This bug has been fixed.
1465
1466 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1467 past the start of the subject in the presence of bytes with the top bit
1468 set, for example "\x8aBCD".
1469
1470 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1471 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1472
1473 14. Optimized (?!) to (*FAIL).
1474
1475 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1476 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1477 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1478 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1479 does: it's just the validity check that is more restrictive.
1480
1481 16. Inserted checks for integer overflows during escape sequence (backslash)
1482 processing, and also fixed erroneous offset values for syntax errors during
1483 backslash processing.
1484
1485 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1486 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1487
1488 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1489 caused an overrun.
1490
1491 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1492 something other than just ASCII characters) inside a group that had an
1493 unlimited repeat caused a loop at compile time (while checking to see
1494 whether the group could match an empty string).
1495
1496 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1497 [\P{Any}] did so. (Error in the code for printing property names.)
1498
1499 21. An orphan \E inside a character class could cause a crash.
1500
1501 22. A repeated capturing bracket such as (A)? could cause a wild memory
1502 reference during compilation.
1503
1504 23. There are several functions in pcre_compile() that scan along a compiled
1505 expression for various reasons (e.g. to see if it's fixed length for look
1506 behind). There were bugs in these functions when a repeated \p or \P was
1507 present in the pattern. These operators have additional parameters compared
1508 with \d, etc, and these were not being taken into account when moving along
1509 the compiled data. Specifically:
1510
1511 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1512 length.
1513
1514 (b) An item such as \pL+ within a repeated group could cause crashes or
1515 loops.
1516
1517 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1518 "reference to non-existent subpattern" error.
1519
1520 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1521
1522 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1523 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1524
1525 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1526
1527 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1528 character were causing crashes (broken optimization).
1529
1530 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1531 \p or \P) caused a compile-time loop.
1532
1533 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1534 break. For example, the unstudied pattern [\r\n]A does not match the string
1535 "\r\nA" because change 7.0/46 below moves the current point on by two
1536 characters after failing to match at the start. However, the pattern \nA
1537 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1538 the same is true. There doesn't seem any very clean way out of this, but
1539 what I have chosen to do makes the common cases work: PCRE now takes note
1540 of whether there can be an explicit match for \r or \n anywhere in the
1541 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1542 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1543 pattern has explicit CR or LF references.
1544
1545 29. Added (*CR) etc for changing newline setting at start of pattern.
1546
1547
1548 Version 7.2 19-Jun-07
1549 ---------------------
1550
1551 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1552 which is apparently normally available under Windows.
1553
1554 2. Re-jig the pcregrep tests with different newline settings in an attempt
1555 to make them independent of the local environment's newline setting.
1556
1557 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1558
1559 4. Some of the "internals" tests were previously cut out when the link size
1560 was not 2, because the output contained actual offsets. The recent new
1561 "Z" feature of pcretest means that these can be cut out, making the tests
1562 usable with all link sizes.
1563
1564 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1565 stack recursion. This gives a massive performance boost under BSD, but just
1566 a small improvement under Linux. However, it saves one field in the frame
1567 in all cases.
1568
1569 6. Added more features from the forthcoming Perl 5.10:
1570
1571 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1572 recursion call. It refers to the nth most recently opened parentheses.
1573
1574 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1575 to be opened parentheses.
1576
1577 (c) Conditions that refer to capturing parentheses can be specified
1578 relatively, for example, (?(-2)... or (?(+3)...
1579
1580 (d) \K resets the start of the current match so that everything before
1581 is not part of it.
1582
1583 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1584
1585 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1586 reference syntax.
1587
1588 (g) (?| introduces a group in which the numbering of parentheses in each
1589 alternative starts with the same number.
1590
1591 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1592
1593 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1594 PCRE_INFO_JCHANGED.
1595
1596 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1597 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1598 for detecting groups that can match an empty string.
1599
1600 9. A pattern with a very large number of alternatives (more than several
1601 hundred) was running out of internal workspace during the pre-compile
1602 phase, where pcre_compile() figures out how much memory will be needed. A
1603 bit of new cunning has reduced the workspace needed for groups with
1604 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1605 workspace instead of running out of the 4096 that are available.
1606
1607 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1608
1609 11. Applied patch from Google to remove an optimization that didn't quite work.
1610 The report of the bug said:
1611
1612 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1613 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1614 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1615
1616 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1617 it matched the wrong number of bytes.
1618
1619
1620 Version 7.1 24-Apr-07
1621 ---------------------
1622
1623 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1624 that is more "standard", making use of automake and other Autotools. There
1625 is some re-arrangement of the files and adjustment of comments consequent
1626 on this.
1627
1628 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1629 for recursive directory scanning broke on some systems because the files
1630 are not scanned in any specific order and on different systems the order
1631 was different. A call to "sort" has been inserted into RunGrepTest for the
1632 approprate test as a short-term fix. In the longer term there may be an
1633 alternative.
1634
1635 3. I had an email from Eric Raymond about problems translating some of PCRE's
1636 man pages to HTML (despite the fact that I distribute HTML pages, some
1637 people do their own conversions for various reasons). The problems
1638 concerned the use of low-level troff macros .br and .in. I have therefore
1639 removed all such uses from the man pages (some were redundant, some could
1640 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1641 HTML has been updated to handle .nf/.fi and to complain if it encounters
1642 .br or .in.
1643
1644 4. Updated comments in configure.ac that get placed in config.h.in and also
1645 arranged for config.h to be included in the distribution, with the name
1646 config.h.generic, for the benefit of those who have to compile without
1647 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1648
1649 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1650 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1651 makevp.bat for the current PCRE, using the additional files
1652 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1653
1654 6. A Windows user reported a minor discrepancy with test 2, which turned out
1655 to be caused by a trailing space on an input line that had got lost in his
1656 copy. The trailing space was an accident, so I've just removed it.
1657
1658 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1659 that is needed.
1660
1661 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1662 as "const" (a) because they are and (b) because it helps the PHP
1663 maintainers who have recently made a script to detect big data structures
1664 in the php code that should be moved to the .rodata section. I remembered
1665 to update Builducptable as well, so it won't revert if ucptable.h is ever
1666 re-created.
1667
1668 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1669 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1670 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1671 support is not required. This saves 1.5-2K of code, which is important in
1672 some applications.
1673
1674 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1675 so as not to refer to the tables, even though these functions will never be
1676 called when UTF-8 support is disabled. Otherwise there are problems with a
1677 shared library.
1678
1679 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1680
1681 (a) It was defining its arguments as char * instead of void *.
1682
1683 (b) It was assuming that all moves were upwards in memory; this was true
1684 a long time ago when I wrote it, but is no longer the case.
1685
1686 The emulated memove() is provided for those environments that have neither
1687 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1688 is clearly not the case, as these two bugs were recently reported.
1689
1690 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1691 and Detrail to create the HTML documentation, the .txt form of the man
1692 pages, and it removes trailing spaces from listed files. It also creates
1693 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1694 case, it wraps all the #defines with #ifndefs. This script should be run
1695 before "make dist".
1696
1697 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1698 with Unicode property support.
1699
1700 (a) For a maximizing quantifier, if the two different cases of the
1701 character were of different lengths in their UTF-8 codings (there are
1702 some cases like this - I found 11), and the matching function had to
1703 back up over a mixture of the two cases, it incorrectly assumed they
1704 were both the same length.
1705
1706 (b) When PCRE was configured to use the heap rather than the stack for
1707 recursion during matching, it was not correctly preserving the data for
1708 the other case of a UTF-8 character when checking ahead for a match
1709 while processing a minimizing repeat. If the check also involved
1710 matching a wide character, but failed, corruption could cause an
1711 erroneous result when trying to check for a repeat of the original
1712 character.
1713
1714 13. Some tidying changes to the testing mechanism:
1715
1716 (a) The RunTest script now detects the internal link size and whether there
1717 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1718 values substituted by "configure". (The RunGrepTest script already did
1719 this for UTF-8.) The configure.ac script no longer substitutes the
1720 relevant variables.
1721
1722 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1723 with length and offset values. This means that the output is different
1724 for different internal link sizes. Test 2 is skipped for link sizes
1725 other than 2 because of this, bypassing the problem. Unfortunately,
1726 there was also a test in test 3 (the locale tests) that used /B and
1727 failed for link sizes other than 2. Rather than cut the whole test out,
1728 I have added a new /Z option to pcretest that replaces the length and
1729 offset values with spaces. This is now used to make test 3 independent
1730 of link size. (Test 2 will be tidied up later.)
1731
1732 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1733 segmentation fault instead of returning the appropriate error message.
1734
1735 15. In multiline mode when the newline sequence was set to "any", the pattern
1736 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1737 This doesn't seem right; it now treats the CRLF combination as the line
1738 ending, and so does not match in that case. It's only a pattern such as ^$
1739 that would hit this one: something like ^ABC$ would have failed after \r
1740 and then tried again after \r\n.
1741
1742 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1743 in an attempt to make files that differ only in their line terminators
1744 compare equal. This works on Linux.
1745
1746 17. Under certain error circumstances pcregrep might try to free random memory
1747 as it exited. This is now fixed, thanks to valgrind.
1748
1749 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1750 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1751 was because its rules for how to advance for /g after matching an empty
1752 string at the end of a line did not allow for this case. They now check for
1753 it specially.
1754
1755 20. pcretest is supposed to handle patterns and data of any length, by
1756 extending its buffers when necessary. It was getting this wrong when the
1757 buffer for a data line had to be extended.
1758
1759 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1760 CRLF as a newline sequence.
1761
1762 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1763 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1764 I have nevertheless tidied it up.
1765
1766 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1767
1768 24. Added a man page for pcre-config.
1769
1770
1771 Version 7.0 19-Dec-06
1772 ---------------------
1773
1774 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1775 moving to gcc 4.1.1.
1776
1777 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1778 sys/time.h, which is documented as needed for this function. It doesn't
1779 seem to matter on Linux, but it showed up on some releases of OS X.
1780
1781 3. It seems that there are systems where bytes whose values are greater than
1782 127 match isprint() in the "C" locale. The "C" locale should be the
1783 default when a C program starts up. In most systems, only ASCII printing
1784 characters match isprint(). This difference caused the output from pcretest
1785 to vary, making some of the tests fail. I have changed pcretest so that:
1786
1787 (a) When it is outputting text in the compiled version of a pattern, bytes
1788 other than 32-126 are always shown as hex escapes.
1789
1790 (b) When it is outputting text that is a matched part of a subject string,
1791 it does the same, unless a different locale has been set for the match
1792 (using the /L modifier). In this case, it uses isprint() to decide.
1793
1794 4. Fixed a major bug that caused incorrect computation of the amount of memory
1795 required for a compiled pattern when options that changed within the
1796 pattern affected the logic of the preliminary scan that determines the
1797 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1798 that the computed length was too small. The symptoms of this bug were
1799 either the PCRE error "internal error: code overflow" from pcre_compile(),
1800 or a glibc crash with a message such as "pcretest: free(): invalid next
1801 size (fast)". Examples of patterns that provoked this bug (shown in
1802 pcretest format) are:
1803
1804 /(?-x: )/x
1805 /(?x)(?-x: \s*#\s*)/
1806 /((?i)[\x{c0}])/8
1807 /(?i:[\x{c0}])/8
1808
1809 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1810 is now done differently.
1811
1812 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1813 wrapper classes; (b) implement a new function in the C++ scanner that is
1814 more efficient than the old way of doing things because it avoids levels of
1815 recursion in the regex matching; (c) add a paragraph to the documentation
1816 for the FullMatch() function.
1817
1818 6. The escape sequence \n was being treated as whatever was defined as
1819 "newline". Not only was this contrary to the documentation, which states
1820 that \n is character 10 (hex 0A), but it also went horribly wrong when
1821 "newline" was defined as CRLF. This has been fixed.
1822
1823 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1824 was being set to -1 for the "end of line" case (supposedly a value that no
1825 character can have). Though this value is never used (the check for end of
1826 line is "zero bytes in current character"), it caused compiler complaints.
1827 I've changed it to 0xffffffff.
1828
1829 8. In pcre_version.c, the version string was being built by a sequence of
1830 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1831 string (as it is for production releases) called a macro with an empty
1832 argument. The C standard says the result of this is undefined. The gcc
1833 compiler treats it as an empty string (which was what was wanted) but it is
1834 reported that Visual C gives an error. The source has been hacked around to
1835 avoid this problem.
1836
1837 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1838 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1839 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1840 of them did).
1841
1842 10. Originally, pcretest opened its input and output without "b"; then I was
1843 told that "b" was needed in some environments, so it was added for release
1844 5.0 to both the input and output. (It makes no difference on Unix-like
1845 systems.) Later I was told that it is wrong for the input on Windows. I've
1846 now abstracted the modes into two macros, to make it easier to fiddle with
1847 them, and removed "b" from the input mode under Windows.
1848
1849 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1850
1851 12. Added -help and --help to pcretest as an official way of being reminded
1852 of the options.
1853
1854 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1855 and pcrecpp.cc because they annoy compilers at high warning levels.
1856
1857 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1858
1859 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1860 scripts are not C programs :-) and which was not noticed because it works
1861 on Linux.
1862
1863 16. pcretest is supposed to handle any length of pattern and data line (as one
1864 line or as a continued sequence of lines) by extending its input buffer if
1865 necessary. This feature was broken for very long pattern lines, leading to
1866 a string of junk being passed to pcre_compile() if the pattern was longer
1867 than about 50K.
1868
1869 17. I have done a major re-factoring of the way pcre_compile() computes the
1870 amount of memory needed for a compiled pattern. Previously, there was code
1871 that made a preliminary scan of the pattern in order to do this. That was
1872 OK when PCRE was new, but as the facilities have expanded, it has become
1873 harder and harder to keep it in step with the real compile phase, and there
1874 have been a number of bugs (see for example, 4 above). I have now found a
1875 cunning way of running the real compile function in a "fake" mode that
1876 enables it to compute how much memory it would need, while actually only
1877 ever using a few hundred bytes of working memory and without too many
1878 tests of the mode. This should make future maintenance and development
1879 easier. A side effect of this work is that the limit of 200 on the nesting
1880 depth of parentheses has been removed (though this was never a serious
1881 limitation, I suspect). However, there is a downside: pcre_compile() now
1882 runs more slowly than before (30% or more, depending on the pattern). I
1883 hope this isn't a big issue. There is no effect on runtime performance.
1884
1885 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1886 newline (only possible for the last line of a file) and it was a
1887 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1888
1889 19. Added additional timing features to pcretest. (1) The -tm option now times
1890 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1891 separate command line item, by a number that specifies the number of
1892 repeats to use when timing. The default is 50000; this gives better
1893 precision, but takes uncomfortably long for very large patterns.
1894
1895 20. Extended pcre_study() to be more clever in cases where a branch of a
1896 subpattern has no definite first character. For example, (a*|b*)[cd] would
1897 previously give no result from pcre_study(). Now it recognizes that the
1898 first character must be a, b, c, or d.
1899
1900 21. There was an incorrect error "recursive call could loop indefinitely" if
1901 a subpattern (or the entire pattern) that was being tested for matching an
1902 empty string contained only one non-empty item after a nested subpattern.
1903 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1904 incorrectly, because the \d was being skipped in the check.
1905
1906 22. The pcretest program now has a new pattern option /B and a command line
1907 option -b, which is equivalent to adding /B to every pattern. This causes
1908 it to show the compiled bytecode, without the additional information that
1909 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1910 is the same as /B/I).
1911
1912 23. A new optimization is now able automatically to treat some sequences such
1913 as a*b as a*+b. More specifically, if something simple (such as a character
1914 or a simple class like \d) has an unlimited quantifier, and is followed by
1915 something that cannot possibly match the quantified thing, the quantifier
1916 is automatically "possessified".
1917
1918 24. A recursive reference to a subpattern whose number was greater than 39
1919 went wrong under certain circumstances in UTF-8 mode. This bug could also
1920 have affected the operation of pcre_study().
1921
1922 25. Realized that a little bit of performance could be had by replacing
1923 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1924
1925 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1926
1927 27. Possessive quantifiers such as a++ were previously implemented by turning
1928 them into atomic groups such as ($>a+). Now they have their own opcodes,
1929 which improves performance. This includes the automatically created ones
1930 from 23 above.
1931
1932 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1933 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1934 the first matched character to be a colon. This applied both to named and
1935 numbered groups.
1936
1937 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1938
1939 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1940 building PCRE on Windows easier, so I have included it in the distribution.
1941
1942 31. There is now a check in pcretest against a ridiculously large number being
1943 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1944 loop, the loop is abandoned.
1945
1946 32. Forward references to subpatterns in conditions such as (?(2)...) where
1947 subpattern 2 is defined later cause pcre_compile() to search forwards in
1948 the pattern for the relevant set of parentheses. This search went wrong
1949 when there were unescaped parentheses in a character class, parentheses
1950 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1951
1952 33. "Subroutine" calls and backreferences were previously restricted to
1953 referencing subpatterns earlier in the regex. This restriction has now
1954 been removed.
1955
1956 34. Added a number of extra features that are going to be in Perl 5.10. On the
1957 whole, these are just syntactic alternatives for features that PCRE had
1958 previously implemented using the Python syntax or my own invention. The
1959 other formats are all retained for compatibility.
1960
1961 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1962 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1963 also .NET compatible.
1964
1965 (b) A recursion or subroutine call to a named group can now be defined as
1966 (?&name) as well as (?P>name).
1967
1968 (c) A backreference to a named group can now be defined as \k<name> or
1969 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1970 5.10, are also .NET compatible.
1971
1972 (d) A conditional reference to a named group can now use the syntax
1973 (?(<name>) or (?('name') as well as (?(name).
1974
1975 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1976 groups (named and numbered) that are never evaluated inline, but can be
1977 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1978 is always false. There may be only one alternative in such a group.
1979
1980 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1981 as the simple (?(R). The condition is true only if the most recent
1982 recursion is that of the given number or name. It does not search out
1983 through the entire recursion stack.
1984
1985 (g) The escape \gN or \g{N} has been added, where N is a positive or
1986 negative number, specifying an absolute or relative reference.
1987
1988 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1989 some "unreachable code" warnings.
1990
1991 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1992 things, this adds five new scripts.
1993
1994 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1995 There were also incompatibilities regarding the handling of \Q..\E inside
1996 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1997 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1998
1999 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2000 matches an empty string, and forcibly breaks the loop. There were bugs in
2001 this code in non-simple cases. For a pattern such as ^(a()*)* matched
2002 against aaaa the result was just "a" rather than "aaaa", for example. Two
2003 separate and independent bugs (that affected different cases) have been
2004 fixed.
2005
2006 39. Refactored the code to abolish the use of different opcodes for small
2007 capturing bracket numbers. This is a tidy that I avoided doing when I
2008 removed the limit on the number of capturing brackets for 3.5 back in 2001.
2009 The new approach is not only tidier, it makes it possible to reduce the
2010 memory needed to fix the previous bug (38).
2011
2012 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2013 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2014 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2015 mode.
2016
2017 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2018 report.
2019
2020 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2021 copy construction and assignment in the C++ wrapper.
2022
2023 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2024 couple of bugs that could have given wrong results in the "--newline=crlf"
2025 case.
2026
2027 44. Added a number of casts and did some reorganization of signed/unsigned int
2028 variables following suggestions from Dair Grant. Also renamed the variable
2029 "this" as "item" because it is a C++ keyword.
2030
2031 45. Arranged for dftables to add
2032
2033 #include "pcre_internal.h"
2034
2035 to pcre_chartables.c because without it, gcc 4.x may remove the array
2036 definition from the final binary if PCRE is built into a static library and
2037 dead code stripping is activated.
2038
2039 46. For an unanchored pattern, if a match attempt fails at the start of a
2040 newline sequence, and the newline setting is CRLF or ANY, and the next two
2041 characters are CRLF, advance by two characters instead of one.
2042
2043
2044 Version 6.7 04-Jul-06
2045 ---------------------
2046
2047 1. In order to handle tests when input lines are enormously long, pcretest has
2048 been re-factored so that it automatically extends its buffers when
2049 necessary. The code is crude, but this _is_ just a test program. The
2050 default size has been increased from 32K to 50K.
2051
2052 2. The code in pcre_study() was using the value of the re argument before
2053 testing it for NULL. (Of course, in any sensible call of the function, it
2054 won't be NULL.)
2055
2056 3. The memmove() emulation function in pcre_internal.h, which is used on
2057 systems that lack both memmove() and bcopy() - that is, hardly ever -
2058 was missing a "static" storage class specifier.
2059
2060 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2061 containing an extended class (one that cannot be represented by a bitmap
2062 because it contains high-valued characters or Unicode property items, e.g.
2063 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2064 pattern, but PCRE should not loop if you do not (it no longer does).
2065 [Detail: two cases were found: (a) a repeated subpattern containing an
2066 extended class; (b) a recursive reference to a subpattern that followed a
2067 previous extended class. It wasn't skipping over the extended class
2068 correctly when UTF-8 mode was not set.]
2069
2070 5. A negated single-character class was not being recognized as fixed-length
2071 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2072 compile error "lookbehind assertion is not fixed length".
2073
2074 6. The RunPerlTest auxiliary script was showing an unexpected difference
2075 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2076 write a Perl script that can interpret lines of an input file either as
2077 byte characters or as UTF-8, which is what "perltest" was being required to
2078 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2079 can't do is switch easily at run time between having the "use utf8;" pragma
2080 or not. In the end, I fudged it by using the RunPerlTest script to insert
2081 "use utf8;" explicitly for the UTF-8 tests.
2082
2083 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2084 the end of the subject string, contrary to the documentation and to what
2085 Perl does. This was true of both matching functions. Now it matches only at
2086 the start of the subject and immediately after *internal* newlines.
2087
2088 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2089 a pointer to an int instead of a pointer to an unsigned long int. This
2090 caused problems on 64-bit systems.
2091
2092 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2093 instance of the 'standard' template library not being so standard".
2094
2095 10. There was no check on the number of named subpatterns nor the maximum
2096 length of a subpattern name. The product of these values is used to compute
2097 the size of the memory block for a compiled pattern. By supplying a very
2098 long subpattern name and a large number of named subpatterns, the size
2099 computation could be caused to overflow. This is now prevented by limiting
2100 the length of names to 32 characters, and the number of named subpatterns
2101 to 10,000.
2102
2103 11. Subpatterns that are repeated with specific counts have to be replicated in
2104 the compiled pattern. The size of memory for this was computed from the
2105 length of the subpattern and the repeat count. The latter is limited to
2106 65535, but there was no limit on the former, meaning that integer overflow
2107 could in principle occur. The compiled length of a repeated subpattern is
2108 now limited to 30,000 bytes in order to prevent this.
2109
2110 12. Added the optional facility to have named substrings with the same name.
2111
2112 13. Added the ability to use a named substring as a condition, using the
2113 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2114 are numbers (not recommended). Forward references are permitted.
2115
2116 14. Added forward references in named backreferences (if you see what I mean).
2117
2118 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2119 pattern could run off the end of the subject. For example, the pattern
2120 "(?s)(.{1,5})"8 did this with the subject "ab".
2121
2122 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2123 PCRE_CASELESS was set when matching characters that were quantified with ?
2124 or *.
2125
2126 17. A character class other than a single negated character that had a minimum
2127 but no maximum quantifier - for example [ab]{6,} - was not handled
2128 correctly by pce_dfa_exec(). It would match only one character.
2129
2130 18. A valid (though odd) pattern that looked like a POSIX character
2131 class but used an invalid character after [ (for example [[,abc,]]) caused
2132 pcre_compile() to give the error "Failed: internal error: code overflow" or
2133 in some cases to crash with a glibc free() error. This could even happen if
2134 the pattern terminated after [[ but there just happened to be a sequence of
2135 letters, a binary zero, and a closing ] in the memory that followed.
2136
2137 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2138 over the years. Originally (before any Unicode support), just the bottom 8
2139 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2140 output from "man perlunicode" includes this:
2141
2142 The regular expression compiler produces polymorphic opcodes. That
2143 is, the pattern adapts to the data and automatically switches to
2144 the Unicode character scheme when presented with Unicode data--or
2145 instead uses a traditional byte scheme when presented with byte
2146 data.
2147
2148 Sadly, a wide octal escape does not cause a switch, and in a string with
2149 no other multibyte characters, these octal escapes are treated as before.
2150 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2151 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2152 Unicode string.
2153
2154 I have not perpetrated such confusion in PCRE. Up till now, it took just
2155 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2156 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2157 translate to the appropriate multibyte character.
2158
2159 29. Applied some refactoring to reduce the number of warnings from Microsoft
2160 and Borland compilers. This has included removing the fudge introduced
2161 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2162 a warning about an unused variable.
2163
2164 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2165 characters since release 4.0, because Perl (from release 5.004) does not.
2166 [Or at least, is documented not to: some releases seem to be in conflict
2167 with the documentation.] However, when a pattern was studied with
2168 pcre_study() and all its branches started with \s, PCRE still included VT
2169 as a possible starting character. Of course, this did no harm; it just
2170 caused an unnecessary match attempt.
2171
2172 22. Removed a now-redundant internal flag bit that recorded the fact that case
2173 dependency changed within the pattern. This was once needed for "required
2174 byte" processing, but is no longer used. This recovers a now-scarce options
2175 bit. Also moved the least significant internal flag bit to the most-
2176 significant bit of the word, which was not previously used (hangover from
2177 the days when it was an int rather than a uint) to free up another bit for
2178 the future.
2179
2180 23. Added support for CRLF line endings as well as CR and LF. As well as the
2181 default being selectable at build time, it can now be changed at runtime
2182 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2183 specify that it is scanning data with non-default line endings.
2184
2185 24. Changed the definition of CXXLINK to make it agree with the definition of
2186 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2187
2188 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2189 recursions. This makes a big different to stack usage for some patterns.
2190
2191 26. If a subpattern containing a named recursion or subroutine reference such
2192 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2193 the space required for the compiled pattern went wrong and gave too small a
2194 value. Depending on the environment, this could lead to "Failed: internal
2195 error: code overflow at offset 49" or "glibc detected double free or
2196 corruption" errors.
2197
2198 27. Applied patches from Google (a) to support the new newline modes and (b) to
2199 advance over multibyte UTF-8 characters in GlobalReplace.
2200
2201 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2202 difference for some implementation of PCRE in some Windows version.
2203
2204 29. Added some extra testing facilities to pcretest:
2205
2206 \q<number> in a data line sets the "match limit" value
2207 \Q<number> in a data line sets the "match recursion limt" value
2208 -S <number> sets the stack size, where <number> is in megabytes
2209
2210 The -S option isn't available for Windows.
2211
2212
2213 Version 6.6 06-Feb-06
2214 ---------------------
2215
2216 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2217 in pcreposix.h. I have copied the definition from pcre.h.
2218
2219 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2220 because pcre.h is no longer a built file.
2221
2222 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2223 not normally included in the compiled code.
2224
2225
2226 Version 6.5 01-Feb-06
2227 ---------------------
2228
2229 1. When using the partial match feature with pcre_dfa_exec(), it was not
2230 anchoring the second and subsequent partial matches at the new starting
2231 point. This could lead to incorrect results. For example, with the pattern
2232 /1234/, partially matching against "123" and then "a4" gave a match.
2233
2234 2. Changes to pcregrep:
2235
2236 (a) All non-match returns from pcre_exec() were being treated as failures
2237 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2238 error message is output. Some extra information is given for the
2239 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2240 probably the only errors that are likely to be caused by users (by
2241 specifying a regex that has nested indefinite repeats, for instance).
2242 If there are more than 20 of these errors, pcregrep is abandoned.
2243
2244 (b) A binary zero was treated as data while matching, but terminated the
2245 output line if it was written out. This has been fixed: binary zeroes
2246 are now no different to any other data bytes.
2247
2248 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2249 used to set a locale for matching. The --locale=xxxx long option has
2250 been added (no short equivalent) to specify a locale explicitly on the
2251 pcregrep command, overriding the environment variables.
2252
2253 (d) When -B was used with -n, some line numbers in the output were one less
2254 than they should have been.
2255
2256 (e) Added the -o (--only-matching) option.
2257
2258 (f) If -A or -C was used with -c (count only), some lines of context were
2259 accidentally printed for the final match.
2260
2261 (g) Added the -H (--with-filename) option.
2262
2263 (h) The combination of options -rh failed to suppress file names for files
2264 that were found from directory arguments.
2265
2266 (i) Added the -D (--devices) and -d (--directories) options.
2267
2268 (j) Added the -F (--fixed-strings) option.
2269
2270 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2271
2272 (l) Added the --colo(u)r option.
2273
2274 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2275 is not present by default.
2276
2277 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2278 items such as (?R) or (?1), when the recursion could match a number of
2279 alternatives. If it matched one of the alternatives, but subsequently,
2280 outside the recursion, there was a failure, the code tried to back up into
2281 the recursion. However, because of the way PCRE is implemented, this is not
2282 possible, and the result was an incorrect result from the match.
2283
2284 In order to prevent this happening, the specification of recursion has
2285 been changed so that all such subpatterns are automatically treated as
2286 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2287
2288 4. I had overlooked the fact that, in some locales, there are characters for
2289 which isalpha() is true but neither isupper() nor islower() are true. In
2290 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2291 and ordfeminine) are like this. This affected the treatment of \w and \W
2292 when they appeared in character classes, but not when they appeared outside
2293 a character class. The bit map for "word" characters is now created
2294 separately from the results of isalnum() instead of just taking it from the
2295 upper, lower, and digit maps. (Plus the underscore character, of course.)
2296
2297 5. The above bug also affected the handling of POSIX character classes such as
2298 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2299 permanent tables. Instead, the bit maps for such a class were previously
2300 created as the appropriate unions of the upper, lower, and digit bitmaps.
2301 Now they are created by subtraction from the [[:word:]] class, which has
2302 its own bitmap.
2303
2304 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2305 It is created by subtracting the vertical space characters (\x09, \x0a,
2306 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2307 subtraction was done in the overall bitmap for a character class, meaning
2308 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2309 be recognized. This bug has been fixed.
2310
2311 7. Patches from the folks at Google:
2312
2313 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2314 real life, but is still worth protecting against".
2315
2316 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2317 regular expressions".
2318
2319 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2320 have it.
2321
2322 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2323 "configure" and the latter not, in order to fix a problem somebody had
2324 with compiling the Arg class on HP-UX.
2325
2326 (e) Improve the error-handling of the C++ wrapper a little bit.
2327
2328 (f) New tests for checking recursion limiting.
2329
2330 8. The pcre_memmove() function, which is used only if the environment does not
2331 have a standard memmove() function (and is therefore rarely compiled),
2332 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2333 returning a result (though PCRE never actually uses the result).
2334
2335 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2336 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2337 returned instead of calling malloc() with an overflowing number that would
2338 most likely cause subsequent chaos.
2339
2340 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2341
2342 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2343 with this option is matched, the nmatch and pmatch options of regexec() are
2344 ignored.
2345
2346 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2347 provided in case anyone wants to the the POSIX interface with UTF-8
2348 strings.
2349
2350 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2351 C++ linking (needed for some HP-UX environments).
2352
2353 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2354 (unused parameter) and in the pcre_printint() function (omitted "default"
2355 switch label when the default is to do nothing).
2356
2357 15. Added some code to make it possible, when PCRE is compiled as a C++
2358 library, to replace subject pointers for pcre_exec() with a smart pointer
2359 class, thus making it possible to process discontinuous strings.
2360
2361 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2362 much the same function. They were added by different people who were trying
2363 to make PCRE easy to compile on non-Unix systems. It has been suggested
2364 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2365 for compiling on Windows systems. I have therefore replaced it with
2366 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2367 defaults to "extern" for C or "extern C" for C++, which works fine on
2368 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2369 SCOPE with something explicit in config.h. In addition:
2370
2371 (a) pcreposix.h still had just "extern" instead of either of these macros;
2372 I have replaced it with PCRE_DATA_SCOPE.
2373
2374 (b) Functions such as _pcre_xclass(), which are internal to the library,
2375 but external in the C sense, all had PCRE_EXPORT in their definitions.
2376 This is apparently wrong for the Windows case, so I have removed it.
2377 (It makes no difference on Unix-like systems.)
2378
2379 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2380 of recursive calls to match(). This is different to MATCH_LIMIT because
2381 that limits the total number of calls to match(), not all of which increase
2382 the depth of recursion. Limiting the recursion depth limits the amount of
2383 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2384 when PCRE is compiled, and changed at run time. A patch from Google adds
2385 this functionality to the C++ interface.
2386
2387 18. Changes to the handling of Unicode character properties:
2388
2389 (a) Updated the table to Unicode 4.1.0.
2390
2391 (b) Recognize characters that are not in the table as "Cn" (undefined).
2392
2393 (c) I revised the way the table is implemented to a much improved format
2394 which includes recognition of ranges. It now supports the ranges that
2395 are defined in UnicodeData.txt, and it also amalgamates other
2396 characters into ranges. This has reduced the number of entries in the
2397 table from around 16,000 to around 3,000, thus reducing its size
2398 considerably. I realized I did not need to use a tree structure after
2399 all - a binary chop search is just as efficient. Having reduced the
2400 number of entries, I extended their size from 6 bytes to 8 bytes to
2401 allow for more data.
2402
2403 (d) Added support for Unicode script names via properties such as \p{Han}.
2404
2405 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2406 matching that character.
2407
2408 20. When matching a repeated Unicode property with a minimum greater than zero,
2409 (for example \pL{2,}), PCRE could look past the end of the subject if it
2410 reached it while seeking the minimum number of characters. This could
2411 happen only if some of the characters were more than one byte long, because
2412 there is a check for at least the minimum number of bytes.
2413
2414 21. Refactored the implementation of \p and \P so as to be more general, to
2415 allow for more different types of property in future. This has changed the
2416 compiled form incompatibly. Anybody with saved compiled patterns that use
2417 \p or \P will have to recompile them.
2418
2419 22. Added "Any" and "L&" to the supported property types.
2420
2421 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2422 but give a compile time error if the value is greater than 0xff.
2423
2424 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2425 accidentally not being installed or uninstalled.
2426
2427 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2428 made were to insert the current release number. This seemed silly, because
2429 it made things harder for people building PCRE on systems that don't run
2430 "configure". I have turned pcre.h into a distributed file, no longer built
2431 by "configure", with the version identification directly included. There is
2432 no longer a pcre.h.in file.
2433
2434 However, this change necessitated a change to the pcre-config script as
2435 well. It is built from pcre-config.in, and one of the substitutions was the
2436 release number. I have updated configure.ac so that ./configure now finds
2437 the release number by grepping pcre.h.
2438
2439 26. Added the ability to run the tests under valgrind.
2440
2441
2442 Version 6.4 05-Sep-05
2443 ---------------------
2444
2445 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2446 "--" to be printed when multiple files were scanned, even when none of the
2447 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2448 consider it to be a bug, and have restored the previous behaviour.
2449
2450 2. A couple of code tidies to get rid of compiler warnings.
2451
2452 3. The pcretest program used to cheat by referring to symbols in the library
2453 whose names begin with _pcre_. These are internal symbols that are not
2454 really supposed to be visible externally, and in some environments it is
2455 possible to suppress them. The cheating is now confined to including
2456 certain files from the library's source, which is a bit cleaner.
2457
2458 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2459 file's purpose clearer.
2460
2461 5. Reorganized pcre_ucp_findchar().
2462
2463
2464 Version 6.3 15-Aug-05
2465 ---------------------
2466
2467 1. The file libpcre.pc.in did not have general read permission in the tarball.
2468
2469 2. There were some problems when building without C++ support:
2470
2471 (a) If C++ support was not built, "make install" and "make test" still
2472 tried to test it.
2473
2474 (b) There were problems when the value of CXX was explicitly set. Some
2475 changes have been made to try to fix these, and ...
2476
2477 (c) --disable-cpp can now be used to explicitly disable C++ support.
2478
2479 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2480 backslash in a target when C++ was disabled. This confuses some
2481 versions of "make", apparently. Using an intermediate variable solves
2482 this. (Same for CPP_LOBJ.)
2483
2484 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2485 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2486 necessary on certain architectures.
2487
2488 4. Added a setting of -export-symbols-regex to the link command to remove
2489 those symbols that are exported in the C sense, but actually are local
2490 within the library, and not documented. Their names all begin with
2491 "_pcre_". This is not a perfect job, because (a) we have to except some
2492 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2493 available (and never for static libraries). I have made a note to try to
2494 find a way round (a) in the future.
2495
2496
2497 Version 6.2 01-Aug-05
2498 ---------------------
2499
2500 1. There was no test for integer overflow of quantifier values. A construction
2501 such as {1111111111111111} would give undefined results. What is worse, if
2502 a minimum quantifier for a parenthesized subpattern overflowed and became
2503 negative, the calculation of the memory size went wrong. This could have
2504 led to memory overwriting.
2505
2506 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2507
2508 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2509 operating environments where this matters.
2510
2511 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2512 PCRE options from within the C++ wrapper.
2513
2514 5. Named capturing subpatterns were not being correctly counted when a pattern
2515 was compiled. This caused two problems: (a) If there were more than 100
2516 such subpatterns, the calculation of the memory needed for the whole
2517 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2518 back references of the form \12, where the number was greater than 9, were
2519 not recognized as back references, even though there were sufficient
2520 previous subpatterns.
2521
2522 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2523 versions of gcc, e.g. 2.95.4.
2524
2525
2526 Version 6.1 21-Jun-05
2527 ---------------------
2528
2529 1. There was one reference to the variable "posix" in pcretest.c that was not
2530 surrounded by "#if !defined NOPOSIX".
2531
2532 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2533 the cross-check on the old pcre_info() function, for the benefit of the
2534 cut-down version of PCRE that is currently imported into Exim.
2535
2536 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2537 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2538 patterns that start (?i) but I don't think that matters. The use of (?i) is
2539 just an example; this all applies to the other options as well.
2540
2541 4. Since libtool seems to echo the compile commands it is issuing, the output
2542 from "make" can be reduced a bit by putting "@" in front of each libtool
2543 compile command.
2544
2545 5. Patch from the folks at Google for configure.in to be a bit more thorough
2546 in checking for a suitable C++ installation before trying to compile the
2547 C++ stuff. This should fix a reported problem when a compiler was present,
2548 but no suitable headers.
2549
2550 6. The man pages all had just "PCRE" as their title. I have changed them to
2551 be the relevant file name. I have also arranged that these names are
2552 retained in the file doc/pcre.txt, which is a concatenation in text format
2553 of all the man pages except the little individual ones for each function.
2554
2555 7. The NON-UNIX-USE file had not been updated for the different set of source
2556 files that come with release 6. I also added a few comments about the C++
2557 wrapper.
2558
2559
2560 Version 6.0 07-Jun-05
2561 ---------------------
2562
2563 1. Some minor internal re-organization to help with my DFA experiments.
2564
2565 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2566 didn't matter for the library itself when fully configured, but did matter
2567 when compiling without UCP support, or within Exim, where the ucp files are
2568 not imported.
2569
2570 3. Refactoring of the library code to split up the various functions into
2571 different source modules. The addition of the new DFA matching code (see
2572 below) to a single monolithic source would have made it really too
2573 unwieldy, quite apart from causing all the code to be include in a
2574 statically linked application, when only some functions are used. This is
2575 relevant even without the DFA addition now that patterns can be compiled in
2576 one application and matched in another.
2577
2578 The downside of splitting up is that there have to be some external
2579 functions and data tables that are used internally in different modules of
2580 the library but which are not part of the API. These have all had their
2581 names changed to start with "_pcre_" so that they are unlikely to clash
2582 with other external names.
2583
2584 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2585 a different (DFA) algorithm. Although it is slower than the original
2586 function, it does have some advantages for certain types of matching
2587 problem.
2588
2589 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2590 including restarting after a partial match.
2591
2592 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2593 defined when compiling for Windows was sent to me. I have put it into the
2594 code, though I have no means of testing or verifying it.
2595
2596 7. Added the pcre_refcount() auxiliary function.
2597
2598 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2599 match before or at the first newline in the subject string. In pcretest,
2600 the /f option on a pattern can be used to set this.
2601
2602 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2603 would behave wrongly. This has been present in PCRE since release 4.0.
2604
2605 10. A number of changes to the pcregrep command:
2606
2607 (a) Refactored how -x works; insert ^(...)$ instead of setting
2608 PCRE_ANCHORED and checking the length, in preparation for adding
2609 something similar for -w.
2610
2611 (b) Added the -w (match as a word) option.
2612
2613 (c) Refactored the way lines are read and buffered so as to have more
2614 than one at a time available.
2615
2616 (d) Implemented a pcregrep test script.
2617
2618 (e) Added the -M (multiline match) option. This allows patterns to match
2619 over several lines of the subject. The buffering ensures that at least
2620 8K, or the rest of the document (whichever is the shorter) is available
2621 for matching (and similarly the previous 8K for lookbehind assertions).
2622
2623 (f) Changed the --help output so that it now says
2624
2625 -w, --word-regex(p)
2626
2627 instead of two lines, one with "regex" and the other with "regexp"
2628 because that confused at least one person since the short forms are the
2629 same. (This required a bit of code, as the output is generated
2630 automatically from a table. It wasn't just a text change.)
2631
2632 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2633 option but starts with a hyphen. Could be a pattern or a path name
2634 starting with a hyphen, for instance.
2635
2636 (h) "-" can be given as a file name to represent stdin.
2637
2638 (i) When file names are being printed, "(standard input)" is used for
2639 the standard input, for compatibility with GNU grep. Previously
2640 "<stdin>" was used.
2641
2642 (j) The option --label=xxx can be used to supply a name to be used for
2643 stdin when file names are being printed. There is no short form.
2644
2645 (k) Re-factored the options decoding logic because we are going to add
2646 two more options that take data. Such options can now be given in four
2647 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2648
2649 (l) Added the -A, -B, and -C options for requesting that lines of context
2650 around matches be printed.
2651
2652 (m) Added the -L option to print the names of files that do not contain
2653 any matching lines, that is, the complement of -l.
2654
2655 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2656 continue to scan other files.
2657
2658 (o) The -s option was incorrectly implemented. For compatibility with other
2659 greps, it now suppresses the error message for a non-existent or non-
2660 accessible file (but not the return code). There is a new option called
2661 -q that suppresses the output of matching lines, which was what -s was
2662 previously doing.
2663
2664 (p) Added --include and --exclude options to specify files for inclusion
2665 and exclusion when recursing.
2666
2667 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2668 Hopefully, it now does.
2669
2670 12. Missing cast in pcre_study().
2671
2672 13. Added an "uninstall" target to the makefile.
2673
2674 14. Replaced "extern" in the function prototypes in Makefile.in with
2675 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2676 world, but is set differently for Windows.
2677
2678 15. Added a second compiling function called pcre_compile2(). The only
2679 difference is that it has an extra argument, which is a pointer to an
2680 integer error code. When there is a compile-time failure, this is set
2681 non-zero, in addition to the error test pointer being set to point to an
2682 error message. The new argument may be NULL if no error number is required
2683 (but then you may as well call pcre_compile(), which is now just a
2684 wrapper). This facility is provided because some applications need a
2685 numeric error indication, but it has also enabled me to tidy up the way
2686 compile-time errors are handled in the POSIX wrapper.
2687
2688 16. Added VPATH=.libs to the makefile; this should help when building with one
2689 prefix path and installing with another. (Or so I'm told by someone who
2690 knows more about this stuff than I do.)
2691
2692 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2693 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2694 match everything, including newlines. This is not POSIX-compatible, but
2695 somebody wanted the feature. From pcretest it can be activated by using
2696 both the P and the s flags.
2697
2698 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2699
2700 19. libpcre.pc was being incorrectly installed as executable.
2701
2702 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2703 it now also looks for '\r' so that it will work unmodified on Windows.
2704
2705 21. Added Google's contributed C++ wrapper to the distribution.
2706
2707 22. Added some untidy missing memory free() calls in pcretest, to keep
2708 Electric Fence happy when testing.
2709
2710
2711
2712 Version 5.0 13-Sep-04
2713 ---------------------
2714
2715 1. Internal change: literal characters are no longer packed up into items
2716 containing multiple characters in a single byte-string. Each character
2717 is now matched using a separate opcode. However, there may be more than one
2718 byte in the character in UTF-8 mode.
2719
2720 2. The pcre_callout_block structure has two new fields: pattern_position and
2721 next_item_length. These contain the offset in the pattern to the next match
2722 item, and its length, respectively.
2723
2724 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2725 insertion of callouts before each pattern item. Added the /C option to
2726 pcretest to make use of this.
2727
2728 4. On the advice of a Windows user, the lines
2729
2730 #if defined(_WIN32) || defined(WIN32)
2731 _setmode( _fileno( stdout ), 0x8000 );
2732 #endif /* defined(_WIN32) || defined(WIN32) */
2733
2734 have been added to the source of pcretest. This apparently does useful
2735 magic in relation to line terminators.
2736
2737 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2738 for the benefit of those environments where the "b" makes a difference.
2739
2740 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2741 to know about it. I have put a hack into configure.in that adds in code
2742 to set GCC=yes if CC=icc. This seems to end up at a point in the
2743 generated configure script that is early enough to affect the setting of
2744 compiler options, which is what is needed, but I have no means of testing
2745 whether it really works. (The user who reported this had patched the
2746 generated configure script, which of course I cannot do.)
2747
2748 LATER: After change 22 below (new libtool files), the configure script
2749 seems to know about icc (and also ecc). Therefore, I have commented out
2750 this hack in configure.in.
2751
2752 7. Added support for pkg-config (2 patches were sent in).
2753
2754 8. Negated POSIX character classes that used a combination of internal tables
2755 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2756 [[:^ascii]]. Typically, they would match almost any characters. The other
2757 POSIX classes were not broken in this way.
2758
2759 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2760 to find the match, as PCRE was deluded into thinking that the match had to
2761 start at the start point or following a newline. The same bug applied to
2762 patterns with negative forward assertions or any backward assertions
2763 preceding ".*" at the start, unless the pattern required a fixed first
2764 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2765
2766 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2767 starting at the last subject character, bytes beyond the end of the subject
2768 string were read.
2769
2770 11. Renamed the variable "class" as "classbits" to make life easier for C++
2771 users. (Previously there was a macro definition, but it apparently wasn't
2772 enough.)
2773
2774 12. Added the new field "tables" to the extra data so that tables can be passed
2775 in at exec time, or the internal tables can be re-selected. This allows
2776 a compiled regex to be saved and re-used at a later time by a different
2777 program that might have everything at different addresses.
2778
2779 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2780 -R library as well as a -L library.
2781
2782 14. The debugging options of pcretest (-d on the command line or D on a
2783 pattern) showed incorrect output for anything following an extended class
2784 that contained multibyte characters and which was followed by a quantifier.
2785
2786 15. Added optional support for general category Unicode character properties
2787 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2788 support. It adds about 90K to the size of the library. The meanings of the
2789 inbuilt class escapes such as \d and \s have NOT been changed.
2790
2791 16. Updated pcredemo.c to include calls to free() to release the memory for the
2792 compiled pattern.
2793
2794 17. The generated file chartables.c was being created in the source directory
2795 instead of in the building directory. This caused the build to fail if the
2796 source directory was different from the building directory, and was
2797 read-only.
2798
2799 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2800 file. No doubt somebody will tell me if they don't make sense... Also added
2801 Dan Mooney's comments about building on OpenVMS.
2802
2803 19. Added support for partial matching via the PCRE_PARTIAL option for
2804 pcre_exec() and the \P data escape in pcretest.
2805
2806 20. Extended pcretest with 3 new pattern features:
2807
2808 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2809 write the compiled pattern to the file whose name is "rest-of-line".
2810 This is a straight binary dump of the data, with the saved pointer to
2811 the character tables forced to be NULL. The study data, if any, is
2812 written too. After writing, pcretest reads a new pattern.
2813
2814 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2815 compiled pattern from the given file. There must not be any
2816 occurrences of "<" in the file name (pretty unlikely); if there are,
2817 pcretest will instead treat the initial "<" as a pattern delimiter.
2818 After reading in the pattern, pcretest goes on to read data lines as
2819 usual.
2820
2821 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2822 and 16-bit fields in a compiled pattern, to simulate a pattern that
2823 was compiled on a host of opposite endianness.
2824
2825 21. The pcre-exec() function can now cope with patterns that were compiled on
2826 hosts of opposite endianness, with this restriction:
2827
2828 As for any compiled expression that is saved and used later, the tables
2829 pointer field cannot be preserved; the extra_data field in the arguments
2830 to pcre_exec() should be used to pass in a tables address if a value
2831 other than the default internal tables were used at compile time.
2832
2833 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2834 now diagnosed as an error. Previously, most of the time, a negative number
2835 would have been treated as zero, but if in addition "ovector" was passed as
2836 NULL, a crash could occur.
2837
2838 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2839 new versions from the libtool 1.5 distribution (the last one is a copy of
2840 a file called libtool.m4). This seems to have fixed the need to patch
2841 "configure" to support Darwin 1.3 (which I used to do). However, I still
2842 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2843 workstation).
2844
2845 24. Changed the PCRE licence to be the more standard "BSD" licence.
2846
2847
2848 Version 4.5 01-Dec-03
2849 ---------------------
2850
2851 1. There has been some re-arrangement of the code for the match() function so
2852 that it can be compiled in a version that does not call itself recursively.
2853 Instead, it keeps those local variables that need separate instances for
2854 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2855 needs to "recurse". Keeping track of where control must go is done by means
2856 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2857 hide most of the details from the main code, and operates only if
2858 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2859 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2860 operating.
2861
2862 To make it easier for callers to provide specially tailored get/free
2863 functions for this usage, two new functions, pcre_stack_malloc, and
2864 pcre_stack_free, are used. They are always called in strict stacking order,
2865 and the size of block requested is always the same.
2866
2867 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2868 PCRE has been compiled to use the stack or the heap for recursion. The
2869 -C option of pcretest uses this to show which version is compiled.
2870
2871 A new data escape \S, is added to pcretest; it causes the amounts of store
2872 obtained and freed by both kinds of malloc/free at match time to be added
2873 to the output.
2874
2875 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2876 what's available on my current Linux desktop machine.
2877
2878 3. When matching a UTF-8 string, the test for a valid string at the start has
2879 been extended. If start_offset is not zero, PCRE now checks that it points
2880 to a byte that is the start of a UTF-8 character. If not, it returns
2881 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2882 this is necessary because there may be backward assertions in the pattern.
2883 When matching the same subject several times, it may save resources to use
2884 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2885
2886 4. The code for checking the validity of UTF-8 strings has been tightened so
2887 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2888 containing "overlong sequences".
2889
2890 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2891 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2892 should have been "&", but it just so happened that all the cases this let
2893 through by mistake were picked up later in the function.
2894
2895 6. I had used a variable called "isblank" - this is a C99 function, causing
2896 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2897
2898 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2899 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2900 so the version is shown; (c) stop "make test" echoing "./RunTest".
2901
2902 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2903
2904 9. The prototype for memmove() for systems that don't have it was using
2905 size_t, but the inclusion of the header that defines size_t was later. I've
2906 moved the #includes for the C headers earlier to avoid this.
2907
2908 10. Added some adjustments to the code to make it easier to compiler on certain
2909 special systems:
2910
2911 (a) Some "const" qualifiers were missing.
2912 (b) Added the macro EXPORT before all exported functions; by default this
2913 is defined to be empty.
2914 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2915 that it reads its output file name as an argument instead of writing
2916 to the standard output and assuming this can be redirected.
2917
2918 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2919 class containing characters with values greater than 255, PCRE compilation
2920 went into a loop.
2921
2922 12. A recursive reference to a subpattern that was within another subpattern
2923 that had a minimum quantifier of zero caused PCRE to crash. For example,
2924 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2925 recursion. If the recursively-called subpattern itself had a zero repeat,
2926 that was OK.
2927
2928 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2929 buffer into which it was copied (for escape processing) was still set at
2930 1024, so long lines caused crashes.
2931
2932 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2933 "internal error: code overflow...". This applied to any character class
2934 that was followed by a possessive quantifier.
2935
2936 15. Modified the Makefile to add libpcre.la as a prerequisite for
2937 libpcreposix.la because I was told this is needed for a parallel build to
2938 work.
2939
2940 16. If a pattern that contained .* following optional items at the start was
2941 studied, the wrong optimizing data was generated, leading to matching
2942 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2943 matching string must start with a or b or c. The correct conclusion for
2944 this pattern is that a match can start with any character.
2945
2946
2947 Version 4.4 13-Aug-03
2948 ---------------------
2949
2950 1. In UTF-8 mode, a character class containing characters with values between
2951 127 and 255 was not handled correctly if the compiled pattern was studied.
2952 In fixing this, I have also improved the studying algorithm for such
2953 classes (slightly).
2954
2955 2. Three internal functions had redundant arguments passed to them. Removal
2956 might give a very teeny performance improvement.
2957
2958 3. Documentation bug: the value of the capture_top field in a callout is *one
2959 more than* the number of the hightest numbered captured substring.
2960
2961 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2962 in incorrectly linking with a previously installed version. They now link
2963 explicitly with libpcre.la.
2964
2965 5. configure.in no longer needs to recognize Cygwin specially.
2966
2967 6. A problem in pcre.in for Windows platforms is fixed.
2968
2969 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2970 pcretest, it used to include the size of the study block as part of its
2971 output. Unfortunately, the structure contains a field that has a different
2972 size on different hardware architectures. This meant that the tests that
2973 showed this size failed. As the block is currently always of a fixed size,
2974 this information isn't actually particularly useful in pcretest output, so
2975 I have just removed it.
2976
2977 8. Three pre-processor statements accidentally did not start in column 1.
2978 Sadly, there are *still* compilers around that complain, even though
2979 standard C has not required this for well over a decade. Sigh.
2980
2981 9. In pcretest, the code for checking callouts passed small integers in the
2982 callout_data field, which is a void * field. However, some picky compilers
2983 complained about the casts involved for this on 64-bit systems. Now
2984 pcretest passes the address of the small integer instead, which should get
2985 rid of the warnings.
2986
2987 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2988 both compile and run time, and gives an error if an invalid UTF-8 sequence
2989 is found. There is a option for disabling this check in cases where the
2990 string is known to be correct and/or the maximum performance is wanted.
2991
2992 11. In response to a bug report, I changed one line in Makefile.in from
2993
2994 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2995 to
2996 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2997
2998 to look similar to other lines, but I have no way of telling whether this
2999 is the right thing to do, as I do not use Windows. No doubt I'll get told
3000 if it's wrong...
3001
3002
3003 Version 4.3 21-May-03
3004 ---------------------
3005
3006 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3007 Makefile.
3008
3009 2. Some refactoring to improve the quality of the code:
3010
3011 (i) The utf8_table... variables are now declared "const".
3012
3013 (ii) The code for \cx, which used the "case flipping" table to upper case
3014 lower case letters, now just substracts 32. This is ASCII-specific,
3015 but the whole concept of \cx is ASCII-specific, so it seems
3016 reasonable.
3017
3018 (iii) PCRE was using its character types table to recognize decimal and
3019 hexadecimal digits in the pattern. This is silly, because it handles
3020 only 0-9, a-f, and A-F, but the character types table is locale-
3021 specific, which means strange things might happen. A private
3022 table is now used for this - though it costs 256 bytes, a table is
3023 much faster than multiple explicit tests. Of course, the standard
3024 character types table is still used for matching digits in subject
3025 strings against \d.
3026
3027 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3028 ending in _t are). So I've renamed it as ESC_tee.
3029
3030 3. The first argument for regexec() in the POSIX wrapper should have been
3031 defined as "const".
3032
3033 4. Changed pcretest to use malloc() for its buffers so that they can be
3034 Electric Fenced for debugging.
3035
3036 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3037 to read one or more bytes before the start of the subject string. Often this
3038 had no effect on PCRE's behaviour, but in some circumstances it could
3039 provoke a segmentation fault.
3040
3041 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3042 to try to read one or more bytes before the start of the subject string.
3043
3044 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3045 UTF-8 support could misbehave in various ways if the subject string
3046 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3047 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3048 back over UTF-8 characters.)
3049
3050
3051 Version 4.2 14-Apr-03
3052 ---------------------
3053
3054 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3055
3056 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3057 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3058 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3059 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3060 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3061 and BUILD_EXEEXT
3062 Note: automatic setting of the BUILD variables is not yet working
3063 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3064 compile-time but not at link-time
3065 [LINK]: use for linking executables only
3066 make different versions for Windows and non-Windows
3067 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3068 libraries
3069 [LINK_FOR_BUILD]: new variable
3070 [OBJEXT]: use throughout
3071 [EXEEXT]: use throughout
3072 <winshared>: new target
3073 <wininstall>: new target
3074 <dftables.o>: use native compiler
3075 <dftables>: use native linker
3076 <install>: handle Windows platform correctly
3077 <clean>: ditto
3078 <check>: ditto
3079 copy DLL to top builddir before testing
3080
3081 As part of these changes, -no-undefined was removed again. This was reported
3082 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3083 in any case.
3084
3085 3. Some tidies to get rid of compiler warnings:
3086
3087 . In the match_data structure, match_limit was an unsigned long int, whereas
3088 match_call_count was an int. I've made them both unsigned long ints.
3089
3090 . In pcretest the fact that a const uschar * doesn't automatically cast to
3091 a void * provoked a warning.
3092
3093 . Turning on some more compiler warnings threw up some "shadow" variables
3094 and a few more missing casts.
3095
3096 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3097 option, a class that contained a single character with a value between 128
3098 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3099
3100 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3101 option, a class that contained several characters, but with at least one
3102 whose value was between 128 and 255 caused PCRE to crash.
3103
3104
3105 Version 4.1 12-Mar-03
3106 ---------------------
3107
3108 1. Compiling with gcc -pedantic found a couple of places where casts were
3109 needed, and a string in dftables.c that was longer than standard compilers are
3110 required to support.
3111
3112 2. Compiling with Sun's compiler found a few more places where the code could
3113 be tidied up in order to avoid warnings.
3114
3115 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3116 first of these names is deprecated in the latest Autoconf in favour of the name
3117 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3118 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3119 analogy I have changed it to CFLAGS_FOR_BUILD.
3120
3121 4. Added -no-undefined to the linking command in the Makefile, because this is
3122 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3123 linking step for the pcreposix library.
3124
3125 5. PCRE was failing to diagnose the case of two named groups with the same
3126 name.
3127
3128 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3129 literal character that is needed in the subject for a match, and scans along to
3130 ensure that it is present before embarking on the full matching process. This
3131 saves time in cases of nested unlimited repeats that are never going to match.
3132 Problem: the scan can take a lot of time if the subject is very long (e.g.
3133 megabytes), thus penalizing straightforward matches. It is now done only if the
3134 amount of subject to be scanned is less than 1000 bytes.
3135
3136 7. A lesser problem with the same optimization is that it was recording the
3137 first character of an anchored pattern as "needed", thus provoking a search
3138 right along the subject, even when the first match of the pattern was going to
3139 fail. The "needed" character is now not set for anchored patterns, unless it
3140 follows something in the pattern that is of non-fixed length. Thus, it still
3141 fulfils its original purpose of finding quick non-matches in cases of nested
3142 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3143
3144
3145 Version 4.0 17-Feb-03
3146 ---------------------
3147
3148 1. If a comment in an extended regex that started immediately after a meta-item
3149 extended to the end of string, PCRE compiled incorrect data. This could lead to
3150 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3151
3152 2. Moved to autoconf 2.53 and libtool 1.4.2.
3153
3154 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3155 the special perltest8 script is no longer needed - all the tests can be run
3156 from a single perltest script.
3157
3158 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3159 by \s. It has now been removed in PCRE. This means it isn't recognized as
3160 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3161 class [:space:] *does* include VT, thereby creating a mess.
3162
3163 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3164 space and tab.
3165
3166 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3167 its new features into the main test script, reducing the number of scripts.
3168
3169 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3170 were backward compatible, and made the (?i) apply to the whole pattern, as if
3171 /i were given. Now it behaves more logically, and applies the option setting
3172 only to what follows. PCRE has been changed to follow suit. However, if it
3173 finds options settings right at the start of the pattern, it extracts them into
3174 the global options, as before. Thus, they show up in the info data.
3175
3176 8. Added support for the \Q...\E escape sequence. Characters in between are
3177 treated as literals. This is slightly different from Perl in that $ and @ are
3178 also handled as literals inside the quotes. In Perl, they will cause variable
3179 interpolation. Note the following examples:
3180
3181 Pattern PCRE matches Perl matches
3182
3183 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3184 \Qabc\$xyz\E abc\$xyz abc\$xyz
3185 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3186
3187 For compatibility with Perl, \Q...\E sequences are recognized inside character
3188 classes as well as outside them.
3189
3190 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3191 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3192 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3193 signed/unsigned warnings.
3194
3195 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3196 option for pcretest, so I've replaced it by a simple function that does just
3197 that job.
3198
3199 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3200 "pcregrep -".
3201
3202 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3203 Java package. This provides some syntactic sugar for simple cases of what my
3204 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3205 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3206 item, you can use this simplified notation. Note that only makes sense with
3207 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3208 greediness, whatever the setting of the PCRE_UNGREEDY option.
3209
3210 13. A change of greediness default within a pattern was not taking effect at
3211 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3212 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3213 was abstracted outside.
3214
3215 14. PCRE now supports the \G assertion. It is true when the current matching
3216 position is at the start point of the match. This differs from \A when the
3217 starting offset is non-zero. Used with the /g option of pcretest (or similar
3218 code), it works in the same way as it does for Perl's /g option. If all
3219 alternatives of a regex begin with \G, the expression is anchored to the start
3220 match position, and the "anchored" flag is set in the compiled expression.
3221
3222 15. Some bugs concerning the handling of certain option changes within patterns
3223 have been fixed. These applied to options other than (?ims). For example,
3224 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3225 been the other way round. Some of this was related to change 7 above.
3226
3227 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3228 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3229 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3230 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3231
3232 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3233 mode. Unlike ".", it always matches newline, whatever the setting of
3234 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3235 assertions. Perl allows it, but it doesn't (in general) work because it can't
3236 calculate the length of the lookbehind. At least, that's the case for Perl
3237 5.8.0 - I've been told they are going to document that it doesn't work in
3238 future.
3239
3240 18. Added an error diagnosis for escapes that PCRE does not support: these are
3241 \L, \l, \N, \P, \p, \U, \u, and \X.
3242
3243 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3244 reading past the end of the pattern in cases such as /[abcd/.
3245
3246 20. PCRE was getting more memory than necessary for patterns with classes that
3247 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3248
3249 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3250 compiling PCRE for use with Virtual Pascal.
3251
3252 22. Small fix to the Makefile to make it work properly if the build is done
3253 outside the source tree.
3254
3255 23. Added a new extension: a condition to go with recursion. If a conditional
3256 subpattern starts with (?(R) the "true" branch is used if recursion has
3257 happened, whereas the "false" branch is used only at the top level.
3258
3259 24. When there was a very long string of literal characters (over 255 bytes
3260 without UTF support, over 250 bytes with UTF support), the computation of how
3261 much memory was required could be incorrect, leading to segfaults or other
3262 strange effects.
3263
3264 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3265 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3266 there was a subsequent back reference to those brackets. This meant that, for
3267 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3268 possible to check for precisely this case. All we can do is abandon the
3269 optimization if .* occurs inside capturing brackets when there are any back
3270 references whatsoever. (See below for a better fix that came later.)
3271
3272 26. The handling of the optimization for finding the first character of a
3273 non-anchored pattern, and for finding a character that is required later in the
3274 match were failing in some cases. This didn't break the matching; it just
3275 failed to optimize when it could. The way this is done has been re-implemented.
3276
3277 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3278
3279 28. Added a new feature that provides some of the functionality that Perl
3280 provides with (?{...}). The facility is termed a "callout". The way it is done
3281 in PCRE is for the caller to provide an optional function, by setting
3282 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3283 global variable. By default it is unset, which disables all calling out. To get
3284 the function called, the regex must include (?C) at appropriate points. This
3285 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3286 This provides a means of identifying different callout points. When PCRE
3287 reaches such a point in the regex, if pcre_callout has been set, the external
3288 function is called. It is provided with data in a structure called
3289 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3290 matching continues; if it returns a non-zero value, the match at the current
3291 point fails. However, backtracking will occur if possible. [This was changed
3292 later and other features added - see item 49 below.]
3293
3294 29. pcretest is upgraded to test the callout functionality. It provides a
3295 callout function that displays information. By default, it shows the start of
3296 the match and the current position in the text. There are some new data escapes
3297 to vary what happens:
3298
3299 \C+ in addition, show current contents of captured substrings
3300 \C- do not supply a callout function
3301 \C!n return 1 when callout number n is reached
3302 \C!n!m return 1 when callout number n is reached for the mth time
3303
3304 30. If pcregrep was called with the -l option and just a single file name, it
3305 output "<stdin>" if a match was found, instead of the file name.
3306
3307 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3308 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3309 pcre_exec(). This saves a malloc/free per call. The default value of
3310 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3311 when configuring.
3312
3313 32. The default maximum size of a compiled pattern is 64K. There have been a
3314 few cases of people hitting this limit. The code now uses macros to handle the
3315 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3316 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3317 configuring. Tests 2 and 5 work only with 2-byte links because they output
3318 debugging information about compiled patterns.
3319
3320 33. Internal code re-arrangements:
3321
3322 (a) Moved the debugging function for printing out a compiled regex into
3323 its own source file (printint.c) and used #include to pull it into
3324 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3325 separate copies.
3326
3327 (b) Defined the list of op-code names for debugging as a macro in
3328 internal.h so that it is next to the definition of the opcodes.
3329
3330 (c) Defined a table of op-code lengths for simpler skipping along compiled
3331 code. This is again a macro in internal.h so that it is next to the
3332 definition of the opcodes.
3333
3334 34. Added support for recursive calls to individual subpatterns, along the
3335 lines of Robin Houston's patch (but implemented somewhat differently).
3336
3337 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3338 allow it to read and process whole directories in Win32. This code was
3339 contributed by Lionel Fourquaux; it has not been tested by me.
3340
3341 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3342 used to name a group. Names consist of alphanumerics and underscores, and must
3343 be unique. Back references use the syntax (?P=name) and recursive calls use
3344 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3345 numbers. The function pcre_fullinfo() can be used after compilation to extract
3346 a name/number map. There are three relevant calls:
3347
3348 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3349 PCRE_INFO_NAMECOUNT yields the number of entries
3350 PCRE_INFO_NAMETABLE yields a pointer to the map.
3351
3352 The map is a vector of fixed-size entries. The size of each entry depends on
3353 the length of the longest name used. The first two bytes of each entry are the
3354 group number, most significant byte first. There follows the corresponding
3355 name, zero terminated. The names are in alphabetical order.
3356
3357 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3358 case instead of 255. Making it the same both with and without UTF-8 support
3359 means that the same test output works with both.
3360
3361 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3362 calling malloc() with a zero argument.
3363
3364 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3365 optimization. I've improved things by keeping a bitmap of backreferences with
3366 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3367 fact referenced, the optimization can be applied. It is unlikely that a
3368 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3369 the match to follow \n) will appear inside brackets with a number greater than
3370 31, but if it does, any back reference > 31 suppresses the optimization.
3371
3372 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3373 of disabling numbered capturing parentheses. Any opening parenthesis that is
3374 not followed by ? behaves as if it were followed by ?: but named parentheses
3375 can still be used for capturing (and they will acquire numbers in the usual
3376 way).
3377
3378 41. Redesigned the return codes from the match() function into yes/no/error so
3379 that errors can be passed back from deep inside the nested calls. A malloc
3380 failure while inside a recursive subpattern call now causes the
3381 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3382
3383 42. It is now possible to set a limit on the number of times the match()
3384 function is called in a call to pcre_exec(). This facility makes it possible to
3385 limit the amount of recursion and backtracking, though not in a directly
3386 obvious way, because the match() function is used in a number of different
3387 circumstances. The count starts from zero for each position in the subject
3388 string (for non-anchored patterns). The default limit is, for compatibility, a
3389 large number, namely 10 000 000. You can change this in two ways:
3390
3391 (a) When configuring PCRE before making, you can use --with-match-limit=n
3392 to set a default value for the compiled library.
3393
3394 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3395 a different value is set. See 45 below.
3396
3397 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3398
3399 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3400 of things that can be changed at compile time. The first argument specifies
3401 what is wanted and the second points to where the information is to be placed.
3402 The current list of available information is:
3403
3404 PCRE_CONFIG_UTF8
3405
3406 The output is an integer that is set to one if UTF-8 support is available;
3407 otherwise it is set to zero.
3408
3409 PCRE_CONFIG_NEWLINE
3410
3411 The output is an integer that it set to the value of the code that is used for
3412 newline. It is either LF (10) or CR (13).
3413
3414 PCRE_CONFIG_LINK_SIZE
3415
3416 The output is an integer that contains the number of bytes used for internal
3417 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3418
3419 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3420
3421 The output is an integer that contains the threshold above which the POSIX
3422 interface uses malloc() for output vectors. See item 31 above.
3423
3424 PCRE_CONFIG_MATCH_LIMIT
3425
3426 The output is an unsigned integer that contains the default limit of the number
3427 of match() calls in a pcre_exec() execution. See 42 above.
3428
3429 44. pcretest has been upgraded by the addition of the -C option. This causes it
3430 to extract all the available output from the new pcre_config() function, and to
3431 output it. The program then exits immediately.
3432
3433 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3434 order to support additional features. One way would have been to define
3435 pcre_exec2() (for example) with extra arguments, but this would not have been
3436 extensible, and would also have required all calls to the original function to
3437 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3438 is used for passing in "extra" data from pcre_study().
3439
3440 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3441 contains the following fields:
3442
3443 flags a bitmap indicating which of the following fields are set
3444 study_data opaque data from pcre_study()
3445 match_limit a way of specifying a limit on match() calls for a specific
3446 call to pcre_exec()
3447 callout_data data for callouts (see 49 below)
3448
3449 The flag bits are also defined in pcre.h, and are
3450
3451 PCRE_EXTRA_STUDY_DATA
3452 PCRE_EXTRA_MATCH_LIMIT
3453 PCRE_EXTRA_CALLOUT_DATA
3454
3455 The pcre_study() function now returns one of these new pcre_extra blocks, with
3456 the actual study data pointed to by the study_data field, and the
3457 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3458 before. That is, this change is entirely upwards-compatible and requires no
3459 change to existing code.
3460
3461 If you want to pass in additional data to pcre_exec(), you can either place it
3462 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3463 block.
3464
3465 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3466 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3467 times with different match limits, until it finds the minimum value needed for
3468 pcre_exec() to complete. The value is then output. This can be instructive; for
3469 most simple matches the number is quite small, but for pathological cases it
3470 gets very large very quickly.
3471
3472 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3473 returns the size of the data block pointed to by the study_data field in a
3474 pcre_extra block, that is, the value that was passed as the argument to
3475 pcre_malloc() when PCRE was getting memory in which to place the information
3476 created by pcre_study(). The fourth argument should point to a size_t variable.
3477 pcretest has been extended so that this information is shown after a successful
3478 pcre_study() call when information about the compiled regex is being displayed.
3479
3480 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3481 because what follows is always an absolute path. (Later: it turns out that this
3482 is more than cosmetic for MinGW, because it doesn't like empty path
3483 components.)
3484
3485 49. Some changes have been made to the callout feature (see 28 above):
3486
3487 (i) A callout function now has three choices for what it returns:
3488
3489 0 => success, carry on matching
3490 > 0 => failure at this point, but backtrack if possible
3491 < 0 => serious error, return this value from pcre_exec()
3492
3493 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3494 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3495 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3496 use by callout functions. It will never be used by PCRE itself.
3497
3498 (ii) The pcre_extra structure (see 45 above) has a void * field called
3499 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3500 pcre_callout_block structure has a field of the same name. The contents of
3501 the field passed in the pcre_extra structure are passed to the callout
3502 function in the corresponding field in the callout block. This makes it
3503 easier to use the same callout-containing regex from multiple threads. For
3504 testing, the pcretest program has a new data escape
3505
3506 \C*n pass the number n (may be negative) as callout_data
3507
3508 If the callout function in pcretest receives a non-zero value as
3509 callout_data, it returns that value.
3510
3511 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3512 there were some redundant $(CFLAGS) in commands that are now specified as
3513 $(LINK), which already includes $(CFLAGS).
3514
3515 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3516 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3517 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3518 one-byte characters throughout. Note that case-insensitive matching applies
3519 only to characters whose values are less than 256. PCRE doesn't support the
3520 notion of cases for higher-valued characters.
3521
3522 (i) A character class whose characters are all within 0-255 is handled as
3523 a bit map, and the map is inverted for negative classes. Previously, a
3524 character > 255 always failed to match such a class; however it should
3525 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3526
3527 (ii) A negated character class with a single character < 255 is coded as
3528 "not this character" (OP_NOT). This wasn't working properly when the test
3529 character was multibyte, either singly or repeated.
3530
3531 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3532 mode, for example: \x{100}{2,3}.
3533
3534 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3535 singly or repeated) now correctly test multibyte characters. However,
3536 PCRE doesn't recognize any characters with values greater than 255 as
3537 digits, spaces, or word characters. Such characters always match \D, \S,
3538 and \W, and never match \d, \s, or \w.
3539
3540 (v) Classes may now contain characters and character ranges with values
3541 greater than 255. For example: [ab\x{100}-\x{400}].
3542
3543 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3544 PCRE in UTF-8 mode.
3545
3546 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3547 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3548 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3549 value.)
3550
3551 53. The single man page has become too large. I have therefore split it up into
3552 a number of separate man pages. These also give rise to individual HTML pages;
3553 these are now put in a separate directory, and there is an index.html page that
3554 lists them all. Some hyperlinking between the pages has been installed.
3555
3556 54. Added convenience functions for handling named capturing parentheses.
3557
3558 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3559 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3560 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3561 are faulted.
3562
3563 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3564 calling configure. These values are used when compiling the dftables.c program
3565 which is run to generate the source of the default character tables. They
3566 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3567 you will need to set these values.
3568
3569 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3570
3571
3572 Version 3.9 02-Jan-02
3573 ---------------------
3574
3575 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3576
3577 2. If --disable-static was given, the building process failed when trying to
3578 build pcretest and pcregrep. (For some reason it was using libtool to compile
3579 them, which is not right, as they aren't part of the library.)
3580
3581
3582 Version 3.8 18-Dec-01
3583 ---------------------
3584
3585 1. The experimental UTF-8 code was completely screwed up. It was packing the
3586 bytes in the wrong order. How dumb can you get?
3587
3588
3589 Version 3.7 29-Oct-01
3590 ---------------------
3591
3592 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3593 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3594 this didn't happen under Solaris 8, where I normally test things.
3595
3596 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3597 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3598 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3599 this fix an hour or so after the initial 3.7 release.)
3600
3601
3602 Version 3.6 23-Oct-01
3603 ---------------------
3604
3605 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3606 offsets passed as NULL with zero offset count.
3607
3608 2. The config.guess and config.sub files had not been updated when I moved to
3609 the latest autoconf.
3610
3611
3612 Version 3.5 15-Aug-01
3613 ---------------------
3614
3615 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3616 had been forgotten.
3617
3618 2. By using declared but undefined structures, we can avoid using "void"
3619 definitions in pcre.h while keeping the internal definitions of the structures
3620 private.
3621
3622 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3623 user point of view, this means that both static and shared libraries are built
3624 by default, but this can be individually controlled. More of the work of
3625 handling this static/shared cases is now inside libtool instead of PCRE's make
3626 file.
3627
3628 4. The pcretest utility is now installed along with pcregrep because it is
3629 useful for users (to test regexs) and by doing this, it automatically gets
3630 relinked by libtool. The documentation has been turned into a man page, so
3631 there are now .1, .txt, and .html versions in /doc.
3632
3633 5. Upgrades to pcregrep:
3634 (i) Added long-form option names like gnu grep.
3635 (ii) Added --help to list all options with an explanatory phrase.
3636 (iii) Added -r, --recursive to recurse into sub-directories.
3637 (iv) Added -f, --file to read patterns from a file.
3638
3639 6. pcre_exec() was referring to its "code" argument before testing that
3640 argument for NULL (and giving an error if it was NULL).
3641
3642 7. Upgraded Makefile.in to allow for compiling in a different directory from
3643 the source directory.
3644
3645 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3646 options bits, the pointer it was passed was to an int instead of to an unsigned
3647 long int. This mattered only on 64-bit systems.
3648
3649 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3650 generated) instead of pcre.in, which it its source. Also made the same change
3651 in several of the .c files.
3652
3653 10. A new release of gcc defines printf() as a macro, which broke pcretest
3654 because it had an ifdef in the middle of a string argument for printf(). Fixed
3655 by using separate calls to printf().
3656
3657 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3658 script, to force use of CR or LF instead of \n in the source. On non-Unix
3659 systems, the value can be set in config.h.
3660
3661 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3662 absolute limit. Changed the text of the error message to make this clear, and
3663 likewise updated the man page.
3664
3665 13. The limit of 99 on the number of capturing subpatterns has been removed.
3666 The new limit is 65535, which I hope will not be a "real" limit.
3667
3668
3669 Version 3.4 22-Aug-00
3670 ---------------------
3671
3672 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3673
3674 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3675
3676
3677 Version 3.3 01-Aug-00
3678 ---------------------
3679
3680 1. If an octal character was given, but the value was greater than \377, it
3681 was not getting masked to the least significant bits, as documented. This could
3682 lead to crashes in some systems.
3683
3684 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3685 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3686
3687 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3688 These just pass their arguments on to (pcre_free)(), but they are provided
3689 because some uses of PCRE bind it to non-C systems that can call its functions,
3690 but cannot call free() or pcre_free() directly.
3691
3692 4. Add "make test" as a synonym for "make check". Corrected some comments in
3693 the Makefile.
3694
3695 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3696 Makefile.
3697
3698 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3699 command called pgrep for grepping around the active processes.
3700
3701 7. Added the beginnings of support for UTF-8 character strings.
3702
3703 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3704 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3705 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3706 out for the ar command.)
3707
3708
3709 Version 3.2 12-May-00
3710 ---------------------
3711
3712 This is purely a bug fixing release.
3713
3714 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3715 of ZA. This was just one example of several cases that could provoke this bug,
3716 which was introduced by change 9 of version 2.00. The code for breaking
3717 infinite loops after an iteration that matches an empty string was't working
3718 correctly.
3719
3720 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3721 when matched against abbab (for example). After matching an empty string, it
3722 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3723 caused it to match further down the string than it should.
3724
3725 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3726 was there because it doesn't seem to be needed, and it causes trouble on some
3727 systems, as it is not a Standard C header. It has been removed.
3728
3729 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3730 were reported on the Macintosh. The changes were from
3731
3732 while ((c = *(++ptr)) != 0 && c != '\n');
3733 to
3734 while ((c = *(++ptr)) != 0 && c != '\n') ;
3735
3736 Totally extraordinary, but if that's what it takes...
3737
3738 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3739 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3740 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3741 assumes the way PCRE uses memmove() (always moving upwards).
3742
3743 6. PCRE is being used in one environment where strchr() is not available. There
3744 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3745 faster code anyway.
3746
3747
3748 Version 3.1 09-Feb-00
3749 ---------------------
3750
3751 The only change in this release is the fixing of some bugs in Makefile.in for
3752 the "install" target:
3753
3754 (1) It was failing to install pcreposix.h.
3755
3756 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3757
3758
3759 Version 3.0 01-Feb-00
3760 ---------------------
3761
3762 1. Add support for the /+ modifier to perltest (to output $` like it does in
3763 pcretest).
3764
3765 2. Add support for the /g modifier to perltest.
3766
3767 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3768 matches null strings.
3769
3770 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3771 pattern. Perl treats empty patterns specially - it reuses the most recent
3772 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3773 effect.
3774
3775 5. The POSIX interface was broken in that it was just handing over the POSIX
3776 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3777 required a bigger vector, with some working space on the end. This means that
3778 the POSIX wrapper now has to get and free some memory, and copy the results.
3779
3780 6. Added some simple autoconf support, placing the test data and the
3781 documentation in separate directories, re-organizing some of the
3782 information files, and making it build pcre-config (a GNU standard). Also added
3783 libtool support for building PCRE as a shared library, which is now the
3784 default.
3785
3786 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3787 09 are not valid octal constants. Single digits will be used for minor values
3788 less than 10.
3789
3790 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3791 existing programs that set these in the POSIX interface can use PCRE without
3792 modification.
3793
3794 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3795 return all that pcre_info() returns, plus additional data. The pcre_info()
3796 function is retained for compatibility, but is considered to be obsolete.
3797
3798 10. Added experimental recursion feature (?R) to handle one common case that
3799 Perl 5.6 will be able to do with (?p{...}).
3800
3801 11. Added support for POSIX character classes like [:alpha:], which Perl is
3802 adopting.
3803
3804
3805 Version 2.08 31-Aug-99
3806 ----------------------
3807
3808 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3809 trying to match at the startoffset position, but instead was moving forward to
3810 the next newline as if a previous match had failed.
3811
3812 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3813 and could get into a loop if a null string was matched other than at the start
3814 of the subject.
3815
3816 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3817 be distinguished at compile time, and for completeness also added PCRE_DATE.
3818
3819 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3820 in GnuWin32 environments.
3821
3822
3823 Version 2.07 29-Jul-99
3824 ----------------------
3825
3826 1. The documentation is now supplied in plain text form and HTML as well as in
3827 the form of man page sources.
3828
3829 2. C++ compilers don't like assigning (void *) values to other pointer types.
3830 In particular this affects malloc(). Although there is no problem in Standard
3831 C, I've put in casts to keep C++ compilers happy.
3832
3833 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3834 should be (const char *).
3835
3836 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3837 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3838 However, I haven't made this a standard facility. The documentation doesn't
3839 mention it, and the Makefile doesn't support it.
3840
3841 5. The Makefile now contains an "install" target, with editable destinations at
3842 the top of the file. The pcretest program is not installed.
3843
3844 6. pgrep -V now gives the PCRE version number and date.
3845
3846 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3847 causing the entire string to be ignored, instead of just the last character.
3848
3849 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3850 non-matching string, it can take a very, very long time, even for strings of
3851 quite modest length, because of the nested recursion. PCRE now does better in
3852 some of these cases. It does this by remembering the last required literal
3853 character in the pattern, and pre-searching the subject to ensure it is present
3854 before running the real match. In other words, it applies a heuristic to detect
3855 some types of certain failure quickly, and in the above example, if presented
3856 with a string that has no trailing " it gives "no match" very quickly.
3857
3858 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3859 other alternatives are tried instead.
3860
3861
3862 Version 2.06 09-Jun-99
3863 ----------------------
3864
3865 1. Change pcretest's output for amount of store used to show just the code
3866 space, because the remainder (the data block) varies in size between 32-bit and
3867 64-bit systems.
3868
3869 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3870 start matching at. This allows lookbehinds to work when searching for multiple
3871 occurrences in a string.
3872
3873 3. Added additional options to pcretest for testing multiple occurrences:
3874
3875 /+ outputs the rest of the string that follows a match
3876 /g loops for multiple occurrences, using the new startoffset argument
3877 /G loops for multiple occurrences by passing an incremented pointer
3878
3879 4. PCRE wasn't doing the "first character" optimization for patterns starting
3880 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3881 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3882 the letter 'x'. On long subject strings, this gives a significant speed-up.
3883
3884
3885 Version 2.05 21-Apr-99
3886 ----------------------
3887
3888 1. Changed the type of magic_number from int to long int so that it works
3889 properly on 16-bit systems.
3890
3891 2. Fixed a bug which caused patterns starting with .* not to work correctly
3892 when the subject string contained newline characters. PCRE was assuming
3893 anchoring for such patterns in all cases, which is not correct because .* will
3894 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3895 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3896 must be retried after every newline in the subject.
3897
3898
3899 Version 2.04 18-Feb-99
3900 ----------------------
3901
3902 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3903 computation of the store needed to hold the pattern was incorrect (too large).
3904 If such patterns were nested a few deep, this could multiply and become a real
3905 problem.
3906
3907 2. Added /M option to pcretest to show the memory requirement of a specific
3908 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3909
3910 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3911 compiled in such a way that the backtracking after subsequent failure was
3912 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3913 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3914
3915
3916 Version 2.03 02-Feb-99
3917 ----------------------
3918
3919 1. Fixed typo and small mistake in man page.
3920
3921 2. Added 4th condition (GPL supersedes if conflict) and created separate
3922 LICENCE file containing the conditions.
3923
3924 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3925 Perl, that is the internal \ allows the delimiter to be included in the
3926 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3927 the final delimiter, add \ to the end of the pattern (to test the error).
3928
3929 4. Added the convenience functions for extracting substrings after a successful
3930 match. Updated pcretest to make it able to test these functions.
3931
3932
3933 Version 2.02 14-Jan-99
3934 ----------------------
3935
3936 1. Initialized the working variables associated with each extraction so that
3937 their saving and restoring doesn't refer to uninitialized store.
3938
3939 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3940 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3941 fix the problem.
3942
3943 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3944 calls, and wasn't printing the correct value for compiling calls. Increased the
3945 default value of LOOPREPEAT, and the number of significant figures in the
3946 times.
3947
3948 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3949
3950 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3951 a building problem on Windows NT with a FAT file system.
3952
3953
3954 Version 2.01 21-Oct-98
3955 ----------------------
3956
3957 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3958 to character tables built by pcre_maketables() in the current locale. If NULL
3959 is passed, the default tables are used.
3960
3961
3962 Version 2.00 24-Sep-98
3963 ----------------------
3964
3965 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3966 it any more.
3967
3968 2. Allow quantification of (?>) groups, and make it work correctly.
3969
3970 3. The first character computation wasn't working for (?>) groups.
3971
3972 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3973 end of the subject) and add 5.005's \z, which really does match only at the
3974 very end of the subject.
3975
3976 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3977
3978 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3979 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3980 localized options. All options to pcre_study() were also removed.
3981
3982 7. Add other new features from 5.005:
3983
3984 $(?<= positive lookbehind
3985 $(?<! negative lookbehind
3986 (?imsx-imsx) added the unsetting capability
3987 such a setting is global if at outer level; local otherwise
3988 (?imsx-imsx:) non-capturing groups with option setting
3989 (?(cond)re|re) conditional pattern matching
3990
3991 A backreference to itself in a repeated group matches the previous
3992 captured string.
3993
3994 8. General tidying up of studying (both automatic and via "study")
3995 consequential on the addition of new assertions.
3996
3997 9. As in 5.005, unlimited repeated groups that could match an empty substring
3998 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3999 runtime if any iteration does actually match an empty substring.
4000
4001 10. Include the RunTest script in the distribution.
4002
4003 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
4004 discrepancies, some of which were old and were also with respect to 5.004. They
4005 have now been fixed.
4006
4007
4008 Version 1.09 28-Apr-98
4009 ----------------------
4010
4011 1. A negated single character class followed by a quantifier with a minimum
4012 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
4013 program crashes, or just wrong answers. This did not apply to negated classes
4014 containing more than one character, or to minima other than one.
4015
4016
4017 Version 1.08 27-Mar-98
4018 ----------------------
4019
4020 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
4021
4022 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
4023 latter must appear before anything that relies on it in the pattern.
4024
4025
4026 Version 1.07 16-Feb-98
4027 ----------------------
4028
4029 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
4030 repeat of a potentially empty string).
4031
4032
4033 Version 1.06 23-Jan-98
4034 ----------------------
4035
4036 1. Added Markus Oberhumer's little patches for C++.
4037
4038 2. Literal strings longer than 255 characters were broken.
4039
4040
4041 Version 1.05 23-Dec-97
4042 ----------------------
4043
4044 1. Negated character classes containing more than one character were failing if
4045 PCRE_CASELESS was set at run time.
4046
4047
4048 Version 1.04 19-Dec-97
4049 ----------------------
4050
4051 1. Corrected the man page, where some "const" qualifiers had been omitted.
4052
4053 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4054 input syntax.
4055
4056 3. Fixed memory leak which occurred when a regex with back references was
4057 matched with an offsets vector that wasn't big enough. The temporary memory
4058 that is used in this case wasn't being freed if the match failed.
4059
4060 4. Tidied pcretest to ensure it frees memory that it gets.
4061
4062 5. Temporary memory was being obtained in the case where the passed offsets
4063 vector was exactly big enough.
4064
4065 6. Corrected definition of offsetof() from change 5 below.
4066
4067 7. I had screwed up change 6 below and broken the rules for the use of
4068 setjmp(). Now fixed.
4069
4070
4071 Version 1.03 18-Dec-97
4072 ----------------------
4073
4074 1. A erroneous regex with a missing opening parenthesis was correctly
4075 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4076 on some systems.
4077
4078 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4079 it was reported that one broken compiler failed on the former because "code" is
4080 also an independent variable.
4081
4082 3. The erroneous regex a[]b caused an array overrun reference.
4083
4084 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4085 fail on data ending with that character. (It was going on too far, and checking
4086 the next character, typically a binary zero.) This was specific to the
4087 optimized code for single-character negative classes.
4088
4089 5. Added a contributed patch from the TIN world which does the following:
4090
4091 + Add an undef for memmove, in case the the system defines a macro for it.
4092
4093 + Add a definition of offsetof(), in case there isn't one. (I don't know
4094 the reason behind this - offsetof() is part of the ANSI standard - but
4095 it does no harm).
4096
4097 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4098 most of the places where whitespace preceded '#'. I have given up and
4099 allowed the remaining 2 cases to be at the margin.
4100
4101 + Rename some variables in pcre to eliminate shadowing. This seems very
4102 pedantic, but does no harm, of course.
4103
4104 6. Moved the call to setjmp() into its own function, to get rid of warnings
4105 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4106
4107 7. Constructs such as \d{8,} were compiling into the equivalent of
4108 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4109 outcome, but in this particular case used more store than had been allocated,
4110 which caused the bug to be discovered because it threw up an internal error.
4111
4112 8. The debugging code in both pcre and pcretest for outputting the compiled
4113 form of a regex was going wrong in the case of back references followed by
4114 curly-bracketed repeats.
4115
4116
4117 Version 1.02 12-Dec-97
4118 ----------------------
4119
4120 1. Typos in pcre.3 and comments in the source fixed.
4121
4122 2. Applied a contributed patch to get rid of places where it used to remove
4123 'const' from variables, and fixed some signed/unsigned and uninitialized
4124 variable warnings.
4125
4126 3. Added the "runtest" target to Makefile.
4127
4128 4. Set default compiler flag to -O2 rather than just -O.
4129
4130
4131 Version 1.01 19-Nov-97
4132 ----------------------
4133
4134 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4135 like /([ab]*)*/, that is, for classes with more than one character in them.
4136
4137 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4138 as /((?>a*))*/ (a PCRE_EXTRA facility).
4139
4140
4141 Version 1.00 18-Nov-97
4142 ----------------------
4143
4144 1. Added compile-time macros to support systems such as SunOS4 which don't have
4145 memmove() or strerror() but have other things that can be used instead.
4146
4147 2. Arranged that "make clean" removes the executables.
4148
4149
4150 Version 0.99 27-Oct-97
4151 ----------------------
4152
4153 1. Fixed bug in code for optimizing classes with only one character. It was
4154 initializing a 32-byte map regardless, which could cause it to run off the end
4155 of the memory it had got.
4156
4157 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4158
4159
4160 Version 0.98 22-Oct-97
4161 ----------------------
4162
4163 1. Fixed bug in code for handling temporary memory usage when there are more
4164 back references than supplied space in the ovector. This could cause segfaults.
4165
4166
4167 Version 0.97 21-Oct-97
4168 ----------------------
4169
4170 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4171
4172 2. Optimized negated single characters not to use a bit map.
4173
4174 3. Brought error texts together as macro definitions; clarified some of them;
4175 fixed one that was wrong - it said "range out of order" when it meant "invalid
4176 escape sequence".
4177
4178 4. Changed some char * arguments to const char *.
4179
4180 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4181
4182 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4183 pcretest.
4184
4185
4186 Version 0.96 16-Oct-97
4187 ----------------------
4188
4189 1. Added a simple "pgrep" utility to the distribution.
4190
4191 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4192 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4193 where "ddd" means "one or more decimal digits".
4194
4195 3. Fixed serious bug. If a pattern had a back reference, but the call to
4196 pcre_exec() didn't supply a large enough ovector to record the related
4197 identifying subpattern, the match always failed. PCRE now remembers the number
4198 of the largest back reference, and gets some temporary memory in which to save
4199 the offsets during matching if necessary, in order to ensure that
4200 backreferences always work.
4201
4202 4. Increased the compatibility with Perl in a number of ways:
4203
4204 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4205 to request this handling. The option can be set at compile or exec time.
4206
4207 (b) $ matches before a terminating newline by default; an option
4208 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4209 mode). The option can be set at compile or exec time.
4210
4211 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4212 the same as Perl's. If the decimal number it represents is less than 10
4213 or there aren't that many previous left capturing parentheses, an octal
4214 escape is read. Inside a character class, it's always an octal escape,
4215 even if it is a single digit.
4216
4217 (d) An escaped but undefined alphabetic character is taken as a literal,
4218 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4219 escapes.
4220
4221 (e) {0} is now permitted. (The previous item is removed from the compiled
4222 pattern).
4223
4224 5. Changed all the names of code files so that the basic parts are no longer
4225 than 10 characters, and abolished the teeny "globals.c" file.
4226
4227 6. Changed the handling of character classes; they are now done with a 32-byte
4228 bit map always.
4229
4230 7. Added the -d and /D options to pcretest to make it possible to look at the
4231 internals of compilation without having to recompile pcre.
4232
4233
4234 Version 0.95 23-Sep-97
4235 ----------------------
4236
4237 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4238 \x20 at the start of a run of normal characters. These were being treated as
4239 real characters, instead of the source characters being re-checked.
4240
4241
4242 Version 0.94 18-Sep-97
4243 ----------------------
4244
4245 1. The functions are now thread-safe, with the caveat that the global variables
4246 containing pointers to malloc() and free() or alternative functions are the
4247 same for all threads.
4248
4249 2. Get pcre_study() to generate a bitmap of initial characters for non-
4250 anchored patterns when this is possible, and use it if passed to pcre_exec().
4251
4252
4253 Version 0.93 15-Sep-97
4254 ----------------------
4255
4256 1. /(b)|(:+)/ was computing an incorrect first character.
4257
4258 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4259 but not actually doing anything yet.
4260
4261 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4262 as Perl does (e.g. [-az] or [az-]).
4263
4264 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4265 all possible positions.
4266
4267 5. Split up into different modules to avoid including unneeded functions in a
4268 compiled binary. However, compile and exec are still in one module. The "study"
4269 function is split off.
4270
4271 6. The character tables are now in a separate module whose source is generated
4272 by an auxiliary program - but can then be edited by hand if required. There are
4273 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4274 toupper() in the code.
4275
4276 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4277 make them global. Abolish the function for setting them, as the caller can now
4278 set them directly.
4279
4280
4281 Version 0.92 11-Sep-97
4282 ----------------------
4283
4284 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4285 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4286
4287 2. Caseless matching was not working in character classes if the characters in
4288 the pattern were in upper case.
4289
4290 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4291
4292 4. Make PCRE_ANCHORED public and accept as a compile option.
4293
4294 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4295 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4296 pass them.
4297
4298 6. Give an error if bad option bits passed at compile or run time.
4299
4300 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4301 pcretest to cause it to pass that flag.
4302
4303 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4304 options, and the first character, if set.
4305
4306 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4307
4308
4309 Version 0.91 10-Sep-97
4310 ----------------------
4311
4312 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4313 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4314
4315 2. PCRE was looping on encountering an indefinitely repeated back reference to
4316 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4317 Perl does - treats the match as successful.
4318
4319 ****

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12