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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 657 - (show annotations) (download)
Mon Aug 15 17:39:09 2011 UTC (2 years, 8 months ago) by ph10
File size: 211307 byte(s)
Fix pcre_study() bug with \b at start of branch.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.13 02-Aug-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, now 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 now
111 *disables* studying, thereby overriding the use of -s on the command line
112 (see 14 above). This is necessary for one or two tests to keep the output
113 identical in both cases.
114
115 21. When (*ACCEPT) was used in an assertion that matched an empty string and
116 PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
117
118 22. When an atomic group that contained a capturing parenthesis was
119 successfully matched, but the branch in which it appeared failed, the
120 capturing was not being forgotten if a higher numbered group was later
121 captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
122 group 1 to "a", when in fact it should be unset. This applied to multi-
123 branched capturing and non-capturing groups, repeated or not, and also to
124 positive assertions (capturing in negative assertions does not happen
125 in PCRE) and also to nested atomic groups.
126
127 23. Add the ++ qualifier feature to pcretest, to show the remainder of the
128 subject after a captured substring, to make it easier to tell which of a
129 number of identical substrings has been captured.
130
131 24. The way atomic groups are processed by pcre_exec() has been changed so that
132 if they are repeated, backtracking one repetition now resets captured
133 values correctly. For example, if ((?>(a+)b)+aabab) is matched against
134 "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
135 "aaa". Previously, it would have been "a". As part of this code
136 refactoring, the way recursive calls are handled has also been changed.
137
138 25. If an assertion condition captured any substrings, they were not passed
139 back unless some other capturing happened later. For example, if
140 (?(?=(a))a) was matched against "a", no capturing was returned.
141
142 26. When studying a pattern that contained subroutine calls or assertions,
143 the code for finding the minimum length of a possible match was handling
144 direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
145 group 1 called group 2 while simultaneously a separate group 2 called group
146 1). A stack overflow occurred in this case. I have fixed this by limiting
147 the recursion depth to 10.
148
149 27. Updated RunTest.bat in the distribution to the version supplied by Tom
150 Fortmann. This supports explicit test numbers on the command line, and has
151 argument validation and error reporting.
152
153 28. An instance of \X with an unlimited repeat could fail if at any point the
154 first character it looked at was a mark character.
155
156 29. Some minor code refactoring concerning Unicode properties and scripts
157 should reduce the stack requirement of match() slightly.
158
159 30. Added the '=' option to pcretest to check the setting of unused capturing
160 slots at the end of the pattern, which are documented as being -1, but are
161 not included in the return count.
162
163 31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
164 compiled something random. Now it gives a compile-time error (as does
165 Perl).
166
167 32. A *MARK encountered during the processing of a positive assertion is now
168 recorded and passed back (compatible with Perl).
169
170 33. If --only-matching or --colour was set on a pcregrep call whose pattern
171 had alternative anchored branches, the search for a second match in a line
172 was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
173 matched the line "0102" twice. The same bug affected patterns that started
174 with a backwards assertion. For example /\b01|\b02/ also matched "0102"
175 twice.
176
177 34. Previously, PCRE did not allow quantification of assertions. However, Perl
178 does, and because of capturing effects, quantifying parenthesized
179 assertions may at times be useful. Quantifiers are now allowed for
180 parenthesized assertions.
181
182 35. A minor code tidy in pcre_compile() when checking options for \R usage.
183
184 36. \g was being checked for fancy things in a character class, when it should
185 just be a literal "g".
186
187 37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
188 appearance of a nested POSIX class supersedes an apparent external class.
189 For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
190 unescaped square brackets may also appear as part of class names. For
191 example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
192 more like Perl.
193
194 38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
195 was because it thought it was \N{name}, which is not supported).
196
197 39. Add minix to OS list not supporting the -S option in pcretest.
198
199 40. PCRE tries to detect cases of infinite recursion at compile time, but it
200 cannot analyze patterns in sufficient detail to catch mutual recursions
201 such as ((?1))((?2)). There is now a runtime test that gives an error if a
202 subgroup is called recursively as a subpattern for a second time at the
203 same position in the subject string. In previous releases this might have
204 been caught by the recursion limit, or it might have run out of stack.
205
206 41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
207 happen only once. PCRE was, however incorrectly giving a compile time error
208 "recursive call could loop indefinitely" because it cannot analyze the
209 pattern in sufficient detail. The compile time test no longer happens when
210 PCRE is compiling a conditional subpattern, but actual runaway loops are
211 now caught at runtime (see 40 above).
212
213 42. It seems that Perl allows any characters other than a closing parenthesis
214 to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
215 has been changed to be the same.
216
217 43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
218 as not to get warnings when autogen.sh is called. Also changed
219 AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
220
221 44. To help people who use pcregrep to scan files containing exceedingly long
222 lines, the following changes have been made:
223
224 (a) The default value of the buffer size parameter has been increased from
225 8K to 20K. (The actual buffer used is three times this size.)
226
227 (b) The default can be changed by ./configure --with-pcregrep-bufsiz when
228 PCRE is built.
229
230 (c) A --buffer-size=n option has been added to pcregrep, to allow the size
231 to be set at run time.
232
233 (d) Numerical values in pcregrep options can be followed by K or M, for
234 example --buffer-size=50K.
235
236 (e) If a line being scanned overflows pcregrep's buffer, an error is now
237 given and the return code is set to 2.
238
239 45. Add a pointer to the latest mark to the callout data block.
240
241 46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
242 partial match of an empty string instead of no match. This was specific to
243 the use of ".".
244
245 47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
246 complete match instead of a partial match. This bug was dependent on both
247 the PCRE_UTF8 and PCRE_DOTALL options being set.
248
249 48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
250 starting byte set, because \b was not being ignored.
251
252
253 Version 8.12 15-Jan-2011
254 ------------------------
255
256 1. Fixed some typos in the markup of the man pages, and wrote a script that
257 checks for such things as part of the documentation building process.
258
259 2. On a big-endian 64-bit system, pcregrep did not correctly process the
260 --match-limit and --recursion-limit options (added for 8.11). In
261 particular, this made one of the standard tests fail. (The integer value
262 went into the wrong half of a long int.)
263
264 3. If the --colour option was given to pcregrep with -v (invert match), it
265 did strange things, either producing crazy output, or crashing. It should,
266 of course, ignore a request for colour when reporting lines that do not
267 match.
268
269 4. Another pcregrep bug caused similar problems if --colour was specified with
270 -M (multiline) and the pattern match finished with a line ending.
271
272 5. In pcregrep, when a pattern that ended with a literal newline sequence was
273 matched in multiline mode, the following line was shown as part of the
274 match. This seems wrong, so I have changed it.
275
276 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
277 the check for further matches in the same line (so they could be coloured)
278 to overrun the end of the current line. If another match was found, it was
279 incorrectly shown (and then shown again when found in the next line).
280
281 7. If pcregrep was compiled under Windows, there was a reference to the
282 function pcregrep_exit() before it was defined. I am assuming this was
283 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
284 reported by a user. I've moved the definition above the reference.
285
286
287 Version 8.11 10-Dec-2010
288 ------------------------
289
290 1. (*THEN) was not working properly if there were untried alternatives prior
291 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
292 backtracked to try for "b" instead of moving to the next alternative branch
293 at the same level (in this case, to look for "c"). The Perl documentation
294 is clear that when (*THEN) is backtracked onto, it goes to the "next
295 alternative in the innermost enclosing group".
296
297 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
298 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
299 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
300 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
301 (*THEN).
302
303 3. If \s appeared in a character class, it removed the VT character from
304 the class, even if it had been included by some previous item, for example
305 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
306 of \s, but is part of the POSIX "space" class.)
307
308 4. A partial match never returns an empty string (because you can always
309 match an empty string at the end of the subject); however the checking for
310 an empty string was starting at the "start of match" point. This has been
311 changed to the "earliest inspected character" point, because the returned
312 data for a partial match starts at this character. This means that, for
313 example, /(?<=abc)def/ gives a partial match for the subject "abc"
314 (previously it gave "no match").
315
316 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
317 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
318 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
319 has an implication that the given string is incomplete (because a partial
320 match is preferred over a full match). For this reason, these items now
321 give a partial match in this situation. [Aside: previously, the one case
322 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
323 match rather than a full match, which was wrong by the old rules, but is
324 now correct.]
325
326 6. There was a bug in the handling of #-introduced comments, recognized when
327 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
328 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
329 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
330 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
331 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
332 places in pcre_compile().
333
334 7. Related to (6) above, when pcre_compile() was skipping #-introduced
335 comments when looking ahead for named forward references to subpatterns,
336 the only newline sequence it recognized was NL. It now handles newlines
337 according to the set newline convention.
338
339 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
340 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
341 cater for a lack of strerror(). These oversights have been fixed.
342
343 9. Added --match-limit and --recursion-limit to pcregrep.
344
345 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
346
347 11. When the -o option was used, pcregrep was setting a return code of 1, even
348 when matches were found, and --line-buffered was not being honoured.
349
350 12. Added an optional parentheses number to the -o and --only-matching options
351 of pcregrep.
352
353 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
354 can match an empty string. The code to do it in pcretest and pcredemo
355 needed fixing:
356
357 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
358 only one byte after an empty string match just before CRLF (this case
359 just got forgotten; "any" and "anycrlf" were OK).
360
361 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
362 mode when an empty string match preceded an ASCII character followed by
363 a non-ASCII character. (The code for advancing by one character rather
364 than one byte was nonsense.)
365
366 (c) The pcredemo.c sample program did not have any code at all to handle
367 the cases when CRLF is a valid newline sequence.
368
369 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
370 as a starting offset was within the subject string. There is now a new
371 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
372 negative or greater than the length of the string. In order to test this,
373 pcretest is extended to allow the setting of negative starting offsets.
374
375 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
376 starting offset points to the beginning of a UTF-8 character was
377 unnecessarily clumsy. I tidied it up.
378
379 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
380 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
381
382 17. Nobody had reported that the --include_dir option, which was added in
383 release 7.7 should have been called --include-dir (hyphen, not underscore)
384 for compatibility with GNU grep. I have changed it to --include-dir, but
385 left --include_dir as an undocumented synonym, and the same for
386 --exclude-dir, though that is not available in GNU grep, at least as of
387 release 2.5.4.
388
389 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
390 characters from a string of bytes) have been redefined so as not to use
391 loops, in order to improve performance in some environments. At the same
392 time, I abstracted some of the common code into auxiliary macros to save
393 repetition (this should not affect the compiled code).
394
395 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
396 compile-time error is now given if \c is not followed by an ASCII
397 character, that is, a byte less than 128. (In EBCDIC mode, the code is
398 different, and any byte value is allowed.)
399
400 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
401 START_OPTIMIZE option, which is now allowed at compile time - but just
402 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
403 to pcregrep and other applications that have no direct access to PCRE
404 options. The new /Y option in pcretest sets this option when calling
405 pcre_compile().
406
407 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
408 back references. Groups containing recursive back references were forced to
409 be atomic by that change, but in the case of named groups, the amount of
410 memory required was incorrectly computed, leading to "Failed: internal
411 error: code overflow". This has been fixed.
412
413 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
414 pcretest.c, to avoid build problems in some Borland environments.
415
416
417 Version 8.10 25-Jun-2010
418 ------------------------
419
420 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
421 THEN.
422
423 2. (*ACCEPT) was not working when inside an atomic group.
424
425 3. Inside a character class, \B is treated as a literal by default, but
426 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
427 causes the error). The code is unchanged, but I tidied the documentation.
428
429 4. Inside a character class, PCRE always treated \R and \X as literals,
430 whereas Perl faults them if its -w option is set. I have changed PCRE so
431 that it faults them when PCRE_EXTRA is set.
432
433 5. Added support for \N, which always matches any character other than
434 newline. (It is the same as "." when PCRE_DOTALL is not set.)
435
436 6. When compiling pcregrep with newer versions of gcc which may have
437 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
438 declared with attribute warn_unused_result" were given. Just casting the
439 result to (void) does not stop the warnings; a more elaborate fudge is
440 needed. I've used a macro to implement this.
441
442 7. Minor change to pcretest.c to avoid a compiler warning.
443
444 8. Added four artifical Unicode properties to help with an option to make
445 \s etc use properties (see next item). The new properties are: Xan
446 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
447
448 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
449 use Unicode properties. (*UCP) at the start of a pattern can be used to set
450 this option. Modified pcretest to add /W to test this facility. Added
451 REG_UCP to make it available via the POSIX interface.
452
453 10. Added --line-buffered to pcregrep.
454
455 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
456 studied, and the match started with a letter with a code point greater than
457 127 whose first byte was different to the first byte of the other case of
458 the letter, the other case of this starting letter was not recognized
459 (#976).
460
461 12. If a pattern that was studied started with a repeated Unicode property
462 test, for example, \p{Nd}+, there was the theoretical possibility of
463 setting up an incorrect bitmap of starting bytes, but fortunately it could
464 not have actually happened in practice until change 8 above was made (it
465 added property types that matched character-matching opcodes).
466
467 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
468 possible starting bytes for non-anchored patterns.
469
470 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
471 \R, and also a number of cases that involve Unicode properties, both
472 explicit and implicit when PCRE_UCP is set.
473
474 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
475 input, it could crash or give wrong results if characters with values
476 greater than 0xc0 were present in the subject string. (Detail: it assumed
477 UTF-8 input when processing these items.)
478
479 16. Added a lot of (int) casts to avoid compiler warnings in systems where
480 size_t is 64-bit (#991).
481
482 17. Added a check for running out of memory when PCRE is compiled with
483 --disable-stack-for-recursion (#990).
484
485 18. If the last data line in a file for pcretest does not have a newline on
486 the end, a newline was missing in the output.
487
488 19. The default pcre_chartables.c file recognizes only ASCII characters (values
489 less than 128) in its various bitmaps. However, there is a facility for
490 generating tables according to the current locale when PCRE is compiled. It
491 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
492 characters, are recognized by isspace() and therefore were getting set in
493 these tables, and indeed these tables seem to approximate to ISO 8859. This
494 caused a problem in UTF-8 mode when pcre_study() was used to create a list
495 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
496 which of course cannot start UTF-8 characters. I have changed the code so
497 that only real ASCII characters (less than 128) and the correct starting
498 bytes for UTF-8 encodings are set for characters greater than 127 when in
499 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
500 altogether.)
501
502 20. Added the /T option to pcretest so as to be able to run tests with non-
503 standard character tables, thus making it possible to include the tests
504 used for 19 above in the standard set of tests.
505
506 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
507 reference to a subpattern the other side of a comment that contains an
508 opening parenthesis caused either an internal compiling error, or a
509 reference to the wrong subpattern.
510
511
512 Version 8.02 19-Mar-2010
513 ------------------------
514
515 1. The Unicode data tables have been updated to Unicode 5.2.0.
516
517 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
518 configured.
519
520 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
521 original author of that file, following a query about its status.
522
523 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
524 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
525
526 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
527 quantifier applied to a forward-referencing subroutine call, could compile
528 incorrect code or give the error "internal error: previously-checked
529 referenced subpattern not found".
530
531 6. Both MS Visual Studio and Symbian OS have problems with initializing
532 variables to point to external functions. For these systems, therefore,
533 pcre_malloc etc. are now initialized to local functions that call the
534 relevant global functions.
535
536 7. There were two entries missing in the vectors called coptable and poptable
537 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
538 I've fixed the data, and added a kludgy way of testing at compile time that
539 the lengths are correct (equal to the number of opcodes).
540
541 8. Following on from 7, I added a similar kludge to check the length of the
542 eint vector in pcreposix.c.
543
544 9. Error texts for pcre_compile() are held as one long string to avoid too
545 much relocation at load time. To find a text, the string is searched,
546 counting zeros. There was no check for running off the end of the string,
547 which could happen if a new error number was added without updating the
548 string.
549
550 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
551
552 11. \K was not working if it appeared in an atomic group or in a group that
553 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
554 \K is "not well defined" if used in an assertion. PCRE now accepts it if
555 the assertion is positive, but not if it is negative.
556
557 12. Change 11 fortuitously reduced the size of the stack frame used in the
558 "match()" function of pcre_exec.c by one pointer. Forthcoming
559 implementation of support for (*MARK) will need an extra pointer on the
560 stack; I have reserved it now, so that the stack frame size does not
561 decrease.
562
563 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
564 item in branch that calls a recursion is a subroutine call - as in the
565 second branch in the above example - was incorrectly given the compile-
566 time error "recursive call could loop indefinitely" because pcre_compile()
567 was not correctly checking the subroutine for matching a non-empty string.
568
569 14. The checks for overrunning compiling workspace could trigger after an
570 overrun had occurred. This is a "should never occur" error, but it can be
571 triggered by pathological patterns such as hundreds of nested parentheses.
572 The checks now trigger 100 bytes before the end of the workspace.
573
574 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
575
576
577 Version 8.01 19-Jan-2010
578 ------------------------
579
580 1. If a pattern contained a conditional subpattern with only one branch (in
581 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
582 computed the wrong minimum data length (which is of course zero for such
583 subpatterns). This could cause incorrect "no match" results.
584
585 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
586 the pattern is reset in the first branch, pcre_compile() failed with
587 "internal error: code overflow at offset...". This happened only when
588 the reset was to the original external option setting. (An optimization
589 abstracts leading options settings into an external setting, which was the
590 cause of this.)
591
592 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
593 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
594 assertion pattern did not match (meaning that the assertion was true), it
595 was incorrectly treated as false if the SKIP had been reached during the
596 matching. This also applied to assertions used as conditions.
597
598 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
599 assertion subpattern, including such a pattern used as a condition,
600 unpredictable results occurred, instead of the error return
601 PCRE_ERROR_DFA_UITEM.
602
603 5. The C++ GlobalReplace function was not working like Perl for the special
604 situation when an empty string is matched. It now does the fancy magic
605 stuff that is necessary.
606
607 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
608 removed. (These were left over from very, very early versions of PCRE.)
609
610 7. Some cosmetic changes to the code to make life easier when compiling it
611 as part of something else:
612
613 (a) Change DEBUG to PCRE_DEBUG.
614
615 (b) In pcre_compile(), rename the member of the "branch_chain" structure
616 called "current" as "current_branch", to prevent a collision with the
617 Linux macro when compiled as a kernel module.
618
619 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
620 prevent a collision with the Linux macro when compiled as a kernel
621 module.
622
623 8. In pcre_compile() there are some checks for integer overflows that used to
624 cast potentially large values to (double). This has been changed to that
625 when building, a check for int64_t is made, and if it is found, it is used
626 instead, thus avoiding the use of floating point arithmetic. (There is no
627 other use of FP in PCRE.) If int64_t is not found, the fallback is to
628 double.
629
630 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
631 2005 (difference between two addresses compared to an unsigned value).
632
633 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
634 custom one, because of the following reported problem in Windows:
635
636 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
637 under Win32.
638 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
639 therefore missing the function definition.
640 - The compiler thus generates a "C" signature for the test function.
641 - The linker fails to find the "C" function.
642 - PCRE fails to configure if asked to do so against libbz2.
643
644 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
645 messages were output:
646
647 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
648 rerunning libtoolize, to keep the correct libtool macros in-tree.
649 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
650
651 I have done both of these things.
652
653 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
654 most of the time, it *can* run out if it is given a pattern that contains a
655 runaway infinite recursion. I updated the discussion in the pcrestack man
656 page.
657
658 13. Now that we have gone to the x.xx style of version numbers, the minor
659 version may start with zero. Using 08 or 09 is a bad idea because users
660 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
661 interpreted as invalid octal numbers. I've updated the previous comment in
662 configure.ac, and also added a check that gives an error if 08 or 09 are
663 used.
664
665 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
666 causing partial matching to fail when the end of the subject matched \W
667 in a UTF-8 pattern where \W was quantified with a minimum of 3.
668
669 15. There were some discrepancies between the declarations in pcre_internal.h
670 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
671 their definitions. The declarations used "const uschar *" and the
672 definitions used USPTR. Even though USPTR is normally defined as "const
673 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
674 reported that: "This difference in casting confuses some C++ compilers, for
675 example, SunCC recognizes above declarations as different functions and
676 generates broken code for hbpcre." I have changed the declarations to use
677 USPTR.
678
679 16. GNU libtool is named differently on some systems. The autogen.sh script now
680 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
681 (FreeBSD).
682
683 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
684 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
685 comment: "Figure out how to create a longlong from a string: strtoll and
686 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
687 instance, but it only takes 2 args instead of 3!"
688
689 18. A subtle bug concerned with back references has been fixed by a change of
690 specification, with a corresponding code fix. A pattern such as
691 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
692 refers, was giving matches when it shouldn't. For example, xa=xaaa would
693 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
694 same bug. Such groups have to be quantified to be useful, or contained
695 inside another quantified group. (If there's no repetition, the reference
696 can never match.) The problem arises because, having left the group and
697 moved on to the rest of the pattern, a later failure that backtracks into
698 the group uses the captured value from the final iteration of the group
699 rather than the correct earlier one. I have fixed this in PCRE by forcing
700 any group that contains a reference to itself to be an atomic group; that
701 is, there cannot be any backtracking into it once it has completed. This is
702 similar to recursive and subroutine calls.
703
704
705 Version 8.00 19-Oct-09
706 ----------------------
707
708 1. The table for translating pcre_compile() error codes into POSIX error codes
709 was out-of-date, and there was no check on the pcre_compile() error code
710 being within the table. This could lead to an OK return being given in
711 error.
712
713 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
714 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
715 in a Windows environment.
716
717 3. The pcregrep --count option prints the count for each file even when it is
718 zero, as does GNU grep. However, pcregrep was also printing all files when
719 --files-with-matches was added. Now, when both options are given, it prints
720 counts only for those files that have at least one match. (GNU grep just
721 prints the file name in this circumstance, but including the count seems
722 more useful - otherwise, why use --count?) Also ensured that the
723 combination -clh just lists non-zero counts, with no names.
724
725 4. The long form of the pcregrep -F option was incorrectly implemented as
726 --fixed_strings instead of --fixed-strings. This is an incompatible change,
727 but it seems right to fix it, and I didn't think it was worth preserving
728 the old behaviour.
729
730 5. The command line items --regex=pattern and --regexp=pattern were not
731 recognized by pcregrep, which required --regex pattern or --regexp pattern
732 (with a space rather than an '='). The man page documented the '=' forms,
733 which are compatible with GNU grep; these now work.
734
735 6. No libpcreposix.pc file was created for pkg-config; there was just
736 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
737
738 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
739 when UCP support is not needed, by modifying the Python script that
740 generates it from Unicode data files. This should not matter if the module
741 is correctly used as a library, but I received one complaint about 50K of
742 unwanted data. My guess is that the person linked everything into his
743 program rather than using a library. Anyway, it does no harm.
744
745 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
746 was a minimum greater than 1 for a wide character in a possessive
747 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
748 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
749 character. Chaos in the form of incorrect output or a compiling loop could
750 result.
751
752 9. The restrictions on what a pattern can contain when partial matching is
753 requested for pcre_exec() have been removed. All patterns can now be
754 partially matched by this function. In addition, if there are at least two
755 slots in the offset vector, the offset of the earliest inspected character
756 for the match and the offset of the end of the subject are set in them when
757 PCRE_ERROR_PARTIAL is returned.
758
759 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
760 synonymous with PCRE_PARTIAL, for backwards compatibility, and
761 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
762 and may be more useful for multi-segment matching.
763
764 11. Partial matching with pcre_exec() is now more intuitive. A partial match
765 used to be given if ever the end of the subject was reached; now it is
766 given only if matching could not proceed because another character was
767 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
768 string "Z", which now yields "no match" instead of "partial match". In the
769 case of pcre_dfa_exec(), "no match" is given if every matching path for the
770 final character ended with (*FAIL).
771
772 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
773 if the pattern had a "must contain" character that was already found in the
774 earlier partial match, unless partial matching was again requested. For
775 example, with the pattern /dog.(body)?/, the "must contain" character is
776 "g". If the first part-match was for the string "dog", restarting with
777 "sbody" failed. This bug has been fixed.
778
779 13. The string returned by pcre_dfa_exec() after a partial match has been
780 changed so that it starts at the first inspected character rather than the
781 first character of the match. This makes a difference only if the pattern
782 starts with a lookbehind assertion or \b or \B (\K is not supported by
783 pcre_dfa_exec()). It's an incompatible change, but it makes the two
784 matching functions compatible, and I think it's the right thing to do.
785
786 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
787 so that the demonstration program is easily available in environments where
788 PCRE has not been installed from source.
789
790 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
791 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
792 library.
793
794 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
795 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
796 is not the first non-POSIX option to be added. Clearly some people find
797 these options useful.
798
799 17. If a caller to the POSIX matching function regexec() passes a non-zero
800 value for nmatch with a NULL value for pmatch, the value of
801 nmatch is forced to zero.
802
803 18. RunGrepTest did not have a test for the availability of the -u option of
804 the diff command, as RunTest does. It now checks in the same way as
805 RunTest, and also checks for the -b option.
806
807 19. If an odd number of negated classes containing just a single character
808 interposed, within parentheses, between a forward reference to a named
809 subpattern and the definition of the subpattern, compilation crashed with
810 an internal error, complaining that it could not find the referenced
811 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
812 [The bug was that it was starting one character too far in when skipping
813 over the character class, thus treating the ] as data rather than
814 terminating the class. This meant it could skip too much.]
815
816 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
817 /g option in pcretest when the pattern contains \K, which makes it possible
818 to have an empty string match not at the start, even when the pattern is
819 anchored. Updated pcretest and pcredemo to use this option.
820
821 21. If the maximum number of capturing subpatterns in a recursion was greater
822 than the maximum at the outer level, the higher number was returned, but
823 with unset values at the outer level. The correct (outer level) value is
824 now given.
825
826 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
827 PCRE did not set those parentheses (unlike Perl). I have now found a way to
828 make it do so. The string so far is captured, making this feature
829 compatible with Perl.
830
831 23. The tests have been re-organized, adding tests 11 and 12, to make it
832 possible to check the Perl 5.10 features against Perl 5.10.
833
834 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
835 pattern matches a fixed length string. PCRE did not allow this; now it
836 does. Neither allows recursion.
837
838 25. I finally figured out how to implement a request to provide the minimum
839 length of subject string that was needed in order to match a given pattern.
840 (It was back references and recursion that I had previously got hung up
841 on.) This code has now been added to pcre_study(); it finds a lower bound
842 to the length of subject needed. It is not necessarily the greatest lower
843 bound, but using it to avoid searching strings that are too short does give
844 some useful speed-ups. The value is available to calling programs via
845 pcre_fullinfo().
846
847 26. While implementing 25, I discovered to my embarrassment that pcretest had
848 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
849 study optimizations had never been tested with that matching function.
850 Oops. What is worse, even when it was passed study data, there was a bug in
851 pcre_dfa_exec() that meant it never actually used it. Double oops. There
852 were also very few tests of studied patterns with pcre_dfa_exec().
853
854 27. If (?| is used to create subpatterns with duplicate numbers, they are now
855 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
856 on the other side of the coin, they are no longer allowed to have different
857 names, because these cannot be distinguished in PCRE, and this has caused
858 confusion. (This is a difference from Perl.)
859
860 28. When duplicate subpattern names are present (necessarily with different
861 numbers, as required by 27 above), and a test is made by name in a
862 conditional pattern, either for a subpattern having been matched, or for
863 recursion in such a pattern, all the associated numbered subpatterns are
864 tested, and the overall condition is true if the condition is true for any
865 one of them. This is the way Perl works, and is also more like the way
866 testing by number works.
867
868
869 Version 7.9 11-Apr-09
870 ---------------------
871
872 1. When building with support for bzlib/zlib (pcregrep) and/or readline
873 (pcretest), all targets were linked against these libraries. This included
874 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
875 libraries. This caused unwanted dependencies to be created. This problem
876 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
877 pcretest is linked with readline.
878
879 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
880 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
881 moved outside it again, because FALSE and TRUE are already defined in AIX,
882 but BOOL is not.
883
884 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
885 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
886
887 4. The pcregrep documentation said spaces were inserted as well as colons (or
888 hyphens) following file names and line numbers when outputting matching
889 lines. This is not true; no spaces are inserted. I have also clarified the
890 wording for the --colour (or --color) option.
891
892 5. In pcregrep, when --colour was used with -o, the list of matching strings
893 was not coloured; this is different to GNU grep, so I have changed it to be
894 the same.
895
896 6. When --colo(u)r was used in pcregrep, only the first matching substring in
897 each matching line was coloured. Now it goes on to look for further matches
898 of any of the test patterns, which is the same behaviour as GNU grep.
899
900 7. A pattern that could match an empty string could cause pcregrep to loop; it
901 doesn't make sense to accept an empty string match in pcregrep, so I have
902 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
903 seems to be how GNU grep behaves.
904
905 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
906 start or after a newline", because the conditional assertion was not being
907 correctly handled. The rule now is that both the assertion and what follows
908 in the first alternative must satisfy the test.
909
910 9. If auto-callout was enabled in a pattern with a conditional group whose
911 condition was an assertion, PCRE could crash during matching, both with
912 pcre_exec() and pcre_dfa_exec().
913
914 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
915 used for matching.
916
917 11. Unicode property support in character classes was not working for
918 characters (bytes) greater than 127 when not in UTF-8 mode.
919
920 12. Added the -M command line option to pcretest.
921
922 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
923
924 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
925
926 16. Added comments and documentation about mis-use of no_arg in the C++
927 wrapper.
928
929 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
930 from Martin Jerabek that uses macro names for all relevant character and
931 string constants.
932
933 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
934 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
935 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
936 these, but not everybody uses configure.
937
938 19. A conditional group that had only one branch was not being correctly
939 recognized as an item that could match an empty string. This meant that an
940 enclosing group might also not be so recognized, causing infinite looping
941 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
942 with the subject "ab", where knowledge that the repeated group can match
943 nothing is needed in order to break the loop.
944
945 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
946 exec(), but without supplying a callout function, matching went wrong.
947
948 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
949 leak if the size of the offset vector was greater than 30. When the vector
950 is smaller, the saved offsets during recursion go onto a local stack
951 vector, but for larger vectors malloc() is used. It was failing to free
952 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
953 error, in fact).
954
955 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
956 heapframe that is used only when UTF-8 support is enabled. This caused no
957 problem, but was untidy.
958
959 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
960 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
961 included within another project.
962
963 24. Steven Van Ingelgem's patches to add more options to the CMake support,
964 slightly modified by me:
965
966 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
967 not building pcregrep.
968
969 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
970 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
971
972 25. Forward references, both numeric and by name, in patterns that made use of
973 duplicate group numbers, could behave incorrectly or give incorrect errors,
974 because when scanning forward to find the reference group, PCRE was not
975 taking into account the duplicate group numbers. A pattern such as
976 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
977
978 26. Changed a few more instances of "const unsigned char *" to USPTR, making
979 the feature of a custom pointer more persuasive (as requested by a user).
980
981 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
982 pcretest.c, inside #ifndefs, because it seems they are sometimes already
983 pre-defined.
984
985 28. Added support for (*UTF8) at the start of a pattern.
986
987 29. Arrange for flags added by the "release type" setting in CMake to be shown
988 in the configuration summary.
989
990
991 Version 7.8 05-Sep-08
992 ---------------------
993
994 1. Replaced UCP searching code with optimized version as implemented for Ad
995 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
996 stage table and inline lookup instead of a function, giving speed ups of 2
997 to 5 times on some simple patterns that I tested. Permission was given to
998 distribute the MultiStage2.py script that generates the tables (it's not in
999 the tarball, but is in the Subversion repository).
1000
1001 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1002 scripts.
1003
1004 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1005 a group with a zero qualifier. The result of the study could be incorrect,
1006 or the function might crash, depending on the pattern.
1007
1008 4. Caseless matching was not working for non-ASCII characters in back
1009 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1010 It now works when Unicode Property Support is available.
1011
1012 5. In pcretest, an escape such as \x{de} in the data was always generating
1013 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1014 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1015 truncation.
1016
1017 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1018
1019 7. Added two (int) casts to pcregrep when printing the difference of two
1020 pointers, in case they are 64-bit values.
1021
1022 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
1023 test 2 if it fails.
1024
1025 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1026 and a #define of that name to empty if it is not externally set. This is to
1027 allow users of MSVC to set it if necessary.
1028
1029 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1030 the convenience functions in the pcre_get.c source file.
1031
1032 11. An option change at the start of a pattern that had top-level alternatives
1033 could cause overwriting and/or a crash. This command provoked a crash in
1034 some environments:
1035
1036 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1037
1038 This potential security problem was recorded as CVE-2008-2371.
1039
1040 12. For a pattern where the match had to start at the beginning or immediately
1041 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1042 pcre_dfa_exec() could read past the end of the passed subject if there was
1043 no match. To help with detecting such bugs (e.g. with valgrind), I modified
1044 pcretest so that it places the subject at the end of its malloc-ed buffer.
1045
1046 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1047 exec() might read past the end of the data buffer in UTF-8 mode.
1048
1049 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1050 the data contained the byte 0x85 as part of a UTF-8 character within its
1051 first line. This applied both to normal and DFA matching.
1052
1053 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1054 /^[^d]*?$/8 failed to match "abc".
1055
1056 16. Added a missing copyright notice to pcrecpp_internal.h.
1057
1058 17. Make it more clear in the documentation that values returned from
1059 pcre_exec() in ovector are byte offsets, not character counts.
1060
1061 18. Tidied a few places to stop certain compilers from issuing warnings.
1062
1063 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1064 supplied by Stefan Weber. I made a further small update for 7.8 because
1065 there is a change of source arrangements: the pcre_searchfuncs.c module is
1066 replaced by pcre_ucd.c.
1067
1068
1069 Version 7.7 07-May-08
1070 ---------------------
1071
1072 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
1073 a string to a long long, pretend we don't even have a long long." This is
1074 done by checking for the strtoq, strtoll, and _strtoi64 functions.
1075
1076 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1077 pre-7.6 versions, which defined a global no_arg variable instead of putting
1078 it in the RE class. (See also #8 below.)
1079
1080 3. Remove a line of dead code, identified by coverity and reported by Nuno
1081 Lopes.
1082
1083 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
1084
1085 (1) The include/exclude patterns were being applied to the whole pathnames
1086 of files, instead of just to the final components.
1087
1088 (2) If there was more than one level of directory, the subdirectories were
1089 skipped unless they satisfied the include/exclude conditions. This is
1090 inconsistent with GNU grep (and could even be seen as contrary to the
1091 pcregrep specification - which I improved to make it absolutely clear).
1092 The action now is always to scan all levels of directory, and just
1093 apply the include/exclude patterns to regular files.
1094
1095 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1096 --exclude_dir in the tests to avoid scanning .svn directories.
1097
1098 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1099 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1100 doesn't support NULs in patterns.
1101
1102 7. Added some missing "const"s to declarations of static tables in
1103 pcre_compile.c and pcre_dfa_exec.c.
1104
1105 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1106 caused by fix #2 above. (Subsequently also a second patch to fix the
1107 first patch. And a third patch - this was a messy problem.)
1108
1109 9. Applied Craig's patch to remove the use of push_back().
1110
1111 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1112 matching function regexec().
1113
1114 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1115 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1116 references. PCRE supports relative numbers with this syntax (I don't think
1117 Oniguruma does).
1118
1119 12. Previously, a group with a zero repeat such as (...){0} was completely
1120 omitted from the compiled regex. However, this means that if the group
1121 was called as a subroutine from elsewhere in the pattern, things went wrong
1122 (an internal error was given). Such groups are now left in the compiled
1123 pattern, with a new opcode that causes them to be skipped at execution
1124 time.
1125
1126 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1127 to the way PCRE behaves:
1128
1129 (a) A lone ] character is dis-allowed (Perl treats it as data).
1130
1131 (b) A back reference to an unmatched subpattern matches an empty string
1132 (Perl fails the current match path).
1133
1134 (c) A data ] in a character class must be notated as \] because if the
1135 first data character in a class is ], it defines an empty class. (In
1136 Perl it is not possible to have an empty class.) The empty class []
1137 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1138 The negative empty class [^] matches any one character, independently
1139 of the DOTALL setting.
1140
1141 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1142 non-existent subpattern following a character class starting with ']' and
1143 containing () gave an internal compiling error instead of "reference to
1144 non-existent subpattern". Fortunately, when the pattern did exist, the
1145 compiled code was correct. (When scanning forwards to check for the
1146 existencd of the subpattern, it was treating the data ']' as terminating
1147 the class, so got the count wrong. When actually compiling, the reference
1148 was subsequently set up correctly.)
1149
1150 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1151 it was being rejected as not supported by pcre_dfa_exec(), even though
1152 other assertions are supported. I have made pcre_dfa_exec() support
1153 (*FAIL).
1154
1155 16. The implementation of 13c above involved the invention of a new opcode,
1156 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1157 cannot be changed at match time, I realized I could make a small
1158 improvement to matching performance by compiling OP_ALLANY instead of
1159 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1160 on the OP_ANY path.
1161
1162 17. Compiling pcretest on Windows with readline support failed without the
1163 following two fixes: (1) Make the unistd.h include conditional on
1164 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1165
1166 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1167 ncurses library to be included for pcretest when ReadLine support is
1168 requested, but also to allow for it to be overridden. This patch came from
1169 Daniel Bergström.
1170
1171 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1172 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1173 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1174 spotting this.
1175
1176
1177 Version 7.6 28-Jan-08
1178 ---------------------
1179
1180 1. A character class containing a very large number of characters with
1181 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1182 overflow.
1183
1184 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1185 HAVE_LONG_LONG is not defined.
1186
1187 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1188 bring them up to date and include new features. This patch includes:
1189
1190 - Fixed PH's badly added libz and libbz2 support.
1191 - Fixed a problem with static linking.
1192 - Added pcredemo. [But later removed - see 7 below.]
1193 - Fixed dftables problem and added an option.
1194 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1195 HAVE_LONG_LONG.
1196 - Added readline support for pcretest.
1197 - Added an listing of the option settings after cmake has run.
1198
1199 4. A user submitted a patch to Makefile that makes it easy to create
1200 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1201 Makefile.am that cause it to include this special target, without
1202 affecting anything else. Note that the same mingw target plus all
1203 the other distribution libraries and programs are now supported
1204 when configuring with CMake (see 6 below) instead of with
1205 Configure/Make.
1206
1207 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1208 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1209 exported in the Windows port". It has not yet been confirmed that the patch
1210 solves the problem, but it does no harm.
1211
1212 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1213 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1214 with CMake, and also correct the comment about stack recursion.
1215
1216 7. Remove the automatic building of pcredemo from the ./configure system and
1217 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1218 of a program that users should build themselves after PCRE is installed, so
1219 building it automatically is not really right. What is more, it gave
1220 trouble in some build environments.
1221
1222 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1223
1224
1225 Version 7.5 10-Jan-08
1226 ---------------------
1227
1228 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1229 values in parens when parsing an RE using the C++ wrapper."
1230
1231 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1232 Characters greater than 255 were excluded from the class instead of being
1233 included.
1234
1235 3. The same bug as (2) above applied to negated POSIX classes such as
1236 [:^space:].
1237
1238 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1239 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1240 I have changed it.
1241
1242 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1243 first named subpattern) and a construct such as (?&a) would reference the
1244 first named subpattern whose name started with "a" (in other words, the
1245 length check was missing). Both these problems are fixed. "Subpattern name
1246 expected" is now given for (?&) (a zero-length name), and this patch also
1247 makes it give the same error for \k'' (previously it complained that that
1248 was a reference to a non-existent subpattern).
1249
1250 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1251 this is right because (?- can be followed by option settings as well as by
1252 digits. I have, however, made the messages clearer.
1253
1254 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1255 than the number used in the conditional) now cause a compile-time error.
1256 This is actually not compatible with Perl, which accepts such patterns, but
1257 treats the conditional as always being FALSE (as PCRE used to), but it
1258 seems to me that giving a diagnostic is better.
1259
1260 8. Change "alphameric" to the more common word "alphanumeric" in comments
1261 and messages.
1262
1263 9. Fix two occurrences of "backslash" in comments that should have been
1264 "backspace".
1265
1266 10. Remove two redundant lines of code that can never be obeyed (their function
1267 was moved elsewhere).
1268
1269 11. The program that makes PCRE's Unicode character property table had a bug
1270 which caused it to generate incorrect table entries for sequences of
1271 characters that have the same character type, but are in different scripts.
1272 It amalgamated them into a single range, with the script of the first of
1273 them. In other words, some characters were in the wrong script. There were
1274 thirteen such cases, affecting characters in the following ranges:
1275
1276 U+002b0 - U+002c1
1277 U+0060c - U+0060d
1278 U+0061e - U+00612
1279 U+0064b - U+0065e
1280 U+0074d - U+0076d
1281 U+01800 - U+01805
1282 U+01d00 - U+01d77
1283 U+01d9b - U+01dbf
1284 U+0200b - U+0200f
1285 U+030fc - U+030fe
1286 U+03260 - U+0327f
1287 U+0fb46 - U+0fbb1
1288 U+10450 - U+1049d
1289
1290 12. The -o option (show only the matching part of a line) for pcregrep was not
1291 compatible with GNU grep in that, if there was more than one match in a
1292 line, it showed only the first of them. It now behaves in the same way as
1293 GNU grep.
1294
1295 13. If the -o and -v options were combined for pcregrep, it printed a blank
1296 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1297 does the same. The return code can be used to tell if there were any
1298 non-matching lines.
1299
1300 14. Added --file-offsets and --line-offsets to pcregrep.
1301
1302 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1303 infinitely looping recursion. The bug was that positive lookaheads were not
1304 being skipped when checking for a possible empty match (negative lookaheads
1305 and both kinds of lookbehind were skipped).
1306
1307 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1308 inclusion of <windows.h> to before rather than after the definition of
1309 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1310
1311 17. Specifying a possessive quantifier with a specific limit for a Unicode
1312 character property caused pcre_compile() to compile bad code, which led at
1313 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1314 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1315 caused the error; without that there was no problem.
1316
1317 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1318
1319 19. Added --enable-pcretest-libreadline.
1320
1321 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1322 RE::GlobalReplace(). As a result, the number of replacements returned was
1323 double what it should be. I removed one of the increments, but Craig sent a
1324 later patch that removed the other one (the right fix) and added unit tests
1325 that check the return values (which was not done before).
1326
1327 21. Several CMake things:
1328
1329 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1330 the names libpcre and libpcreposix, not just pcre and pcreposix.
1331
1332 (2) The above change means that pcretest and pcregrep are now correctly
1333 linked with the newly-built libraries, not previously installed ones.
1334
1335 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1336
1337 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1338 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1339 UTF-8 newline character). The key issue is that the pattern starts .*;
1340 this means that the match must be either at the beginning, or after a
1341 newline. The bug was in the code for advancing after a failed match and
1342 checking that the new position followed a newline. It was not taking
1343 account of UTF-8 characters correctly.
1344
1345 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1346 character classes. PCRE was not treating the sequence [:...:] as a
1347 character class unless the ... were all letters. Perl, however, seems to
1348 allow any characters between [: and :], though of course it rejects as
1349 unknown any "names" that contain non-letters, because all the known class
1350 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1351 for example, whereas PCRE did not - it did not recognize a POSIX character
1352 class. This seemed a bit dangerous, so the code has been changed to be
1353 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1354 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1355 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1356 Perl does, and where it didn't before.
1357
1358 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1359 Windows environments %n is disabled by default.
1360
1361
1362 Version 7.4 21-Sep-07
1363 ---------------------
1364
1365 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1366 means that a class such as [\s] counted as "explicit reference to CR or
1367 LF". That isn't really right - the whole point of the change was to try to
1368 help when there was an actual mention of one of the two characters. So now
1369 the change happens only if \r or \n (or a literal CR or LF) character is
1370 encountered.
1371
1372 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1373 of both had grown to the point where there were only 3 bits left.
1374 Fortunately, there was spare space in the data structure, and so I have
1375 moved the internal flags into a new 16-bit field to free up more option
1376 bits.
1377
1378 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1379 but did not set the internal JCHANGED flag - either of these is enough to
1380 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1381 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1382 start sets both bits.
1383
1384 4. Added options (at build time, compile time, exec time) to change \R from
1385 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1386
1387 5. doc/pcresyntax.html was missing from the distribution.
1388
1389 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1390 compatibility, even though it is no longer used.
1391
1392 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1393 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1394 windows.h file is present (where different names are used). [This was
1395 reversed later after testing - see 16 below.]
1396
1397 8. Changed all #include <config.h> to #include "config.h". There were also
1398 some further <pcre.h> cases that I changed to "pcre.h".
1399
1400 9. When pcregrep was used with the --colour option, it missed the line ending
1401 sequence off the lines that it output.
1402
1403 10. It was pointed out to me that arrays of string pointers cause lots of
1404 relocations when a shared library is dynamically loaded. A technique of
1405 using a single long string with a table of offsets can drastically reduce
1406 these. I have refactored PCRE in four places to do this. The result is
1407 dramatic:
1408
1409 Originally: 290
1410 After changing UCP table: 187
1411 After changing error message table: 43
1412 After changing table of "verbs" 36
1413 After changing table of Posix names 22
1414
1415 Thanks to the folks working on Gregex for glib for this insight.
1416
1417 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1418 unicode-properties was also set.
1419
1420 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1421
1422 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1423 checked only for CRLF.
1424
1425 14. Added casts to pcretest.c to avoid compiler warnings.
1426
1427 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1428
1429 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1430 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1431 entirely. This removes changes made in 7 above.
1432
1433 17. The CMake files have been updated, and there is now more information about
1434 building with CMake in the NON-UNIX-USE document.
1435
1436
1437 Version 7.3 28-Aug-07
1438 ---------------------
1439
1440 1. In the rejigging of the build system that eventually resulted in 7.1, the
1441 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1442 brackets there is not right, since it causes compilers to look for an
1443 installed pcre.h, not the version that is in the source that is being
1444 compiled (which of course may be different). I have changed it back to:
1445
1446 #include "pcre.h"
1447
1448 I have a vague recollection that the change was concerned with compiling in
1449 different directories, but in the new build system, that is taken care of
1450 by the VPATH setting the Makefile.
1451
1452 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1453 when the subject happened to end in the byte 0x85 (e.g. if the last
1454 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1455 characters but of course it shouldn't be taken as a newline when it is part
1456 of another character. The bug was that, for an unlimited repeat of . in
1457 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1458 characters when looking for a newline.
1459
1460 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1461
1462 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1463 in debug output.
1464
1465 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1466 long printing in the pcrecpp unittest when running under MinGW.
1467
1468 6. ESC_K was left out of the EBCDIC table.
1469
1470 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1471 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1472 limit also applies to "virtual nesting" when a pattern is recursive, and in
1473 this case 1000 isn't so big. I have been able to remove this limit at the
1474 expense of backing off one optimization in certain circumstances. Normally,
1475 when pcre_exec() would call its internal match() function recursively and
1476 immediately return the result unconditionally, it uses a "tail recursion"
1477 feature to save stack. However, when a subpattern that can match an empty
1478 string has an unlimited repetition quantifier, it no longer makes this
1479 optimization. That gives it a stack frame in which to save the data for
1480 checking that an empty string has been matched. Previously this was taken
1481 from the 1000-entry workspace that had been reserved. So now there is no
1482 explicit limit, but more stack is used.
1483
1484 8. Applied Daniel's patches to solve problems with the import/export magic
1485 syntax that is required for Windows, and which was going wrong for the
1486 pcreposix and pcrecpp parts of the library. These were overlooked when this
1487 problem was solved for the main library.
1488
1489 9. There were some crude static tests to avoid integer overflow when computing
1490 the size of patterns that contain repeated groups with explicit upper
1491 limits. As the maximum quantifier is 65535, the maximum group length was
1492 set at 30,000 so that the product of these two numbers did not overflow a
1493 32-bit integer. However, it turns out that people want to use groups that
1494 are longer than 30,000 bytes (though not repeat them that many times).
1495 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1496 made it possible to implement the integer overflow checks in a much more
1497 dynamic way, which I have now done. The artificial limitation on group
1498 length has been removed - we now have only the limit on the total length of
1499 the compiled pattern, which depends on the LINK_SIZE setting.
1500
1501 10. Fixed a bug in the documentation for get/copy named substring when
1502 duplicate names are permitted. If none of the named substrings are set, the
1503 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1504 empty string.
1505
1506 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1507 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1508 because the ] is interpreted as the first data character and the
1509 terminating ] is not found. PCRE has been made compatible with Perl in this
1510 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1511 cause memory overwriting.
1512
1513 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1514 string has been matched (to stop an infinite loop). It was not recognizing
1515 a conditional subpattern that could match an empty string if that
1516 subpattern was within another subpattern. For example, it looped when
1517 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1518 condition was not nested. This bug has been fixed.
1519
1520 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1521 past the start of the subject in the presence of bytes with the top bit
1522 set, for example "\x8aBCD".
1523
1524 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1525 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1526
1527 14. Optimized (?!) to (*FAIL).
1528
1529 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1530 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1531 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1532 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1533 does: it's just the validity check that is more restrictive.
1534
1535 16. Inserted checks for integer overflows during escape sequence (backslash)
1536 processing, and also fixed erroneous offset values for syntax errors during
1537 backslash processing.
1538
1539 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1540 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1541
1542 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1543 caused an overrun.
1544
1545 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1546 something other than just ASCII characters) inside a group that had an
1547 unlimited repeat caused a loop at compile time (while checking to see
1548 whether the group could match an empty string).
1549
1550 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1551 [\P{Any}] did so. (Error in the code for printing property names.)
1552
1553 21. An orphan \E inside a character class could cause a crash.
1554
1555 22. A repeated capturing bracket such as (A)? could cause a wild memory
1556 reference during compilation.
1557
1558 23. There are several functions in pcre_compile() that scan along a compiled
1559 expression for various reasons (e.g. to see if it's fixed length for look
1560 behind). There were bugs in these functions when a repeated \p or \P was
1561 present in the pattern. These operators have additional parameters compared
1562 with \d, etc, and these were not being taken into account when moving along
1563 the compiled data. Specifically:
1564
1565 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1566 length.
1567
1568 (b) An item such as \pL+ within a repeated group could cause crashes or
1569 loops.
1570
1571 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1572 "reference to non-existent subpattern" error.
1573
1574 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1575
1576 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1577 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1578
1579 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1580
1581 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1582 character were causing crashes (broken optimization).
1583
1584 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1585 \p or \P) caused a compile-time loop.
1586
1587 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1588 break. For example, the unstudied pattern [\r\n]A does not match the string
1589 "\r\nA" because change 7.0/46 below moves the current point on by two
1590 characters after failing to match at the start. However, the pattern \nA
1591 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1592 the same is true. There doesn't seem any very clean way out of this, but
1593 what I have chosen to do makes the common cases work: PCRE now takes note
1594 of whether there can be an explicit match for \r or \n anywhere in the
1595 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1596 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1597 pattern has explicit CR or LF references.
1598
1599 29. Added (*CR) etc for changing newline setting at start of pattern.
1600
1601
1602 Version 7.2 19-Jun-07
1603 ---------------------
1604
1605 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1606 which is apparently normally available under Windows.
1607
1608 2. Re-jig the pcregrep tests with different newline settings in an attempt
1609 to make them independent of the local environment's newline setting.
1610
1611 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1612
1613 4. Some of the "internals" tests were previously cut out when the link size
1614 was not 2, because the output contained actual offsets. The recent new
1615 "Z" feature of pcretest means that these can be cut out, making the tests
1616 usable with all link sizes.
1617
1618 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1619 stack recursion. This gives a massive performance boost under BSD, but just
1620 a small improvement under Linux. However, it saves one field in the frame
1621 in all cases.
1622
1623 6. Added more features from the forthcoming Perl 5.10:
1624
1625 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1626 recursion call. It refers to the nth most recently opened parentheses.
1627
1628 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1629 to be opened parentheses.
1630
1631 (c) Conditions that refer to capturing parentheses can be specified
1632 relatively, for example, (?(-2)... or (?(+3)...
1633
1634 (d) \K resets the start of the current match so that everything before
1635 is not part of it.
1636
1637 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1638
1639 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1640 reference syntax.
1641
1642 (g) (?| introduces a group in which the numbering of parentheses in each
1643 alternative starts with the same number.
1644
1645 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1646
1647 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1648 PCRE_INFO_JCHANGED.
1649
1650 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1651 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1652 for detecting groups that can match an empty string.
1653
1654 9. A pattern with a very large number of alternatives (more than several
1655 hundred) was running out of internal workspace during the pre-compile
1656 phase, where pcre_compile() figures out how much memory will be needed. A
1657 bit of new cunning has reduced the workspace needed for groups with
1658 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1659 workspace instead of running out of the 4096 that are available.
1660
1661 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1662
1663 11. Applied patch from Google to remove an optimization that didn't quite work.
1664 The report of the bug said:
1665
1666 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1667 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1668 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1669
1670 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1671 it matched the wrong number of bytes.
1672
1673
1674 Version 7.1 24-Apr-07
1675 ---------------------
1676
1677 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1678 that is more "standard", making use of automake and other Autotools. There
1679 is some re-arrangement of the files and adjustment of comments consequent
1680 on this.
1681
1682 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1683 for recursive directory scanning broke on some systems because the files
1684 are not scanned in any specific order and on different systems the order
1685 was different. A call to "sort" has been inserted into RunGrepTest for the
1686 approprate test as a short-term fix. In the longer term there may be an
1687 alternative.
1688
1689 3. I had an email from Eric Raymond about problems translating some of PCRE's
1690 man pages to HTML (despite the fact that I distribute HTML pages, some
1691 people do their own conversions for various reasons). The problems
1692 concerned the use of low-level troff macros .br and .in. I have therefore
1693 removed all such uses from the man pages (some were redundant, some could
1694 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1695 HTML has been updated to handle .nf/.fi and to complain if it encounters
1696 .br or .in.
1697
1698 4. Updated comments in configure.ac that get placed in config.h.in and also
1699 arranged for config.h to be included in the distribution, with the name
1700 config.h.generic, for the benefit of those who have to compile without
1701 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1702
1703 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1704 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1705 makevp.bat for the current PCRE, using the additional files
1706 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1707
1708 6. A Windows user reported a minor discrepancy with test 2, which turned out
1709 to be caused by a trailing space on an input line that had got lost in his
1710 copy. The trailing space was an accident, so I've just removed it.
1711
1712 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1713 that is needed.
1714
1715 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1716 as "const" (a) because they are and (b) because it helps the PHP
1717 maintainers who have recently made a script to detect big data structures
1718 in the php code that should be moved to the .rodata section. I remembered
1719 to update Builducptable as well, so it won't revert if ucptable.h is ever
1720 re-created.
1721
1722 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1723 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1724 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1725 support is not required. This saves 1.5-2K of code, which is important in
1726 some applications.
1727
1728 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1729 so as not to refer to the tables, even though these functions will never be
1730 called when UTF-8 support is disabled. Otherwise there are problems with a
1731 shared library.
1732
1733 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1734
1735 (a) It was defining its arguments as char * instead of void *.
1736
1737 (b) It was assuming that all moves were upwards in memory; this was true
1738 a long time ago when I wrote it, but is no longer the case.
1739
1740 The emulated memove() is provided for those environments that have neither
1741 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1742 is clearly not the case, as these two bugs were recently reported.
1743
1744 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1745 and Detrail to create the HTML documentation, the .txt form of the man
1746 pages, and it removes trailing spaces from listed files. It also creates
1747 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1748 case, it wraps all the #defines with #ifndefs. This script should be run
1749 before "make dist".
1750
1751 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1752 with Unicode property support.
1753
1754 (a) For a maximizing quantifier, if the two different cases of the
1755 character were of different lengths in their UTF-8 codings (there are
1756 some cases like this - I found 11), and the matching function had to
1757 back up over a mixture of the two cases, it incorrectly assumed they
1758 were both the same length.
1759
1760 (b) When PCRE was configured to use the heap rather than the stack for
1761 recursion during matching, it was not correctly preserving the data for
1762 the other case of a UTF-8 character when checking ahead for a match
1763 while processing a minimizing repeat. If the check also involved
1764 matching a wide character, but failed, corruption could cause an
1765 erroneous result when trying to check for a repeat of the original
1766 character.
1767
1768 13. Some tidying changes to the testing mechanism:
1769
1770 (a) The RunTest script now detects the internal link size and whether there
1771 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1772 values substituted by "configure". (The RunGrepTest script already did
1773 this for UTF-8.) The configure.ac script no longer substitutes the
1774 relevant variables.
1775
1776 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1777 with length and offset values. This means that the output is different
1778 for different internal link sizes. Test 2 is skipped for link sizes
1779 other than 2 because of this, bypassing the problem. Unfortunately,
1780 there was also a test in test 3 (the locale tests) that used /B and
1781 failed for link sizes other than 2. Rather than cut the whole test out,
1782 I have added a new /Z option to pcretest that replaces the length and
1783 offset values with spaces. This is now used to make test 3 independent
1784 of link size. (Test 2 will be tidied up later.)
1785
1786 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1787 segmentation fault instead of returning the appropriate error message.
1788
1789 15. In multiline mode when the newline sequence was set to "any", the pattern
1790 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1791 This doesn't seem right; it now treats the CRLF combination as the line
1792 ending, and so does not match in that case. It's only a pattern such as ^$
1793 that would hit this one: something like ^ABC$ would have failed after \r
1794 and then tried again after \r\n.
1795
1796 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1797 in an attempt to make files that differ only in their line terminators
1798 compare equal. This works on Linux.
1799
1800 17. Under certain error circumstances pcregrep might try to free random memory
1801 as it exited. This is now fixed, thanks to valgrind.
1802
1803 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1804 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1805 was because its rules for how to advance for /g after matching an empty
1806 string at the end of a line did not allow for this case. They now check for
1807 it specially.
1808
1809 20. pcretest is supposed to handle patterns and data of any length, by
1810 extending its buffers when necessary. It was getting this wrong when the
1811 buffer for a data line had to be extended.
1812
1813 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1814 CRLF as a newline sequence.
1815
1816 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1817 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1818 I have nevertheless tidied it up.
1819
1820 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1821
1822 24. Added a man page for pcre-config.
1823
1824
1825 Version 7.0 19-Dec-06
1826 ---------------------
1827
1828 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1829 moving to gcc 4.1.1.
1830
1831 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1832 sys/time.h, which is documented as needed for this function. It doesn't
1833 seem to matter on Linux, but it showed up on some releases of OS X.
1834
1835 3. It seems that there are systems where bytes whose values are greater than
1836 127 match isprint() in the "C" locale. The "C" locale should be the
1837 default when a C program starts up. In most systems, only ASCII printing
1838 characters match isprint(). This difference caused the output from pcretest
1839 to vary, making some of the tests fail. I have changed pcretest so that:
1840
1841 (a) When it is outputting text in the compiled version of a pattern, bytes
1842 other than 32-126 are always shown as hex escapes.
1843
1844 (b) When it is outputting text that is a matched part of a subject string,
1845 it does the same, unless a different locale has been set for the match
1846 (using the /L modifier). In this case, it uses isprint() to decide.
1847
1848 4. Fixed a major bug that caused incorrect computation of the amount of memory
1849 required for a compiled pattern when options that changed within the
1850 pattern affected the logic of the preliminary scan that determines the
1851 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1852 that the computed length was too small. The symptoms of this bug were
1853 either the PCRE error "internal error: code overflow" from pcre_compile(),
1854 or a glibc crash with a message such as "pcretest: free(): invalid next
1855 size (fast)". Examples of patterns that provoked this bug (shown in
1856 pcretest format) are:
1857
1858 /(?-x: )/x
1859 /(?x)(?-x: \s*#\s*)/
1860 /((?i)[\x{c0}])/8
1861 /(?i:[\x{c0}])/8
1862
1863 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1864 is now done differently.
1865
1866 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1867 wrapper classes; (b) implement a new function in the C++ scanner that is
1868 more efficient than the old way of doing things because it avoids levels of
1869 recursion in the regex matching; (c) add a paragraph to the documentation
1870 for the FullMatch() function.
1871
1872 6. The escape sequence \n was being treated as whatever was defined as
1873 "newline". Not only was this contrary to the documentation, which states
1874 that \n is character 10 (hex 0A), but it also went horribly wrong when
1875 "newline" was defined as CRLF. This has been fixed.
1876
1877 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1878 was being set to -1 for the "end of line" case (supposedly a value that no
1879 character can have). Though this value is never used (the check for end of
1880 line is "zero bytes in current character"), it caused compiler complaints.
1881 I've changed it to 0xffffffff.
1882
1883 8. In pcre_version.c, the version string was being built by a sequence of
1884 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1885 string (as it is for production releases) called a macro with an empty
1886 argument. The C standard says the result of this is undefined. The gcc
1887 compiler treats it as an empty string (which was what was wanted) but it is
1888 reported that Visual C gives an error. The source has been hacked around to
1889 avoid this problem.
1890
1891 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1892 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1893 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1894 of them did).
1895
1896 10. Originally, pcretest opened its input and output without "b"; then I was
1897 told that "b" was needed in some environments, so it was added for release
1898 5.0 to both the input and output. (It makes no difference on Unix-like
1899 systems.) Later I was told that it is wrong for the input on Windows. I've
1900 now abstracted the modes into two macros, to make it easier to fiddle with
1901 them, and removed "b" from the input mode under Windows.
1902
1903 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1904
1905 12. Added -help and --help to pcretest as an official way of being reminded
1906 of the options.
1907
1908 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1909 and pcrecpp.cc because they annoy compilers at high warning levels.
1910
1911 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1912
1913 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1914 scripts are not C programs :-) and which was not noticed because it works
1915 on Linux.
1916
1917 16. pcretest is supposed to handle any length of pattern and data line (as one
1918 line or as a continued sequence of lines) by extending its input buffer if
1919 necessary. This feature was broken for very long pattern lines, leading to
1920 a string of junk being passed to pcre_compile() if the pattern was longer
1921 than about 50K.
1922
1923 17. I have done a major re-factoring of the way pcre_compile() computes the
1924 amount of memory needed for a compiled pattern. Previously, there was code
1925 that made a preliminary scan of the pattern in order to do this. That was
1926 OK when PCRE was new, but as the facilities have expanded, it has become
1927 harder and harder to keep it in step with the real compile phase, and there
1928 have been a number of bugs (see for example, 4 above). I have now found a
1929 cunning way of running the real compile function in a "fake" mode that
1930 enables it to compute how much memory it would need, while actually only
1931 ever using a few hundred bytes of working memory and without too many
1932 tests of the mode. This should make future maintenance and development
1933 easier. A side effect of this work is that the limit of 200 on the nesting
1934 depth of parentheses has been removed (though this was never a serious
1935 limitation, I suspect). However, there is a downside: pcre_compile() now
1936 runs more slowly than before (30% or more, depending on the pattern). I
1937 hope this isn't a big issue. There is no effect on runtime performance.
1938
1939 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1940 newline (only possible for the last line of a file) and it was a
1941 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1942
1943 19. Added additional timing features to pcretest. (1) The -tm option now times
1944 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1945 separate command line item, by a number that specifies the number of
1946 repeats to use when timing. The default is 50000; this gives better
1947 precision, but takes uncomfortably long for very large patterns.
1948
1949 20. Extended pcre_study() to be more clever in cases where a branch of a
1950 subpattern has no definite first character. For example, (a*|b*)[cd] would
1951 previously give no result from pcre_study(). Now it recognizes that the
1952 first character must be a, b, c, or d.
1953
1954 21. There was an incorrect error "recursive call could loop indefinitely" if
1955 a subpattern (or the entire pattern) that was being tested for matching an
1956 empty string contained only one non-empty item after a nested subpattern.
1957 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1958 incorrectly, because the \d was being skipped in the check.
1959
1960 22. The pcretest program now has a new pattern option /B and a command line
1961 option -b, which is equivalent to adding /B to every pattern. This causes
1962 it to show the compiled bytecode, without the additional information that
1963 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1964 is the same as /B/I).
1965
1966 23. A new optimization is now able automatically to treat some sequences such
1967 as a*b as a*+b. More specifically, if something simple (such as a character
1968 or a simple class like \d) has an unlimited quantifier, and is followed by
1969 something that cannot possibly match the quantified thing, the quantifier
1970 is automatically "possessified".
1971
1972 24. A recursive reference to a subpattern whose number was greater than 39
1973 went wrong under certain circumstances in UTF-8 mode. This bug could also
1974 have affected the operation of pcre_study().
1975
1976 25. Realized that a little bit of performance could be had by replacing
1977 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1978
1979 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1980
1981 27. Possessive quantifiers such as a++ were previously implemented by turning
1982 them into atomic groups such as ($>a+). Now they have their own opcodes,
1983 which improves performance. This includes the automatically created ones
1984 from 23 above.
1985
1986 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1987 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1988 the first matched character to be a colon. This applied both to named and
1989 numbered groups.
1990
1991 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1992
1993 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1994 building PCRE on Windows easier, so I have included it in the distribution.
1995
1996 31. There is now a check in pcretest against a ridiculously large number being
1997 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1998 loop, the loop is abandoned.
1999
2000 32. Forward references to subpatterns in conditions such as (?(2)...) where
2001 subpattern 2 is defined later cause pcre_compile() to search forwards in
2002 the pattern for the relevant set of parentheses. This search went wrong
2003 when there were unescaped parentheses in a character class, parentheses
2004 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2005
2006 33. "Subroutine" calls and backreferences were previously restricted to
2007 referencing subpatterns earlier in the regex. This restriction has now
2008 been removed.
2009
2010 34. Added a number of extra features that are going to be in Perl 5.10. On the
2011 whole, these are just syntactic alternatives for features that PCRE had
2012 previously implemented using the Python syntax or my own invention. The
2013 other formats are all retained for compatibility.
2014
2015 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2016 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2017 also .NET compatible.
2018
2019 (b) A recursion or subroutine call to a named group can now be defined as
2020 (?&name) as well as (?P>name).
2021
2022 (c) A backreference to a named group can now be defined as \k<name> or
2023 \k'name' as well as (?P=name). The new forms, as well as being in Perl
2024 5.10, are also .NET compatible.
2025
2026 (d) A conditional reference to a named group can now use the syntax
2027 (?(<name>) or (?('name') as well as (?(name).
2028
2029 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2030 groups (named and numbered) that are never evaluated inline, but can be
2031 called as "subroutines" from elsewhere. In effect, the DEFINE condition
2032 is always false. There may be only one alternative in such a group.
2033
2034 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2035 as the simple (?(R). The condition is true only if the most recent
2036 recursion is that of the given number or name. It does not search out
2037 through the entire recursion stack.
2038
2039 (g) The escape \gN or \g{N} has been added, where N is a positive or
2040 negative number, specifying an absolute or relative reference.
2041
2042 35. Tidied to get rid of some further signed/unsigned compiler warnings and
2043 some "unreachable code" warnings.
2044
2045 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2046 things, this adds five new scripts.
2047
2048 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2049 There were also incompatibilities regarding the handling of \Q..\E inside
2050 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2051 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2052
2053 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2054 matches an empty string, and forcibly breaks the loop. There were bugs in
2055 this code in non-simple cases. For a pattern such as ^(a()*)* matched
2056 against aaaa the result was just "a" rather than "aaaa", for example. Two
2057 separate and independent bugs (that affected different cases) have been
2058 fixed.
2059
2060 39. Refactored the code to abolish the use of different opcodes for small
2061 capturing bracket numbers. This is a tidy that I avoided doing when I
2062 removed the limit on the number of capturing brackets for 3.5 back in 2001.
2063 The new approach is not only tidier, it makes it possible to reduce the
2064 memory needed to fix the previous bug (38).
2065
2066 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2067 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2068 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2069 mode.
2070
2071 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2072 report.
2073
2074 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2075 copy construction and assignment in the C++ wrapper.
2076
2077 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2078 couple of bugs that could have given wrong results in the "--newline=crlf"
2079 case.
2080
2081 44. Added a number of casts and did some reorganization of signed/unsigned int
2082 variables following suggestions from Dair Grant. Also renamed the variable
2083 "this" as "item" because it is a C++ keyword.
2084
2085 45. Arranged for dftables to add
2086
2087 #include "pcre_internal.h"
2088
2089 to pcre_chartables.c because without it, gcc 4.x may remove the array
2090 definition from the final binary if PCRE is built into a static library and
2091 dead code stripping is activated.
2092
2093 46. For an unanchored pattern, if a match attempt fails at the start of a
2094 newline sequence, and the newline setting is CRLF or ANY, and the next two
2095 characters are CRLF, advance by two characters instead of one.
2096
2097
2098 Version 6.7 04-Jul-06
2099 ---------------------
2100
2101 1. In order to handle tests when input lines are enormously long, pcretest has
2102 been re-factored so that it automatically extends its buffers when
2103 necessary. The code is crude, but this _is_ just a test program. The
2104 default size has been increased from 32K to 50K.
2105
2106 2. The code in pcre_study() was using the value of the re argument before
2107 testing it for NULL. (Of course, in any sensible call of the function, it
2108 won't be NULL.)
2109
2110 3. The memmove() emulation function in pcre_internal.h, which is used on
2111 systems that lack both memmove() and bcopy() - that is, hardly ever -
2112 was missing a "static" storage class specifier.
2113
2114 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2115 containing an extended class (one that cannot be represented by a bitmap
2116 because it contains high-valued characters or Unicode property items, e.g.
2117 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2118 pattern, but PCRE should not loop if you do not (it no longer does).
2119 [Detail: two cases were found: (a) a repeated subpattern containing an
2120 extended class; (b) a recursive reference to a subpattern that followed a
2121 previous extended class. It wasn't skipping over the extended class
2122 correctly when UTF-8 mode was not set.]
2123
2124 5. A negated single-character class was not being recognized as fixed-length
2125 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2126 compile error "lookbehind assertion is not fixed length".
2127
2128 6. The RunPerlTest auxiliary script was showing an unexpected difference
2129 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2130 write a Perl script that can interpret lines of an input file either as
2131 byte characters or as UTF-8, which is what "perltest" was being required to
2132 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2133 can't do is switch easily at run time between having the "use utf8;" pragma
2134 or not. In the end, I fudged it by using the RunPerlTest script to insert
2135 "use utf8;" explicitly for the UTF-8 tests.
2136
2137 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2138 the end of the subject string, contrary to the documentation and to what
2139 Perl does. This was true of both matching functions. Now it matches only at
2140 the start of the subject and immediately after *internal* newlines.
2141
2142 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2143 a pointer to an int instead of a pointer to an unsigned long int. This
2144 caused problems on 64-bit systems.
2145
2146 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2147 instance of the 'standard' template library not being so standard".
2148
2149 10. There was no check on the number of named subpatterns nor the maximum
2150 length of a subpattern name. The product of these values is used to compute
2151 the size of the memory block for a compiled pattern. By supplying a very
2152 long subpattern name and a large number of named subpatterns, the size
2153 computation could be caused to overflow. This is now prevented by limiting
2154 the length of names to 32 characters, and the number of named subpatterns
2155 to 10,000.
2156
2157 11. Subpatterns that are repeated with specific counts have to be replicated in
2158 the compiled pattern. The size of memory for this was computed from the
2159 length of the subpattern and the repeat count. The latter is limited to
2160 65535, but there was no limit on the former, meaning that integer overflow
2161 could in principle occur. The compiled length of a repeated subpattern is
2162 now limited to 30,000 bytes in order to prevent this.
2163
2164 12. Added the optional facility to have named substrings with the same name.
2165
2166 13. Added the ability to use a named substring as a condition, using the
2167 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2168 are numbers (not recommended). Forward references are permitted.
2169
2170 14. Added forward references in named backreferences (if you see what I mean).
2171
2172 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2173 pattern could run off the end of the subject. For example, the pattern
2174 "(?s)(.{1,5})"8 did this with the subject "ab".
2175
2176 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2177 PCRE_CASELESS was set when matching characters that were quantified with ?
2178 or *.
2179
2180 17. A character class other than a single negated character that had a minimum
2181 but no maximum quantifier - for example [ab]{6,} - was not handled
2182 correctly by pce_dfa_exec(). It would match only one character.
2183
2184 18. A valid (though odd) pattern that looked like a POSIX character
2185 class but used an invalid character after [ (for example [[,abc,]]) caused
2186 pcre_compile() to give the error "Failed: internal error: code overflow" or
2187 in some cases to crash with a glibc free() error. This could even happen if
2188 the pattern terminated after [[ but there just happened to be a sequence of
2189 letters, a binary zero, and a closing ] in the memory that followed.
2190
2191 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2192 over the years. Originally (before any Unicode support), just the bottom 8
2193 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2194 output from "man perlunicode" includes this:
2195
2196 The regular expression compiler produces polymorphic opcodes. That
2197 is, the pattern adapts to the data and automatically switches to
2198 the Unicode character scheme when presented with Unicode data--or
2199 instead uses a traditional byte scheme when presented with byte
2200 data.
2201
2202 Sadly, a wide octal escape does not cause a switch, and in a string with
2203 no other multibyte characters, these octal escapes are treated as before.
2204 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2205 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2206 Unicode string.
2207
2208 I have not perpetrated such confusion in PCRE. Up till now, it took just
2209 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2210 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2211 translate to the appropriate multibyte character.
2212
2213 29. Applied some refactoring to reduce the number of warnings from Microsoft
2214 and Borland compilers. This has included removing the fudge introduced
2215 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2216 a warning about an unused variable.
2217
2218 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2219 characters since release 4.0, because Perl (from release 5.004) does not.
2220 [Or at least, is documented not to: some releases seem to be in conflict
2221 with the documentation.] However, when a pattern was studied with
2222 pcre_study() and all its branches started with \s, PCRE still included VT
2223 as a possible starting character. Of course, this did no harm; it just
2224 caused an unnecessary match attempt.
2225
2226 22. Removed a now-redundant internal flag bit that recorded the fact that case
2227 dependency changed within the pattern. This was once needed for "required
2228 byte" processing, but is no longer used. This recovers a now-scarce options
2229 bit. Also moved the least significant internal flag bit to the most-
2230 significant bit of the word, which was not previously used (hangover from
2231 the days when it was an int rather than a uint) to free up another bit for
2232 the future.
2233
2234 23. Added support for CRLF line endings as well as CR and LF. As well as the
2235 default being selectable at build time, it can now be changed at runtime
2236 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2237 specify that it is scanning data with non-default line endings.
2238
2239 24. Changed the definition of CXXLINK to make it agree with the definition of
2240 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2241
2242 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2243 recursions. This makes a big different to stack usage for some patterns.
2244
2245 26. If a subpattern containing a named recursion or subroutine reference such
2246 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2247 the space required for the compiled pattern went wrong and gave too small a
2248 value. Depending on the environment, this could lead to "Failed: internal
2249 error: code overflow at offset 49" or "glibc detected double free or
2250 corruption" errors.
2251
2252 27. Applied patches from Google (a) to support the new newline modes and (b) to
2253 advance over multibyte UTF-8 characters in GlobalReplace.
2254
2255 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2256 difference for some implementation of PCRE in some Windows version.
2257
2258 29. Added some extra testing facilities to pcretest:
2259
2260 \q<number> in a data line sets the "match limit" value
2261 \Q<number> in a data line sets the "match recursion limt" value
2262 -S <number> sets the stack size, where <number> is in megabytes
2263
2264 The -S option isn't available for Windows.
2265
2266
2267 Version 6.6 06-Feb-06
2268 ---------------------
2269
2270 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2271 in pcreposix.h. I have copied the definition from pcre.h.
2272
2273 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2274 because pcre.h is no longer a built file.
2275
2276 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2277 not normally included in the compiled code.
2278
2279
2280 Version 6.5 01-Feb-06
2281 ---------------------
2282
2283 1. When using the partial match feature with pcre_dfa_exec(), it was not
2284 anchoring the second and subsequent partial matches at the new starting
2285 point. This could lead to incorrect results. For example, with the pattern
2286 /1234/, partially matching against "123" and then "a4" gave a match.
2287
2288 2. Changes to pcregrep:
2289
2290 (a) All non-match returns from pcre_exec() were being treated as failures
2291 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2292 error message is output. Some extra information is given for the
2293 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2294 probably the only errors that are likely to be caused by users (by
2295 specifying a regex that has nested indefinite repeats, for instance).
2296 If there are more than 20 of these errors, pcregrep is abandoned.
2297
2298 (b) A binary zero was treated as data while matching, but terminated the
2299 output line if it was written out. This has been fixed: binary zeroes
2300 are now no different to any other data bytes.
2301
2302 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2303 used to set a locale for matching. The --locale=xxxx long option has
2304 been added (no short equivalent) to specify a locale explicitly on the
2305 pcregrep command, overriding the environment variables.
2306
2307 (d) When -B was used with -n, some line numbers in the output were one less
2308 than they should have been.
2309
2310 (e) Added the -o (--only-matching) option.
2311
2312 (f) If -A or -C was used with -c (count only), some lines of context were
2313 accidentally printed for the final match.
2314
2315 (g) Added the -H (--with-filename) option.
2316
2317 (h) The combination of options -rh failed to suppress file names for files
2318 that were found from directory arguments.
2319
2320 (i) Added the -D (--devices) and -d (--directories) options.
2321
2322 (j) Added the -F (--fixed-strings) option.
2323
2324 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2325
2326 (l) Added the --colo(u)r option.
2327
2328 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2329 is not present by default.
2330
2331 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2332 items such as (?R) or (?1), when the recursion could match a number of
2333 alternatives. If it matched one of the alternatives, but subsequently,
2334 outside the recursion, there was a failure, the code tried to back up into
2335 the recursion. However, because of the way PCRE is implemented, this is not
2336 possible, and the result was an incorrect result from the match.
2337
2338 In order to prevent this happening, the specification of recursion has
2339 been changed so that all such subpatterns are automatically treated as
2340 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2341
2342 4. I had overlooked the fact that, in some locales, there are characters for
2343 which isalpha() is true but neither isupper() nor islower() are true. In
2344 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2345 and ordfeminine) are like this. This affected the treatment of \w and \W
2346 when they appeared in character classes, but not when they appeared outside
2347 a character class. The bit map for "word" characters is now created
2348 separately from the results of isalnum() instead of just taking it from the
2349 upper, lower, and digit maps. (Plus the underscore character, of course.)
2350
2351 5. The above bug also affected the handling of POSIX character classes such as
2352 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2353 permanent tables. Instead, the bit maps for such a class were previously
2354 created as the appropriate unions of the upper, lower, and digit bitmaps.
2355 Now they are created by subtraction from the [[:word:]] class, which has
2356 its own bitmap.
2357
2358 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2359 It is created by subtracting the vertical space characters (\x09, \x0a,
2360 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2361 subtraction was done in the overall bitmap for a character class, meaning
2362 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2363 be recognized. This bug has been fixed.
2364
2365 7. Patches from the folks at Google:
2366
2367 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2368 real life, but is still worth protecting against".
2369
2370 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2371 regular expressions".
2372
2373 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2374 have it.
2375
2376 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2377 "configure" and the latter not, in order to fix a problem somebody had
2378 with compiling the Arg class on HP-UX.
2379
2380 (e) Improve the error-handling of the C++ wrapper a little bit.
2381
2382 (f) New tests for checking recursion limiting.
2383
2384 8. The pcre_memmove() function, which is used only if the environment does not
2385 have a standard memmove() function (and is therefore rarely compiled),
2386 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2387 returning a result (though PCRE never actually uses the result).
2388
2389 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2390 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2391 returned instead of calling malloc() with an overflowing number that would
2392 most likely cause subsequent chaos.
2393
2394 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2395
2396 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2397 with this option is matched, the nmatch and pmatch options of regexec() are
2398 ignored.
2399
2400 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2401 provided in case anyone wants to the the POSIX interface with UTF-8
2402 strings.
2403
2404 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2405 C++ linking (needed for some HP-UX environments).
2406
2407 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2408 (unused parameter) and in the pcre_printint() function (omitted "default"
2409 switch label when the default is to do nothing).
2410
2411 15. Added some code to make it possible, when PCRE is compiled as a C++
2412 library, to replace subject pointers for pcre_exec() with a smart pointer
2413 class, thus making it possible to process discontinuous strings.
2414
2415 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2416 much the same function. They were added by different people who were trying
2417 to make PCRE easy to compile on non-Unix systems. It has been suggested
2418 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2419 for compiling on Windows systems. I have therefore replaced it with
2420 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2421 defaults to "extern" for C or "extern C" for C++, which works fine on
2422 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2423 SCOPE with something explicit in config.h. In addition:
2424
2425 (a) pcreposix.h still had just "extern" instead of either of these macros;
2426 I have replaced it with PCRE_DATA_SCOPE.
2427
2428 (b) Functions such as _pcre_xclass(), which are internal to the library,
2429 but external in the C sense, all had PCRE_EXPORT in their definitions.
2430 This is apparently wrong for the Windows case, so I have removed it.
2431 (It makes no difference on Unix-like systems.)
2432
2433 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2434 of recursive calls to match(). This is different to MATCH_LIMIT because
2435 that limits the total number of calls to match(), not all of which increase
2436 the depth of recursion. Limiting the recursion depth limits the amount of
2437 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2438 when PCRE is compiled, and changed at run time. A patch from Google adds
2439 this functionality to the C++ interface.
2440
2441 18. Changes to the handling of Unicode character properties:
2442
2443 (a) Updated the table to Unicode 4.1.0.
2444
2445 (b) Recognize characters that are not in the table as "Cn" (undefined).
2446
2447 (c) I revised the way the table is implemented to a much improved format
2448 which includes recognition of ranges. It now supports the ranges that
2449 are defined in UnicodeData.txt, and it also amalgamates other
2450 characters into ranges. This has reduced the number of entries in the
2451 table from around 16,000 to around 3,000, thus reducing its size
2452 considerably. I realized I did not need to use a tree structure after
2453 all - a binary chop search is just as efficient. Having reduced the
2454 number of entries, I extended their size from 6 bytes to 8 bytes to
2455 allow for more data.
2456
2457 (d) Added support for Unicode script names via properties such as \p{Han}.
2458
2459 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2460 matching that character.
2461
2462 20. When matching a repeated Unicode property with a minimum greater than zero,
2463 (for example \pL{2,}), PCRE could look past the end of the subject if it
2464 reached it while seeking the minimum number of characters. This could
2465 happen only if some of the characters were more than one byte long, because
2466 there is a check for at least the minimum number of bytes.
2467
2468 21. Refactored the implementation of \p and \P so as to be more general, to
2469 allow for more different types of property in future. This has changed the
2470 compiled form incompatibly. Anybody with saved compiled patterns that use
2471 \p or \P will have to recompile them.
2472
2473 22. Added "Any" and "L&" to the supported property types.
2474
2475 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2476 but give a compile time error if the value is greater than 0xff.
2477
2478 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2479 accidentally not being installed or uninstalled.
2480
2481 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2482 made were to insert the current release number. This seemed silly, because
2483 it made things harder for people building PCRE on systems that don't run
2484 "configure". I have turned pcre.h into a distributed file, no longer built
2485 by "configure", with the version identification directly included. There is
2486 no longer a pcre.h.in file.
2487
2488 However, this change necessitated a change to the pcre-config script as
2489 well. It is built from pcre-config.in, and one of the substitutions was the
2490 release number. I have updated configure.ac so that ./configure now finds
2491 the release number by grepping pcre.h.
2492
2493 26. Added the ability to run the tests under valgrind.
2494
2495
2496 Version 6.4 05-Sep-05
2497 ---------------------
2498
2499 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2500 "--" to be printed when multiple files were scanned, even when none of the
2501 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2502 consider it to be a bug, and have restored the previous behaviour.
2503
2504 2. A couple of code tidies to get rid of compiler warnings.
2505
2506 3. The pcretest program used to cheat by referring to symbols in the library
2507 whose names begin with _pcre_. These are internal symbols that are not
2508 really supposed to be visible externally, and in some environments it is
2509 possible to suppress them. The cheating is now confined to including
2510 certain files from the library's source, which is a bit cleaner.
2511
2512 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2513 file's purpose clearer.
2514
2515 5. Reorganized pcre_ucp_findchar().
2516
2517
2518 Version 6.3 15-Aug-05
2519 ---------------------
2520
2521 1. The file libpcre.pc.in did not have general read permission in the tarball.
2522
2523 2. There were some problems when building without C++ support:
2524
2525 (a) If C++ support was not built, "make install" and "make test" still
2526 tried to test it.
2527
2528 (b) There were problems when the value of CXX was explicitly set. Some
2529 changes have been made to try to fix these, and ...
2530
2531 (c) --disable-cpp can now be used to explicitly disable C++ support.
2532
2533 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2534 backslash in a target when C++ was disabled. This confuses some
2535 versions of "make", apparently. Using an intermediate variable solves
2536 this. (Same for CPP_LOBJ.)
2537
2538 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2539 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2540 necessary on certain architectures.
2541
2542 4. Added a setting of -export-symbols-regex to the link command to remove
2543 those symbols that are exported in the C sense, but actually are local
2544 within the library, and not documented. Their names all begin with
2545 "_pcre_". This is not a perfect job, because (a) we have to except some
2546 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2547 available (and never for static libraries). I have made a note to try to
2548 find a way round (a) in the future.
2549
2550
2551 Version 6.2 01-Aug-05
2552 ---------------------
2553
2554 1. There was no test for integer overflow of quantifier values. A construction
2555 such as {1111111111111111} would give undefined results. What is worse, if
2556 a minimum quantifier for a parenthesized subpattern overflowed and became
2557 negative, the calculation of the memory size went wrong. This could have
2558 led to memory overwriting.
2559
2560 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2561
2562 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2563 operating environments where this matters.
2564
2565 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2566 PCRE options from within the C++ wrapper.
2567
2568 5. Named capturing subpatterns were not being correctly counted when a pattern
2569 was compiled. This caused two problems: (a) If there were more than 100
2570 such subpatterns, the calculation of the memory needed for the whole
2571 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2572 back references of the form \12, where the number was greater than 9, were
2573 not recognized as back references, even though there were sufficient
2574 previous subpatterns.
2575
2576 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2577 versions of gcc, e.g. 2.95.4.
2578
2579
2580 Version 6.1 21-Jun-05
2581 ---------------------
2582
2583 1. There was one reference to the variable "posix" in pcretest.c that was not
2584 surrounded by "#if !defined NOPOSIX".
2585
2586 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2587 the cross-check on the old pcre_info() function, for the benefit of the
2588 cut-down version of PCRE that is currently imported into Exim.
2589
2590 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2591 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2592 patterns that start (?i) but I don't think that matters. The use of (?i) is
2593 just an example; this all applies to the other options as well.
2594
2595 4. Since libtool seems to echo the compile commands it is issuing, the output
2596 from "make" can be reduced a bit by putting "@" in front of each libtool
2597 compile command.
2598
2599 5. Patch from the folks at Google for configure.in to be a bit more thorough
2600 in checking for a suitable C++ installation before trying to compile the
2601 C++ stuff. This should fix a reported problem when a compiler was present,
2602 but no suitable headers.
2603
2604 6. The man pages all had just "PCRE" as their title. I have changed them to
2605 be the relevant file name. I have also arranged that these names are
2606 retained in the file doc/pcre.txt, which is a concatenation in text format
2607 of all the man pages except the little individual ones for each function.
2608
2609 7. The NON-UNIX-USE file had not been updated for the different set of source
2610 files that come with release 6. I also added a few comments about the C++
2611 wrapper.
2612
2613
2614 Version 6.0 07-Jun-05
2615 ---------------------
2616
2617 1. Some minor internal re-organization to help with my DFA experiments.
2618
2619 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2620 didn't matter for the library itself when fully configured, but did matter
2621 when compiling without UCP support, or within Exim, where the ucp files are
2622 not imported.
2623
2624 3. Refactoring of the library code to split up the various functions into
2625 different source modules. The addition of the new DFA matching code (see
2626 below) to a single monolithic source would have made it really too
2627 unwieldy, quite apart from causing all the code to be include in a
2628 statically linked application, when only some functions are used. This is
2629 relevant even without the DFA addition now that patterns can be compiled in
2630 one application and matched in another.
2631
2632 The downside of splitting up is that there have to be some external
2633 functions and data tables that are used internally in different modules of
2634 the library but which are not part of the API. These have all had their
2635 names changed to start with "_pcre_" so that they are unlikely to clash
2636 with other external names.
2637
2638 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2639 a different (DFA) algorithm. Although it is slower than the original
2640 function, it does have some advantages for certain types of matching
2641 problem.
2642
2643 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2644 including restarting after a partial match.
2645
2646 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2647 defined when compiling for Windows was sent to me. I have put it into the
2648 code, though I have no means of testing or verifying it.
2649
2650 7. Added the pcre_refcount() auxiliary function.
2651
2652 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2653 match before or at the first newline in the subject string. In pcretest,
2654 the /f option on a pattern can be used to set this.
2655
2656 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2657 would behave wrongly. This has been present in PCRE since release 4.0.
2658
2659 10. A number of changes to the pcregrep command:
2660
2661 (a) Refactored how -x works; insert ^(...)$ instead of setting
2662 PCRE_ANCHORED and checking the length, in preparation for adding
2663 something similar for -w.
2664
2665 (b) Added the -w (match as a word) option.
2666
2667 (c) Refactored the way lines are read and buffered so as to have more
2668 than one at a time available.
2669
2670 (d) Implemented a pcregrep test script.
2671
2672 (e) Added the -M (multiline match) option. This allows patterns to match
2673 over several lines of the subject. The buffering ensures that at least
2674 8K, or the rest of the document (whichever is the shorter) is available
2675 for matching (and similarly the previous 8K for lookbehind assertions).
2676
2677 (f) Changed the --help output so that it now says
2678
2679 -w, --word-regex(p)
2680
2681 instead of two lines, one with "regex" and the other with "regexp"
2682 because that confused at least one person since the short forms are the
2683 same. (This required a bit of code, as the output is generated
2684 automatically from a table. It wasn't just a text change.)
2685
2686 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2687 option but starts with a hyphen. Could be a pattern or a path name
2688 starting with a hyphen, for instance.
2689
2690 (h) "-" can be given as a file name to represent stdin.
2691
2692 (i) When file names are being printed, "(standard input)" is used for
2693 the standard input, for compatibility with GNU grep. Previously
2694 "<stdin>" was used.
2695
2696 (j) The option --label=xxx can be used to supply a name to be used for
2697 stdin when file names are being printed. There is no short form.
2698
2699 (k) Re-factored the options decoding logic because we are going to add
2700 two more options that take data. Such options can now be given in four
2701 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2702
2703 (l) Added the -A, -B, and -C options for requesting that lines of context
2704 around matches be printed.
2705
2706 (m) Added the -L option to print the names of files that do not contain
2707 any matching lines, that is, the complement of -l.
2708
2709 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2710 continue to scan other files.
2711
2712 (o) The -s option was incorrectly implemented. For compatibility with other
2713 greps, it now suppresses the error message for a non-existent or non-
2714 accessible file (but not the return code). There is a new option called
2715 -q that suppresses the output of matching lines, which was what -s was
2716 previously doing.
2717
2718 (p) Added --include and --exclude options to specify files for inclusion
2719 and exclusion when recursing.
2720
2721 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2722 Hopefully, it now does.
2723
2724 12. Missing cast in pcre_study().
2725
2726 13. Added an "uninstall" target to the makefile.
2727
2728 14. Replaced "extern" in the function prototypes in Makefile.in with
2729 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2730 world, but is set differently for Windows.
2731
2732 15. Added a second compiling function called pcre_compile2(). The only
2733 difference is that it has an extra argument, which is a pointer to an
2734 integer error code. When there is a compile-time failure, this is set
2735 non-zero, in addition to the error test pointer being set to point to an
2736 error message. The new argument may be NULL if no error number is required
2737 (but then you may as well call pcre_compile(), which is now just a
2738 wrapper). This facility is provided because some applications need a
2739 numeric error indication, but it has also enabled me to tidy up the way
2740 compile-time errors are handled in the POSIX wrapper.
2741
2742 16. Added VPATH=.libs to the makefile; this should help when building with one
2743 prefix path and installing with another. (Or so I'm told by someone who
2744 knows more about this stuff than I do.)
2745
2746 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2747 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2748 match everything, including newlines. This is not POSIX-compatible, but
2749 somebody wanted the feature. From pcretest it can be activated by using
2750 both the P and the s flags.
2751
2752 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2753
2754 19. libpcre.pc was being incorrectly installed as executable.
2755
2756 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2757 it now also looks for '\r' so that it will work unmodified on Windows.
2758
2759 21. Added Google's contributed C++ wrapper to the distribution.
2760
2761 22. Added some untidy missing memory free() calls in pcretest, to keep
2762 Electric Fence happy when testing.
2763
2764
2765
2766 Version 5.0 13-Sep-04
2767 ---------------------
2768
2769 1. Internal change: literal characters are no longer packed up into items
2770 containing multiple characters in a single byte-string. Each character
2771 is now matched using a separate opcode. However, there may be more than one
2772 byte in the character in UTF-8 mode.
2773
2774 2. The pcre_callout_block structure has two new fields: pattern_position and
2775 next_item_length. These contain the offset in the pattern to the next match
2776 item, and its length, respectively.
2777
2778 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2779 insertion of callouts before each pattern item. Added the /C option to
2780 pcretest to make use of this.
2781
2782 4. On the advice of a Windows user, the lines
2783
2784 #if defined(_WIN32) || defined(WIN32)
2785 _setmode( _fileno( stdout ), 0x8000 );
2786 #endif /* defined(_WIN32) || defined(WIN32) */
2787
2788 have been added to the source of pcretest. This apparently does useful
2789 magic in relation to line terminators.
2790
2791 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2792 for the benefit of those environments where the "b" makes a difference.
2793
2794 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2795 to know about it. I have put a hack into configure.in that adds in code
2796 to set GCC=yes if CC=icc. This seems to end up at a point in the
2797 generated configure script that is early enough to affect the setting of
2798 compiler options, which is what is needed, but I have no means of testing
2799 whether it really works. (The user who reported this had patched the
2800 generated configure script, which of course I cannot do.)
2801
2802 LATER: After change 22 below (new libtool files), the configure script
2803 seems to know about icc (and also ecc). Therefore, I have commented out
2804 this hack in configure.in.
2805
2806 7. Added support for pkg-config (2 patches were sent in).
2807
2808 8. Negated POSIX character classes that used a combination of internal tables
2809 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2810 [[:^ascii]]. Typically, they would match almost any characters. The other
2811 POSIX classes were not broken in this way.
2812
2813 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2814 to find the match, as PCRE was deluded into thinking that the match had to
2815 start at the start point or following a newline. The same bug applied to
2816 patterns with negative forward assertions or any backward assertions
2817 preceding ".*" at the start, unless the pattern required a fixed first
2818 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2819
2820 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2821 starting at the last subject character, bytes beyond the end of the subject
2822 string were read.
2823
2824 11. Renamed the variable "class" as "classbits" to make life easier for C++
2825 users. (Previously there was a macro definition, but it apparently wasn't
2826 enough.)
2827
2828 12. Added the new field "tables" to the extra data so that tables can be passed
2829 in at exec time, or the internal tables can be re-selected. This allows
2830 a compiled regex to be saved and re-used at a later time by a different
2831 program that might have everything at different addresses.
2832
2833 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2834 -R library as well as a -L library.
2835
2836 14. The debugging options of pcretest (-d on the command line or D on a
2837 pattern) showed incorrect output for anything following an extended class
2838 that contained multibyte characters and which was followed by a quantifier.
2839
2840 15. Added optional support for general category Unicode character properties
2841 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2842 support. It adds about 90K to the size of the library. The meanings of the
2843 inbuilt class escapes such as \d and \s have NOT been changed.
2844
2845 16. Updated pcredemo.c to include calls to free() to release the memory for the
2846 compiled pattern.
2847
2848 17. The generated file chartables.c was being created in the source directory
2849 instead of in the building directory. This caused the build to fail if the
2850 source directory was different from the building directory, and was
2851 read-only.
2852
2853 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2854 file. No doubt somebody will tell me if they don't make sense... Also added
2855 Dan Mooney's comments about building on OpenVMS.
2856
2857 19. Added support for partial matching via the PCRE_PARTIAL option for
2858 pcre_exec() and the \P data escape in pcretest.
2859
2860 20. Extended pcretest with 3 new pattern features:
2861
2862 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2863 write the compiled pattern to the file whose name is "rest-of-line".
2864 This is a straight binary dump of the data, with the saved pointer to
2865 the character tables forced to be NULL. The study data, if any, is
2866 written too. After writing, pcretest reads a new pattern.
2867
2868 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2869 compiled pattern from the given file. There must not be any
2870 occurrences of "<" in the file name (pretty unlikely); if there are,
2871 pcretest will instead treat the initial "<" as a pattern delimiter.
2872 After reading in the pattern, pcretest goes on to read data lines as
2873 usual.
2874
2875 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2876 and 16-bit fields in a compiled pattern, to simulate a pattern that
2877 was compiled on a host of opposite endianness.
2878
2879 21. The pcre-exec() function can now cope with patterns that were compiled on
2880 hosts of opposite endianness, with this restriction:
2881
2882 As for any compiled expression that is saved and used later, the tables
2883 pointer field cannot be preserved; the extra_data field in the arguments
2884 to pcre_exec() should be used to pass in a tables address if a value
2885 other than the default internal tables were used at compile time.
2886
2887 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2888 now diagnosed as an error. Previously, most of the time, a negative number
2889 would have been treated as zero, but if in addition "ovector" was passed as
2890 NULL, a crash could occur.
2891
2892 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2893 new versions from the libtool 1.5 distribution (the last one is a copy of
2894 a file called libtool.m4). This seems to have fixed the need to patch
2895 "configure" to support Darwin 1.3 (which I used to do). However, I still
2896 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2897 workstation).
2898
2899 24. Changed the PCRE licence to be the more standard "BSD" licence.
2900
2901
2902 Version 4.5 01-Dec-03
2903 ---------------------
2904
2905 1. There has been some re-arrangement of the code for the match() function so
2906 that it can be compiled in a version that does not call itself recursively.
2907 Instead, it keeps those local variables that need separate instances for
2908 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2909 needs to "recurse". Keeping track of where control must go is done by means
2910 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2911 hide most of the details from the main code, and operates only if
2912 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2913 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2914 operating.
2915
2916 To make it easier for callers to provide specially tailored get/free
2917 functions for this usage, two new functions, pcre_stack_malloc, and
2918 pcre_stack_free, are used. They are always called in strict stacking order,
2919 and the size of block requested is always the same.
2920
2921 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2922 PCRE has been compiled to use the stack or the heap for recursion. The
2923 -C option of pcretest uses this to show which version is compiled.
2924
2925 A new data escape \S, is added to pcretest; it causes the amounts of store
2926 obtained and freed by both kinds of malloc/free at match time to be added
2927 to the output.
2928
2929 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2930 what's available on my current Linux desktop machine.
2931
2932 3. When matching a UTF-8 string, the test for a valid string at the start has
2933 been extended. If start_offset is not zero, PCRE now checks that it points
2934 to a byte that is the start of a UTF-8 character. If not, it returns
2935 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2936 this is necessary because there may be backward assertions in the pattern.
2937 When matching the same subject several times, it may save resources to use
2938 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2939
2940 4. The code for checking the validity of UTF-8 strings has been tightened so
2941 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2942 containing "overlong sequences".
2943
2944 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2945 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2946 should have been "&", but it just so happened that all the cases this let
2947 through by mistake were picked up later in the function.
2948
2949 6. I had used a variable called "isblank" - this is a C99 function, causing
2950 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2951
2952 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2953 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2954 so the version is shown; (c) stop "make test" echoing "./RunTest".
2955
2956 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2957
2958 9. The prototype for memmove() for systems that don't have it was using
2959 size_t, but the inclusion of the header that defines size_t was later. I've
2960 moved the #includes for the C headers earlier to avoid this.
2961
2962 10. Added some adjustments to the code to make it easier to compiler on certain
2963 special systems:
2964
2965 (a) Some "const" qualifiers were missing.
2966 (b) Added the macro EXPORT before all exported functions; by default this
2967 is defined to be empty.
2968 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2969 that it reads its output file name as an argument instead of writing
2970 to the standard output and assuming this can be redirected.
2971
2972 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2973 class containing characters with values greater than 255, PCRE compilation
2974 went into a loop.
2975
2976 12. A recursive reference to a subpattern that was within another subpattern
2977 that had a minimum quantifier of zero caused PCRE to crash. For example,
2978 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2979 recursion. If the recursively-called subpattern itself had a zero repeat,
2980 that was OK.
2981
2982 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2983 buffer into which it was copied (for escape processing) was still set at
2984 1024, so long lines caused crashes.
2985
2986 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2987 "internal error: code overflow...". This applied to any character class
2988 that was followed by a possessive quantifier.
2989
2990 15. Modified the Makefile to add libpcre.la as a prerequisite for
2991 libpcreposix.la because I was told this is needed for a parallel build to
2992 work.
2993
2994 16. If a pattern that contained .* following optional items at the start was
2995 studied, the wrong optimizing data was generated, leading to matching
2996 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2997 matching string must start with a or b or c. The correct conclusion for
2998 this pattern is that a match can start with any character.
2999
3000
3001 Version 4.4 13-Aug-03
3002 ---------------------
3003
3004 1. In UTF-8 mode, a character class containing characters with values between
3005 127 and 255 was not handled correctly if the compiled pattern was studied.
3006 In fixing this, I have also improved the studying algorithm for such
3007 classes (slightly).
3008
3009 2. Three internal functions had redundant arguments passed to them. Removal
3010 might give a very teeny performance improvement.
3011
3012 3. Documentation bug: the value of the capture_top field in a callout is *one
3013 more than* the number of the hightest numbered captured substring.
3014
3015 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3016 in incorrectly linking with a previously installed version. They now link
3017 explicitly with libpcre.la.
3018
3019 5. configure.in no longer needs to recognize Cygwin specially.
3020
3021 6. A problem in pcre.in for Windows platforms is fixed.
3022
3023 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3024 pcretest, it used to include the size of the study block as part of its
3025 output. Unfortunately, the structure contains a field that has a different
3026 size on different hardware architectures. This meant that the tests that
3027 showed this size failed. As the block is currently always of a fixed size,
3028 this information isn't actually particularly useful in pcretest output, so
3029 I have just removed it.
3030
3031 8. Three pre-processor statements accidentally did not start in column 1.
3032 Sadly, there are *still* compilers around that complain, even though
3033 standard C has not required this for well over a decade. Sigh.
3034
3035 9. In pcretest, the code for checking callouts passed small integers in the
3036 callout_data field, which is a void * field. However, some picky compilers
3037 complained about the casts involved for this on 64-bit systems. Now
3038 pcretest passes the address of the small integer instead, which should get
3039 rid of the warnings.
3040
3041 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3042 both compile and run time, and gives an error if an invalid UTF-8 sequence
3043 is found. There is a option for disabling this check in cases where the
3044 string is known to be correct and/or the maximum performance is wanted.
3045
3046 11. In response to a bug report, I changed one line in Makefile.in from
3047
3048 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3049 to
3050 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3051
3052 to look similar to other lines, but I have no way of telling whether this
3053 is the right thing to do, as I do not use Windows. No doubt I'll get told
3054 if it's wrong...
3055
3056
3057 Version 4.3 21-May-03
3058 ---------------------
3059
3060 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3061 Makefile.
3062
3063 2. Some refactoring to improve the quality of the code:
3064
3065 (i) The utf8_table... variables are now declared "const".
3066
3067 (ii) The code for \cx, which used the "case flipping" table to upper case
3068 lower case letters, now just substracts 32. This is ASCII-specific,
3069 but the whole concept of \cx is ASCII-specific, so it seems
3070 reasonable.
3071
3072 (iii) PCRE was using its character types table to recognize decimal and
3073 hexadecimal digits in the pattern. This is silly, because it handles
3074 only 0-9, a-f, and A-F, but the character types table is locale-
3075 specific, which means strange things might happen. A private
3076 table is now used for this - though it costs 256 bytes, a table is
3077 much faster than multiple explicit tests. Of course, the standard
3078 character types table is still used for matching digits in subject
3079 strings against \d.
3080
3081 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3082 ending in _t are). So I've renamed it as ESC_tee.
3083
3084 3. The first argument for regexec() in the POSIX wrapper should have been
3085 defined as "const".
3086
3087 4. Changed pcretest to use malloc() for its buffers so that they can be
3088 Electric Fenced for debugging.
3089
3090 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3091 to read one or more bytes before the start of the subject string. Often this
3092 had no effect on PCRE's behaviour, but in some circumstances it could
3093 provoke a segmentation fault.
3094
3095 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3096 to try to read one or more bytes before the start of the subject string.
3097
3098 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3099 UTF-8 support could misbehave in various ways if the subject string
3100 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3101 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3102 back over UTF-8 characters.)
3103
3104
3105 Version 4.2 14-Apr-03
3106 ---------------------
3107
3108 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3109
3110 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3111 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3112 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3113 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3114 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3115 and BUILD_EXEEXT
3116 Note: automatic setting of the BUILD variables is not yet working
3117 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3118 compile-time but not at link-time
3119 [LINK]: use for linking executables only
3120 make different versions for Windows and non-Windows
3121 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3122 libraries
3123 [LINK_FOR_BUILD]: new variable
3124 [OBJEXT]: use throughout
3125 [EXEEXT]: use throughout
3126 <winshared>: new target
3127 <wininstall>: new target
3128 <dftables.o>: use native compiler
3129 <dftables>: use native linker
3130 <install>: handle Windows platform correctly
3131 <clean>: ditto
3132 <check>: ditto
3133 copy DLL to top builddir before testing
3134
3135 As part of these changes, -no-undefined was removed again. This was reported
3136 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3137 in any case.
3138
3139 3. Some tidies to get rid of compiler warnings:
3140
3141 . In the match_data structure, match_limit was an unsigned long int, whereas
3142 match_call_count was an int. I've made them both unsigned long ints.
3143
3144 . In pcretest the fact that a const uschar * doesn't automatically cast to
3145 a void * provoked a warning.
3146
3147 . Turning on some more compiler warnings threw up some "shadow" variables
3148 and a few more missing casts.
3149
3150 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3151 option, a class that contained a single character with a value between 128
3152 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3153
3154 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3155 option, a class that contained several characters, but with at least one
3156 whose value was between 128 and 255 caused PCRE to crash.
3157
3158
3159 Version 4.1 12-Mar-03
3160 ---------------------
3161
3162 1. Compiling with gcc -pedantic found a couple of places where casts were
3163 needed, and a string in dftables.c that was longer than standard compilers are
3164 required to support.
3165
3166 2. Compiling with Sun's compiler found a few more places where the code could
3167 be tidied up in order to avoid warnings.
3168
3169 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3170 first of these names is deprecated in the latest Autoconf in favour of the name
3171 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3172 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3173 analogy I have changed it to CFLAGS_FOR_BUILD.
3174
3175 4. Added -no-undefined to the linking command in the Makefile, because this is
3176 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3177 linking step for the pcreposix library.
3178
3179 5. PCRE was failing to diagnose the case of two named groups with the same
3180 name.
3181
3182 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3183 literal character that is needed in the subject for a match, and scans along to
3184 ensure that it is present before embarking on the full matching process. This
3185 saves time in cases of nested unlimited repeats that are never going to match.
3186 Problem: the scan can take a lot of time if the subject is very long (e.g.
3187 megabytes), thus penalizing straightforward matches. It is now done only if the
3188 amount of subject to be scanned is less than 1000 bytes.
3189
3190 7. A lesser problem with the same optimization is that it was recording the
3191 first character of an anchored pattern as "needed", thus provoking a search
3192 right along the subject, even when the first match of the pattern was going to
3193 fail. The "needed" character is now not set for anchored patterns, unless it
3194 follows something in the pattern that is of non-fixed length. Thus, it still
3195 fulfils its original purpose of finding quick non-matches in cases of nested
3196 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3197
3198
3199 Version 4.0 17-Feb-03
3200 ---------------------
3201
3202 1. If a comment in an extended regex that started immediately after a meta-item
3203 extended to the end of string, PCRE compiled incorrect data. This could lead to
3204 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3205
3206 2. Moved to autoconf 2.53 and libtool 1.4.2.
3207
3208 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3209 the special perltest8 script is no longer needed - all the tests can be run
3210 from a single perltest script.
3211
3212 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3213 by \s. It has now been removed in PCRE. This means it isn't recognized as
3214 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3215 class [:space:] *does* include VT, thereby creating a mess.
3216
3217 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3218 space and tab.
3219
3220 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3221 its new features into the main test script, reducing the number of scripts.
3222
3223 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3224 were backward compatible, and made the (?i) apply to the whole pattern, as if
3225 /i were given. Now it behaves more logically, and applies the option setting
3226 only to what follows. PCRE has been changed to follow suit. However, if it
3227 finds options settings right at the start of the pattern, it extracts them into
3228 the global options, as before. Thus, they show up in the info data.
3229
3230 8. Added support for the \Q...\E escape sequence. Characters in between are
3231 treated as literals. This is slightly different from Perl in that $ and @ are
3232 also handled as literals inside the quotes. In Perl, they will cause variable
3233 interpolation. Note the following examples:
3234
3235 Pattern PCRE matches Perl matches
3236
3237 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3238 \Qabc\$xyz\E abc\$xyz abc\$xyz
3239 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3240
3241 For compatibility with Perl, \Q...\E sequences are recognized inside character
3242 classes as well as outside them.
3243
3244 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3245 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3246 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3247 signed/unsigned warnings.
3248
3249 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3250 option for pcretest, so I've replaced it by a simple function that does just
3251 that job.
3252
3253 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3254 "pcregrep -".
3255
3256 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3257 Java package. This provides some syntactic sugar for simple cases of what my
3258 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3259 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3260 item, you can use this simplified notation. Note that only makes sense with
3261 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3262 greediness, whatever the setting of the PCRE_UNGREEDY option.
3263
3264 13. A change of greediness default within a pattern was not taking effect at
3265 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3266 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3267 was abstracted outside.
3268
3269 14. PCRE now supports the \G assertion. It is true when the current matching
3270 position is at the start point of the match. This differs from \A when the
3271 starting offset is non-zero. Used with the /g option of pcretest (or similar
3272 code), it works in the same way as it does for Perl's /g option. If all
3273 alternatives of a regex begin with \G, the expression is anchored to the start
3274 match position, and the "anchored" flag is set in the compiled expression.
3275
3276 15. Some bugs concerning the handling of certain option changes within patterns
3277 have been fixed. These applied to options other than (?ims). For example,
3278 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3279 been the other way round. Some of this was related to change 7 above.
3280
3281 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3282 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3283 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3284 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3285
3286 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3287 mode. Unlike ".", it always matches newline, whatever the setting of
3288 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3289 assertions. Perl allows it, but it doesn't (in general) work because it can't
3290 calculate the length of the lookbehind. At least, that's the case for Perl
3291 5.8.0 - I've been told they are going to document that it doesn't work in
3292 future.
3293
3294 18. Added an error diagnosis for escapes that PCRE does not support: these are
3295 \L, \l, \N, \P, \p, \U, \u, and \X.
3296
3297 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3298 reading past the end of the pattern in cases such as /[abcd/.
3299
3300 20. PCRE was getting more memory than necessary for patterns with classes that
3301 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3302
3303 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3304 compiling PCRE for use with Virtual Pascal.
3305
3306 22. Small fix to the Makefile to make it work properly if the build is done
3307 outside the source tree.
3308
3309 23. Added a new extension: a condition to go with recursion. If a conditional
3310 subpattern starts with (?(R) the "true" branch is used if recursion has
3311 happened, whereas the "false" branch is used only at the top level.
3312
3313 24. When there was a very long string of literal characters (over 255 bytes
3314 without UTF support, over 250 bytes with UTF support), the computation of how
3315 much memory was required could be incorrect, leading to segfaults or other
3316 strange effects.
3317
3318 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3319 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3320 there was a subsequent back reference to those brackets. This meant that, for
3321 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3322 possible to check for precisely this case. All we can do is abandon the
3323 optimization if .* occurs inside capturing brackets when there are any back
3324 references whatsoever. (See below for a better fix that came later.)
3325
3326 26. The handling of the optimization for finding the first character of a
3327 non-anchored pattern, and for finding a character that is required later in the
3328 match were failing in some cases. This didn't break the matching; it just
3329 failed to optimize when it could. The way this is done has been re-implemented.
3330
3331 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3332
3333 28. Added a new feature that provides some of the functionality that Perl
3334 provides with (?{...}). The facility is termed a "callout". The way it is done
3335 in PCRE is for the caller to provide an optional function, by setting
3336 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3337 global variable. By default it is unset, which disables all calling out. To get
3338 the function called, the regex must include (?C) at appropriate points. This
3339 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3340 This provides a means of identifying different callout points. When PCRE
3341 reaches such a point in the regex, if pcre_callout has been set, the external
3342 function is called. It is provided with data in a structure called
3343 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3344 matching continues; if it returns a non-zero value, the match at the current
3345 point fails. However, backtracking will occur if possible. [This was changed
3346 later and other features added - see item 49 below.]
3347
3348 29. pcretest is upgraded to test the callout functionality. It provides a
3349 callout function that displays information. By default, it shows the start of
3350 the match and the current position in the text. There are some new data escapes
3351 to vary what happens:
3352
3353 \C+ in addition, show current contents of captured substrings
3354 \C- do not supply a callout function
3355 \C!n return 1 when callout number n is reached
3356 \C!n!m return 1 when callout number n is reached for the mth time
3357
3358 30. If pcregrep was called with the -l option and just a single file name, it
3359 output "<stdin>" if a match was found, instead of the file name.
3360
3361 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3362 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3363 pcre_exec(). This saves a malloc/free per call. The default value of
3364 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3365 when configuring.
3366
3367 32. The default maximum size of a compiled pattern is 64K. There have been a
3368 few cases of people hitting this limit. The code now uses macros to handle the
3369 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3370 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3371 configuring. Tests 2 and 5 work only with 2-byte links because they output
3372 debugging information about compiled patterns.
3373
3374 33. Internal code re-arrangements:
3375
3376 (a) Moved the debugging function for printing out a compiled regex into
3377 its own source file (printint.c) and used #include to pull it into
3378 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3379 separate copies.
3380
3381 (b) Defined the list of op-code names for debugging as a macro in
3382 internal.h so that it is next to the definition of the opcodes.
3383
3384 (c) Defined a table of op-code lengths for simpler skipping along compiled
3385 code. This is again a macro in internal.h so that it is next to the
3386 definition of the opcodes.
3387
3388 34. Added support for recursive calls to individual subpatterns, along the
3389 lines of Robin Houston's patch (but implemented somewhat differently).
3390
3391 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3392 allow it to read and process whole directories in Win32. This code was
3393 contributed by Lionel Fourquaux; it has not been tested by me.
3394
3395 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3396 used to name a group. Names consist of alphanumerics and underscores, and must
3397 be unique. Back references use the syntax (?P=name) and recursive calls use
3398 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3399 numbers. The function pcre_fullinfo() can be used after compilation to extract
3400 a name/number map. There are three relevant calls:
3401
3402 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3403 PCRE_INFO_NAMECOUNT yields the number of entries
3404 PCRE_INFO_NAMETABLE yields a pointer to the map.
3405
3406 The map is a vector of fixed-size entries. The size of each entry depends on
3407 the length of the longest name used. The first two bytes of each entry are the
3408 group number, most significant byte first. There follows the corresponding
3409 name, zero terminated. The names are in alphabetical order.
3410
3411 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3412 case instead of 255. Making it the same both with and without UTF-8 support
3413 means that the same test output works with both.
3414
3415 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3416 calling malloc() with a zero argument.
3417
3418 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3419 optimization. I've improved things by keeping a bitmap of backreferences with
3420 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3421 fact referenced, the optimization can be applied. It is unlikely that a
3422 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3423 the match to follow \n) will appear inside brackets with a number greater than
3424 31, but if it does, any back reference > 31 suppresses the optimization.
3425
3426 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3427 of disabling numbered capturing parentheses. Any opening parenthesis that is
3428 not followed by ? behaves as if it were followed by ?: but named parentheses
3429 can still be used for capturing (and they will acquire numbers in the usual
3430 way).
3431
3432 41. Redesigned the return codes from the match() function into yes/no/error so
3433 that errors can be passed back from deep inside the nested calls. A malloc
3434 failure while inside a recursive subpattern call now causes the
3435 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3436
3437 42. It is now possible to set a limit on the number of times the match()
3438 function is called in a call to pcre_exec(). This facility makes it possible to
3439 limit the amount of recursion and backtracking, though not in a directly
3440 obvious way, because the match() function is used in a number of different
3441 circumstances. The count starts from zero for each position in the subject
3442 string (for non-anchored patterns). The default limit is, for compatibility, a
3443 large number, namely 10 000 000. You can change this in two ways:
3444
3445 (a) When configuring PCRE before making, you can use --with-match-limit=n
3446 to set a default value for the compiled library.
3447
3448 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3449 a different value is set. See 45 below.
3450
3451 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3452
3453 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3454 of things that can be changed at compile time. The first argument specifies
3455 what is wanted and the second points to where the information is to be placed.
3456 The current list of available information is:
3457
3458 PCRE_CONFIG_UTF8
3459
3460 The output is an integer that is set to one if UTF-8 support is available;
3461 otherwise it is set to zero.
3462
3463 PCRE_CONFIG_NEWLINE
3464
3465 The output is an integer that it set to the value of the code that is used for
3466 newline. It is either LF (10) or CR (13).
3467
3468 PCRE_CONFIG_LINK_SIZE
3469
3470 The output is an integer that contains the number of bytes used for internal
3471 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3472
3473 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3474
3475 The output is an integer that contains the threshold above which the POSIX
3476 interface uses malloc() for output vectors. See item 31 above.
3477
3478 PCRE_CONFIG_MATCH_LIMIT
3479
3480 The output is an unsigned integer that contains the default limit of the number
3481 of match() calls in a pcre_exec() execution. See 42 above.
3482
3483 44. pcretest has been upgraded by the addition of the -C option. This causes it
3484 to extract all the available output from the new pcre_config() function, and to
3485 output it. The program then exits immediately.
3486
3487 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3488 order to support additional features. One way would have been to define
3489 pcre_exec2() (for example) with extra arguments, but this would not have been
3490 extensible, and would also have required all calls to the original function to
3491 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3492 is used for passing in "extra" data from pcre_study().
3493
3494 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3495 contains the following fields:
3496
3497 flags a bitmap indicating which of the following fields are set
3498 study_data opaque data from pcre_study()
3499 match_limit a way of specifying a limit on match() calls for a specific
3500 call to pcre_exec()
3501 callout_data data for callouts (see 49 below)
3502
3503 The flag bits are also defined in pcre.h, and are
3504
3505 PCRE_EXTRA_STUDY_DATA
3506 PCRE_EXTRA_MATCH_LIMIT
3507 PCRE_EXTRA_CALLOUT_DATA
3508
3509 The pcre_study() function now returns one of these new pcre_extra blocks, with
3510 the actual study data pointed to by the study_data field, and the
3511 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3512 before. That is, this change is entirely upwards-compatible and requires no
3513 change to existing code.
3514
3515 If you want to pass in additional data to pcre_exec(), you can either place it
3516 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3517 block.
3518
3519 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3520 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3521 times with different match limits, until it finds the minimum value needed for
3522 pcre_exec() to complete. The value is then output. This can be instructive; for
3523 most simple matches the number is quite small, but for pathological cases it
3524 gets very large very quickly.
3525
3526 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3527 returns the size of the data block pointed to by the study_data field in a
3528 pcre_extra block, that is, the value that was passed as the argument to
3529 pcre_malloc() when PCRE was getting memory in which to place the information
3530 created by pcre_study(). The fourth argument should point to a size_t variable.
3531 pcretest has been extended so that this information is shown after a successful
3532 pcre_study() call when information about the compiled regex is being displayed.
3533
3534 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3535 because what follows is always an absolute path. (Later: it turns out that this
3536 is more than cosmetic for MinGW, because it doesn't like empty path
3537 components.)
3538
3539 49. Some changes have been made to the callout feature (see 28 above):
3540
3541 (i) A callout function now has three choices for what it returns:
3542
3543 0 => success, carry on matching
3544 > 0 => failure at this point, but backtrack if possible
3545 < 0 => serious error, return this value from pcre_exec()
3546
3547 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3548 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3549 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3550 use by callout functions. It will never be used by PCRE itself.
3551
3552 (ii) The pcre_extra structure (see 45 above) has a void * field called
3553 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3554 pcre_callout_block structure has a field of the same name. The contents of
3555 the field passed in the pcre_extra structure are passed to the callout
3556 function in the corresponding field in the callout block. This makes it
3557 easier to use the same callout-containing regex from multiple threads. For
3558 testing, the pcretest program has a new data escape
3559
3560 \C*n pass the number n (may be negative) as callout_data
3561
3562 If the callout function in pcretest receives a non-zero value as
3563 callout_data, it returns that value.
3564
3565 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3566 there were some redundant $(CFLAGS) in commands that are now specified as
3567 $(LINK), which already includes $(CFLAGS).
3568
3569 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3570 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3571 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3572 one-byte characters throughout. Note that case-insensitive matching applies
3573 only to characters whose values are less than 256. PCRE doesn't support the
3574 notion of cases for higher-valued characters.
3575
3576 (i) A character class whose characters are all within 0-255 is handled as
3577 a bit map, and the map is inverted for negative classes. Previously, a
3578 character > 255 always failed to match such a class; however it should
3579 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3580
3581 (ii) A negated character class with a single character < 255 is coded as
3582 "not this character" (OP_NOT). This wasn't working properly when the test
3583 character was multibyte, either singly or repeated.
3584
3585 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3586 mode, for example: \x{100}{2,3}.
3587
3588 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3589 singly or repeated) now correctly test multibyte characters. However,
3590 PCRE doesn't recognize any characters with values greater than 255 as
3591 digits, spaces, or word characters. Such characters always match \D, \S,
3592 and \W, and never match \d, \s, or \w.
3593
3594 (v) Classes may now contain characters and character ranges with values
3595 greater than 255. For example: [ab\x{100}-\x{400}].
3596
3597 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3598 PCRE in UTF-8 mode.
3599
3600 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3601 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3602 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3603 value.)
3604
3605 53. The single man page has become too large. I have therefore split it up into
3606 a number of separate man pages. These also give rise to individual HTML pages;
3607 these are now put in a separate directory, and there is an index.html page that
3608 lists them all. Some hyperlinking between the pages has been installed.
3609
3610 54. Added convenience functions for handling named capturing parentheses.
3611
3612 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3613 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3614 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3615 are faulted.
3616
3617 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3618 calling configure. These values are used when compiling the dftables.c program
3619 which is run to generate the source of the default character tables. They
3620 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3621 you will need to set these values.
3622
3623 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3624
3625
3626 Version 3.9 02-Jan-02
3627 ---------------------
3628
3629 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3630
3631 2. If --disable-static was given, the building process failed when trying to
3632 build pcretest and pcregrep. (For some reason it was using libtool to compile
3633 them, which is not right, as they aren't part of the library.)
3634
3635
3636 Version 3.8 18-Dec-01
3637 ---------------------
3638
3639 1. The experimental UTF-8 code was completely screwed up. It was packing the
3640 bytes in the wrong order. How dumb can you get?
3641
3642
3643 Version 3.7 29-Oct-01
3644 ---------------------
3645
3646 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3647 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3648 this didn't happen under Solaris 8, where I normally test things.
3649
3650 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3651 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3652 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3653 this fix an hour or so after the initial 3.7 release.)
3654
3655
3656 Version 3.6 23-Oct-01
3657 ---------------------
3658
3659 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3660 offsets passed as NULL with zero offset count.
3661
3662 2. The config.guess and config.sub files had not been updated when I moved to
3663 the latest autoconf.
3664
3665
3666 Version 3.5 15-Aug-01
3667 ---------------------
3668
3669 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3670 had been forgotten.
3671
3672 2. By using declared but undefined structures, we can avoid using "void"
3673 definitions in pcre.h while keeping the internal definitions of the structures
3674 private.
3675
3676 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3677 user point of view, this means that both static and shared libraries are built
3678 by default, but this can be individually controlled. More of the work of
3679 handling this static/shared cases is now inside libtool instead of PCRE's make
3680 file.
3681
3682 4. The pcretest utility is now installed along with pcregrep because it is
3683 useful for users (to test regexs) and by doing this, it automatically gets
3684 relinked by libtool. The documentation has been turned into a man page, so
3685 there are now .1, .txt, and .html versions in /doc.
3686
3687 5. Upgrades to pcregrep:
3688 (i) Added long-form option names like gnu grep.
3689 (ii) Added --help to list all options with an explanatory phrase.
3690 (iii) Added -r, --recursive to recurse into sub-directories.
3691 (iv) Added -f, --file to read patterns from a file.
3692
3693 6. pcre_exec() was referring to its "code" argument before testing that
3694 argument for NULL (and giving an error if it was NULL).
3695
3696 7. Upgraded Makefile.in to allow for compiling in a different directory from
3697 the source directory.
3698
3699 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3700 options bits, the pointer it was passed was to an int instead of to an unsigned
3701 long int. This mattered only on 64-bit systems.
3702
3703 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3704 generated) instead of pcre.in, which it its source. Also made the same change
3705 in several of the .c files.
3706
3707 10. A new release of gcc defines printf() as a macro, which broke pcretest
3708 because it had an ifdef in the middle of a string argument for printf(). Fixed
3709 by using separate calls to printf().
3710
3711 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3712 script, to force use of CR or LF instead of \n in the source. On non-Unix
3713 systems, the value can be set in config.h.
3714
3715 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3716 absolute limit. Changed the text of the error message to make this clear, and
3717 likewise updated the man page.
3718
3719 13. The limit of 99 on the number of capturing subpatterns has been removed.
3720 The new limit is 65535, which I hope will not be a "real" limit.
3721
3722
3723 Version 3.4 22-Aug-00
3724 ---------------------
3725
3726 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3727
3728 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3729
3730
3731 Version 3.3 01-Aug-00
3732 ---------------------
3733
3734 1. If an octal character was given, but the value was greater than \377, it
3735 was not getting masked to the least significant bits, as documented. This could
3736 lead to crashes in some systems.
3737
3738 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3739 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3740
3741 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3742 These just pass their arguments on to (pcre_free)(), but they are provided
3743 because some uses of PCRE bind it to non-C systems that can call its functions,
3744 but cannot call free() or pcre_free() directly.
3745
3746 4. Add "make test" as a synonym for "make check". Corrected some comments in
3747 the Makefile.
3748
3749 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3750 Makefile.
3751
3752 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3753 command called pgrep for grepping around the active processes.
3754
3755 7. Added the beginnings of support for UTF-8 character strings.
3756
3757 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3758 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3759 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3760 out for the ar command.)
3761
3762
3763 Version 3.2 12-May-00
3764 ---------------------
3765
3766 This is purely a bug fixing release.
3767
3768 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3769 of ZA. This was just one example of several cases that could provoke this bug,
3770 which was introduced by change 9 of version 2.00. The code for breaking
3771 infinite loops after an iteration that matches an empty string was't working
3772 correctly.
3773
3774 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3775 when matched against abbab (for example). After matching an empty string, it
3776 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3777 caused it to match further down the string than it should.
3778
3779 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3780 was there because it doesn't seem to be needed, and it causes trouble on some
3781 systems, as it is not a Standard C header. It has been removed.
3782
3783 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3784 were reported on the Macintosh. The changes were from
3785
3786 while ((c = *(++ptr)) != 0 && c != '\n');
3787 to
3788 while ((c = *(++ptr)) != 0 && c != '\n') ;
3789
3790 Totally extraordinary, but if that's what it takes...
3791
3792 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3793 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3794 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3795 assumes the way PCRE uses memmove() (always moving upwards).
3796
3797 6. PCRE is being used in one environment where strchr() is not available. There
3798 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3799 faster code anyway.
3800
3801
3802 Version 3.1 09-Feb-00
3803 ---------------------
3804
3805 The only change in this release is the fixing of some bugs in Makefile.in for
3806 the "install" target:
3807
3808 (1) It was failing to install pcreposix.h.
3809
3810 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3811
3812
3813 Version 3.0 01-Feb-00
3814 ---------------------
3815
3816 1. Add support for the /+ modifier to perltest (to output $` like it does in
3817 pcretest).
3818
3819 2. Add support for the /g modifier to perltest.
3820
3821 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3822 matches null strings.
3823
3824 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3825 pattern. Perl treats empty patterns specially - it reuses the most recent
3826 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3827 effect.
3828
3829 5. The POSIX interface was broken in that it was just handing over the POSIX
3830 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3831 required a bigger vector, with some working space on the end. This means that
3832 the POSIX wrapper now has to get and free some memory, and copy the results.
3833
3834 6. Added some simple autoconf support, placing the test data and the
3835 documentation in separate directories, re-organizing some of the
3836 information files, and making it build pcre-config (a GNU standard). Also added
3837 libtool support for building PCRE as a shared library, which is now the
3838 default.
3839
3840 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3841 09 are not valid octal constants. Single digits will be used for minor values
3842 less than 10.
3843
3844 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3845 existing programs that set these in the POSIX interface can use PCRE without
3846 modification.
3847
3848 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3849 return all that pcre_info() returns, plus additional data. The pcre_info()
3850 function is retained for compatibility, but is considered to be obsolete.
3851
3852 10. Added experimental recursion feature (?R) to handle one common case that
3853 Perl 5.6 will be able to do with (?p{...}).
3854
3855 11. Added support for POSIX character classes like [:alpha:], which Perl is
3856 adopting.
3857
3858
3859 Version 2.08 31-Aug-99
3860 ----------------------
3861
3862 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3863 trying to match at the startoffset position, but instead was moving forward to
3864 the next newline as if a previous match had failed.
3865
3866 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3867 and could get into a loop if a null string was matched other than at the start
3868 of the subject.
3869
3870 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3871 be distinguished at compile time, and for completeness also added PCRE_DATE.
3872
3873 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3874 in GnuWin32 environments.
3875
3876
3877 Version 2.07 29-Jul-99
3878 ----------------------
3879
3880 1. The documentation is now supplied in plain text form and HTML as well as in
3881 the form of man page sources.
3882
3883 2. C++ compilers don't like assigning (void *) values to other pointer types.
3884 In particular this affects malloc(). Although there is no problem in Standard
3885 C, I've put in casts to keep C++ compilers happy.
3886
3887 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3888 should be (const char *).
3889
3890 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3891 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3892 However, I haven't made this a standard facility. The documentation doesn't
3893 mention it, and the Makefile doesn't support it.
3894
3895 5. The Makefile now contains an "install" target, with editable destinations at
3896 the top of the file. The pcretest program is not installed.
3897
3898 6. pgrep -V now gives the PCRE version number and date.
3899
3900 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3901 causing the entire string to be ignored, instead of just the last character.
3902
3903 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3904 non-matching string, it can take a very, very long time, even for strings of
3905 quite modest length, because of the nested recursion. PCRE now does better in
3906 some of these cases. It does this by remembering the last required literal
3907 character in the pattern, and pre-searching the subject to ensure it is present
3908 before running the real match. In other words, it applies a heuristic to detect
3909 some types of certain failure quickly, and in the above example, if presented
3910 with a string that has no trailing " it gives "no match" very quickly.
3911
3912 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3913 other alternatives are tried instead.
3914
3915
3916 Version 2.06 09-Jun-99
3917 ----------------------
3918
3919 1. Change pcretest's output for amount of store used to show just the code
3920 space, because the remainder (the data block) varies in size between 32-bit and
3921 64-bit systems.
3922
3923 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3924 start matching at. This allows lookbehinds to work when searching for multiple
3925 occurrences in a string.
3926
3927 3. Added additional options to pcretest for testing multiple occurrences:
3928
3929 /+ outputs the rest of the string that follows a match
3930 /g loops for multiple occurrences, using the new startoffset argument
3931 /G loops for multiple occurrences by passing an incremented pointer
3932
3933 4. PCRE wasn't doing the "first character" optimization for patterns starting
3934 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3935 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3936 the letter 'x'. On long subject strings, this gives a significant speed-up.
3937
3938
3939 Version 2.05 21-Apr-99
3940 ----------------------
3941
3942 1. Changed the type of magic_number from int to long int so that it works
3943 properly on 16-bit systems.
3944
3945 2. Fixed a bug which caused patterns starting with .* not to work correctly
3946 when the subject string contained newline characters. PCRE was assuming
3947 anchoring for such patterns in all cases, which is not correct because .* will
3948 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3949 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3950 must be retried after every newline in the subject.
3951
3952
3953 Version 2.04 18-Feb-99
3954 ----------------------
3955
3956 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3957 computation of the store needed to hold the pattern was incorrect (too large).
3958 If such patterns were nested a few deep, this could multiply and become a real
3959 problem.
3960
3961 2. Added /M option to pcretest to show the memory requirement of a specific
3962 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3963
3964 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3965 compiled in such a way that the backtracking after subsequent failure was
3966 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3967 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3968
3969
3970 Version 2.03 02-Feb-99
3971 ----------------------
3972
3973 1. Fixed typo and small mistake in man page.
3974
3975 2. Added 4th condition (GPL supersedes if conflict) and created separate
3976 LICENCE file containing the conditions.
3977
3978 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3979 Perl, that is the internal \ allows the delimiter to be included in the
3980 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3981 the final delimiter, add \ to the end of the pattern (to test the error).
3982
3983 4. Added the convenience functions for extracting substrings after a successful
3984 match. Updated pcretest to make it able to test these functions.
3985
3986
3987 Version 2.02 14-Jan-99
3988 ----------------------
3989
3990 1. Initialized the working variables associated with each extraction so that
3991 their saving and restoring doesn't refer to uninitialized store.
3992
3993 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3994 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3995 fix the problem.
3996
3997 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3998 calls, and wasn't printing the correct value for compiling calls. Increased the
3999 default value of LOOPREPEAT, and the number of significant figures in the
4000 times.
4001
4002 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
4003
4004 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
4005 a building problem on Windows NT with a FAT file system.
4006
4007
4008 Version 2.01 21-Oct-98
4009 ----------------------
4010
4011 1. Changed the API for pcre_compile() to allow for the provision of a pointer
4012 to character tables built by pcre_maketables() in the current locale. If NULL
4013 is passed, the default tables are used.
4014
4015
4016 Version 2.00 24-Sep-98
4017 ----------------------
4018
4019 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
4020 it any more.
4021
4022 2. Allow quantification of (?>) groups, and make it work correctly.
4023
4024 3. The first character computation wasn't working for (?>) groups.
4025
4026 4. Correct the implementation of \Z (it is permitted to match on the \n at the
4027 end of the subject) and add 5.005's \z, which really does match only at the
4028 very end of the subject.
4029
4030 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
4031
4032 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
4033 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
4034 localized options. All options to pcre_study() were also removed.
4035
4036 7. Add other new features from 5.005:
4037
4038 $(?<= positive lookbehind
4039 $(?<! negative lookbehind
4040 (?imsx-imsx) added the unsetting capability
4041 such a setting is global if at outer level; local otherwise
4042 (?imsx-imsx:) non-capturing groups with option setting
4043 (?(cond)re|re) conditional pattern matching
4044
4045 A backreference to itself in a repeated group matches the previous
4046 captured string.
4047
4048 8. General tidying up of studying (both automatic and via "study")
4049 consequential on the addition of new assertions.
4050
4051 9. As in 5.005, unlimited repeated groups that could match an empty substring
4052 are no longer faulted at compile time. Instead, the loop is forcibly broken at
4053 runtime if any iteration does actually match an empty substring.
4054
4055 10. Include the RunTest script in the distribution.
4056
4057 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
4058 discrepancies, some of which were old and were also with respect to 5.004. They
4059 have now been fixed.
4060
4061
4062 Version 1.09 28-Apr-98
4063 ----------------------
4064
4065 1. A negated single character class followed by a quantifier with a minimum
4066 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
4067 program crashes, or just wrong answers. This did not apply to negated classes
4068 containing more than one character, or to minima other than one.
4069
4070
4071 Version 1.08 27-Mar-98
4072 ----------------------
4073
4074 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
4075
4076 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
4077 latter must appear before anything that relies on it in the pattern.
4078
4079
4080 Version 1.07 16-Feb-98
4081 ----------------------
4082
4083 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
4084 repeat of a potentially empty string).
4085
4086
4087 Version 1.06 23-Jan-98
4088 ----------------------
4089
4090 1. Added Markus Oberhumer's little patches for C++.
4091
4092 2. Literal strings longer than 255 characters were broken.
4093
4094
4095 Version 1.05 23-Dec-97
4096 ----------------------
4097
4098 1. Negated character classes containing more than one character were failing if
4099 PCRE_CASELESS was set at run time.
4100
4101
4102 Version 1.04 19-Dec-97
4103 ----------------------
4104
4105 1. Corrected the man page, where some "const" qualifiers had been omitted.
4106
4107 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4108 input syntax.
4109
4110 3. Fixed memory leak which occurred when a regex with back references was
4111 matched with an offsets vector that wasn't big enough. The temporary memory
4112 that is used in this case wasn't being freed if the match failed.
4113
4114 4. Tidied pcretest to ensure it frees memory that it gets.
4115
4116 5. Temporary memory was being obtained in the case where the passed offsets
4117 vector was exactly big enough.
4118
4119 6. Corrected definition of offsetof() from change 5 below.
4120
4121 7. I had screwed up change 6 below and broken the rules for the use of
4122 setjmp(). Now fixed.
4123
4124
4125 Version 1.03 18-Dec-97
4126 ----------------------
4127
4128 1. A erroneous regex with a missing opening parenthesis was correctly
4129 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4130 on some systems.
4131
4132 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4133 it was reported that one broken compiler failed on the former because "code" is
4134 also an independent variable.
4135
4136 3. The erroneous regex a[]b caused an array overrun reference.
4137
4138 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4139 fail on data ending with that character. (It was going on too far, and checking
4140 the next character, typically a binary zero.) This was specific to the
4141 optimized code for single-character negative classes.
4142
4143 5. Added a contributed patch from the TIN world which does the following:
4144
4145 + Add an undef for memmove, in case the the system defines a macro for it.
4146
4147 + Add a definition of offsetof(), in case there isn't one. (I don't know
4148 the reason behind this - offsetof() is part of the ANSI standard - but
4149 it does no harm).
4150
4151 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4152 most of the places where whitespace preceded '#'. I have given up and
4153 allowed the remaining 2 cases to be at the margin.
4154
4155 + Rename some variables in pcre to eliminate shadowing. This seems very
4156 pedantic, but does no harm, of course.
4157
4158 6. Moved the call to setjmp() into its own function, to get rid of warnings
4159 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4160
4161 7. Constructs such as \d{8,} were compiling into the equivalent of
4162 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4163 outcome, but in this particular case used more store than had been allocated,
4164 which caused the bug to be discovered because it threw up an internal error.
4165
4166 8. The debugging code in both pcre and pcretest for outputting the compiled
4167 form of a regex was going wrong in the case of back references followed by
4168 curly-bracketed repeats.
4169
4170
4171 Version 1.02 12-Dec-97
4172 ----------------------
4173
4174 1. Typos in pcre.3 and comments in the source fixed.
4175
4176 2. Applied a contributed patch to get rid of places where it used to remove
4177 'const' from variables, and fixed some signed/unsigned and uninitialized
4178 variable warnings.
4179
4180 3. Added the "runtest" target to Makefile.
4181
4182 4. Set default compiler flag to -O2 rather than just -O.
4183
4184
4185 Version 1.01 19-Nov-97
4186 ----------------------
4187
4188 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4189 like /([ab]*)*/, that is, for classes with more than one character in them.
4190
4191 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4192 as /((?>a*))*/ (a PCRE_EXTRA facility).
4193
4194
4195 Version 1.00 18-Nov-97
4196 ----------------------
4197
4198 1. Added compile-time macros to support systems such as SunOS4 which don't have
4199 memmove() or strerror() but have other things that can be used instead.
4200
4201 2. Arranged that "make clean" removes the executables.
4202
4203
4204 Version 0.99 27-Oct-97
4205 ----------------------
4206
4207 1. Fixed bug in code for optimizing classes with only one character. It was
4208 initializing a 32-byte map regardless, which could cause it to run off the end
4209 of the memory it had got.
4210
4211 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4212
4213
4214 Version 0.98 22-Oct-97
4215 ----------------------
4216
4217 1. Fixed bug in code for handling temporary memory usage when there are more
4218 back references than supplied space in the ovector. This could cause segfaults.
4219
4220
4221 Version 0.97 21-Oct-97
4222 ----------------------
4223
4224 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4225
4226 2. Optimized negated single characters not to use a bit map.
4227
4228 3. Brought error texts together as macro definitions; clarified some of them;
4229 fixed one that was wrong - it said "range out of order" when it meant "invalid
4230 escape sequence".
4231
4232 4. Changed some char * arguments to const char *.
4233
4234 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4235
4236 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4237 pcretest.
4238
4239
4240 Version 0.96 16-Oct-97
4241 ----------------------
4242
4243 1. Added a simple "pgrep" utility to the distribution.
4244
4245 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4246 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4247 where "ddd" means "one or more decimal digits".
4248
4249 3. Fixed serious bug. If a pattern had a back reference, but the call to
4250 pcre_exec() didn't supply a large enough ovector to record the related
4251 identifying subpattern, the match always failed. PCRE now remembers the number
4252 of the largest back reference, and gets some temporary memory in which to save
4253 the offsets during matching if necessary, in order to ensure that
4254 backreferences always work.
4255
4256 4. Increased the compatibility with Perl in a number of ways:
4257
4258 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4259 to request this handling. The option can be set at compile or exec time.
4260
4261 (b) $ matches before a terminating newline by default; an option
4262 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4263 mode). The option can be set at compile or exec time.
4264
4265 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4266 the same as Perl's. If the decimal number it represents is less than 10
4267 or there aren't that many previous left capturing parentheses, an octal
4268 escape is read. Inside a character class, it's always an octal escape,
4269 even if it is a single digit.
4270
4271 (d) An escaped but undefined alphabetic character is taken as a literal,
4272 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4273 escapes.
4274
4275 (e) {0} is now permitted. (The previous item is removed from the compiled
4276 pattern).
4277
4278 5. Changed all the names of code files so that the basic parts are no longer
4279 than 10 characters, and abolished the teeny "globals.c" file.
4280
4281 6. Changed the handling of character classes; they are now done with a 32-byte
4282 bit map always.
4283
4284 7. Added the -d and /D options to pcretest to make it possible to look at the
4285 internals of compilation without having to recompile pcre.
4286
4287
4288 Version 0.95 23-Sep-97
4289 ----------------------
4290
4291 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4292 \x20 at the start of a run of normal characters. These were being treated as
4293 real characters, instead of the source characters being re-checked.
4294
4295
4296 Version 0.94 18-Sep-97
4297 ----------------------
4298
4299 1. The functions are now thread-safe, with the caveat that the global variables
4300 containing pointers to malloc() and free() or alternative functions are the
4301 same for all threads.
4302
4303 2. Get pcre_study() to generate a bitmap of initial characters for non-
4304 anchored patterns when this is possible, and use it if passed to pcre_exec().
4305
4306
4307 Version 0.93 15-Sep-97
4308 ----------------------
4309
4310 1. /(b)|(:+)/ was computing an incorrect first character.
4311
4312 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4313 but not actually doing anything yet.
4314
4315 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4316 as Perl does (e.g. [-az] or [az-]).
4317
4318 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4319 all possible positions.
4320
4321 5. Split up into different modules to avoid including unneeded functions in a
4322 compiled binary. However, compile and exec are still in one module. The "study"
4323 function is split off.
4324
4325 6. The character tables are now in a separate module whose source is generated
4326 by an auxiliary program - but can then be edited by hand if required. There are
4327 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4328 toupper() in the code.
4329
4330 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4331 make them global. Abolish the function for setting them, as the caller can now
4332 set them directly.
4333
4334
4335 Version 0.92 11-Sep-97
4336 ----------------------
4337
4338 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4339 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4340
4341 2. Caseless matching was not working in character classes if the characters in
4342 the pattern were in upper case.
4343
4344 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4345
4346 4. Make PCRE_ANCHORED public and accept as a compile option.
4347
4348 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4349 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4350 pass them.
4351
4352 6. Give an error if bad option bits passed at compile or run time.
4353
4354 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4355 pcretest to cause it to pass that flag.
4356
4357 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4358 options, and the first character, if set.
4359
4360 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4361
4362
4363 Version 0.91 10-Sep-97
4364 ----------------------
4365
4366 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4367 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4368
4369 2. PCRE was looping on encountering an indefinitely repeated back reference to
4370 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4371 Perl does - treats the match as successful.
4372
4373 ****

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