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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 640 - (show annotations) (download)
Mon Jul 25 10:50:28 2011 UTC (3 years, 1 month ago) by ph10
File size: 208806 byte(s)
Fix three compile-time bugs (Bugzilla #1123).

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

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