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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 648 - (show annotations)
Mon Aug 1 11:02:08 2011 UTC (3 years, 11 months ago) by ph10
File size: 211222 byte(s)
Error occurred while calculating annotation data.
Fix bug that caused /.(*F)/ to give a partial match instead of no match.
1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.13 30-Apr-2011
5 ------------------------
6
7 1. The Unicode data tables have been updated to Unicode 6.0.0.
8
9 2. Two minor typos in pcre_internal.h have been fixed.
10
11 3. Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12 pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13 in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14
15 4. There were a number of related bugs in the code for matching backrefences
16 caselessly in UTF-8 mode when codes for the characters concerned were
17 different numbers of bytes. For example, U+023A and U+2C65 are an upper
18 and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19 (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20 code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21 2-byte code at the end of the subject (it thought there wasn't enough data
22 left).
23
24 5. Comprehensive information about what went wrong is now returned by
25 pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26 as the output vector has at least 2 elements. The offset of the start of
27 the failing character and a reason code are placed in the vector.
28
29 6. When the UTF-8 string check fails for pcre_compile(), the offset that is
30 now returned is for the first byte of the failing character, instead of the
31 last byte inspected. This is an incompatible change, but I hope it is small
32 enough not to be a problem. It makes the returned offset consistent with
33 pcre_exec() and pcre_dfa_exec().
34
35 7. pcretest now gives a text phrase as well as the error number when
36 pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37 failure, the offset and reason code are output.
38
39 8. When \R was used with a maximizing quantifier it failed to skip backwards
40 over a \r\n pair if the subsequent match failed. Instead, it just skipped
41 back over a single character (\n). This seems wrong (because it treated the
42 two characters as a single entity when going forwards), conflicts with the
43 documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44 behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45 has been changed.
46
47 9. Some internal refactoring has changed the processing so that the handling
48 of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49 time (the PCRE_DOTALL option was changed this way some time ago: version
50 7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51 which was always a bit of a fudge. It also means that there is one less
52 argument for the match() function, which reduces its stack requirements
53 slightly. This change also fixes an incompatibility with Perl: the pattern
54 (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55
56 10. More internal refactoring has drastically reduced the number of recursive
57 calls to match() for possessively repeated groups such as (abc)++ when
58 using pcre_exec().
59
60 11. While implementing 10, a number of bugs in the handling of groups were
61 discovered and fixed:
62
63 (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64 (a|)*(?1) gave a compile-time internal error.
65 ((a|)+)+ did not notice that the outer group could match an empty string.
66 (^a|^)+ was not marked as anchored.
67 (.*a|.*)+ was not marked as matching at start or after a newline.
68
69 12. Yet more internal refactoring has removed another argument from the match()
70 function. Special calls to this function are now indicated by setting a
71 value in a variable in the "match data" data block.
72
73 13. Be more explicit in pcre_study() instead of relying on "default" for
74 opcodes that mean there is no starting character; this means that when new
75 ones are added and accidentally left out of pcre_study(), testing should
76 pick them up.
77
78 14. The -s option of pcretest has been documented for ages as being an old
79 synonym of -m (show memory usage). I have changed it to mean "force study
80 for every regex", that is, assume /S for every regex. This is similar to -i
81 and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82 using it. It makes it easier to run collections of tests with and without
83 study enabled, and thereby test pcre_study() more easily. All the standard
84 tests are now run with and without -s (but some patterns can be marked as
85 "never study" - see 20 below).
86
87 15. When (*ACCEPT) was used in a subpattern that was called recursively, the
88 restoration of the capturing data to the outer values was not happening
89 correctly.
90
91 16. If a recursively called subpattern ended with (*ACCEPT) and matched an
92 empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
93 pattern had matched an empty string, and so incorrectly returned a no
94 match.
95
96 17. There was optimizing code for the last branch of non-capturing parentheses,
97 and also for the obeyed branch of a conditional subexpression, which used
98 tail recursion to cut down on stack usage. Unfortunately, not that there is
99 the possibility of (*THEN) occurring in these branches, tail recursion is
100 no longer possible because the return has to be checked for (*THEN). These
101 two optimizations have therefore been removed.
102
103 18. If a pattern containing \R was studied, it was assumed that \R always
104 matched two bytes, thus causing the minimum subject length to be
105 incorrectly computed because \R can also match just one byte.
106
107 19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108 was incorrectly computed.
109
110 20. If /S is present twice on a test pattern in pcretest input, it *disables*
111 studying, thereby overriding the use of -s on the command line. This is
112 necessary for one or two tests to keep the output identical in both cases.
113
114 21. When (*ACCEPT) was used in an assertion that matched an empty string and
115 PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
116
117 22. When an atomic group that contained a capturing parenthesis was
118 successfully matched, but the branch in which it appeared failed, the
119 capturing was not being forgotten if a higher numbered group was later
120 captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
121 group 1 to "a", when in fact it should be unset. This applied to multi-
122 branched capturing and non-capturing groups, repeated or not, and also to
123 positive assertions (capturing in negative assertions is not well defined
124 in PCRE) and also to nested atomic groups.
125
126 23. Add the ++ qualifier feature to pcretest, to show the remainder of the
127 subject after a captured substring (to make it easier to tell which of a
128 number of identical substrings has been captured).
129
130 24. The way atomic groups are processed by pcre_exec() has been changed so that
131 if they are repeated, backtracking one repetition now resets captured
132 values correctly. For example, if ((?>(a+)b)+aabab) is matched against
133 "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
134 "aaa". Previously, it would have been "a". As part of this code
135 refactoring, the way recursive calls are handled has also been changed.
136
137 24. If an assertion condition captured any substrings, they were not passed
138 back unless some other capturing happened later. For example, if
139 (?(?=(a))a) was matched against "a", no capturing was returned.
140
141 25. When studying a pattern that contained subroutine calls or assertions,
142 the code for finding the minimum length of a possible match was handling
143 direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
144 group 1 called group 2 while simultaneously a separate group 2 called group
145 1). A stack overflow occurred in this case. I have fixed this by limiting
146 the recursion depth to 10.
147
148 26. Updated RunTest.bat in the distribution to the version supplied by Tom
149 Fortmann. This supports explicit test numbers on the command line, and has
150 argument validation and error reporting.
151
152 27. An instance of \X with an unlimited repeat could fail if at any point the
153 first character it looked at was a mark character.
154
155 28. Some minor code refactoring concerning Unicode properties and scripts
156 should reduce the stack requirement of match() slightly.
157
158 29. Added the '=' option to pcretest to check the setting of unused capturing
159 slots at the end of the pattern, which are documented as being -1, but are
160 not included in the return count.
161
162 30. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
163 compiled something random. Now it gives a compile-time error (as does
164 Perl).
165
166 31. A *MARK encountered during the processing of a positive assertion is now
167 recorded and passed back (compatible with Perl).
168
169 32. If --only-matching or --colour was set on a pcregrep call whose pattern
170 had alternative anchored branches, the search for a second match in a line
171 was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
172 matched the line "0102" twice. The same bug affected patterns that started
173 with a backwards assertion. For example /\b01|\b02/ also matched "0102"
174 twice.
175
176 33. Previously, PCRE did not allow quantification of assertions. However, Perl
177 does, and because of capturing effects, quantifying parenthesized
178 assertions may at times be useful. Quantifiers are now allowed for
179 parenthesized assertions.
180
181 34. A minor code tidy in pcre_compile() when checking options for \R usage.
182
183 35. \g was being checked for fancy things in a character class, when it should
184 just be a literal "g".
185
186 36. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
187 appearance of a nested POSIX class supersedes an apparent external class.
188 For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
189 unescaped square brackets may also appear as part of class names. For
190 example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
191 more like Perl.
192
193 37. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
194 was because it thought it was \N{name}, which is not supported).
195
196 38. Add minix to OS list not supporting the -S option in pcretest.
197
198 39. PCRE tries to detect cases of infinite recursion at compile time, but it
199 cannot analyze patterns in sufficient detail to catch mutual recursions
200 such as ((?1))((?2)). There is now a runtime test that gives an error if a
201 subgroup is called recursively as a subpattern for a second time at the
202 same position in the subject string. In previous releases this might have
203 been caught by the recursion limit, or it might have run out of stack.
204
205 40. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
206 happen only once. PCRE was, however incorrectly giving a compile time error
207 "recursive call could loop indefinitely" because it cannot analyze the
208 pattern in sufficient detail. The compile time test no longer happens when
209 PCRE is compiling a conditional subpattern, but actual runaway loops are
210 now caught at runtime (see 39 above).
211
212 41. It seems that Perl allows any characters other than a closing parenthesis
213 to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
214 has been changed to be the same.
215
216 42. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
217 as not to get warnings when autogen.sh is called. Also changed
218 AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
219
220 43. To help people who use pcregrep to scan files containing exceedingly long
221 lines, the following changes have been made:
222
223 (a) The default value of the buffer size parameter has been increased from
224 8K to 20K. (A buffer three times this size is actually used.)
225
226 (b) The default can be changed by ./configure --with-pcregrep-bufsiz when
227 PCRE is built.
228
229 (c) A --buffer-size=n option has been added to pcregrep, to allow the size
230 to be set at run time.
231
232 (d) Numerical values in pcregrep options can be followed by K or M, for
233 example --buffer-size=50K.
234
235 (e) If a line being scanned overflows pcregrep's buffer, an error is now
236 given and the return code is set to 2.
237
238 44. Add a pointer to the latest mark to the callout data block.
239
240 45. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
241 partial match of an empty string instead of no match. This was specific to
242 the use of ".".
243
244
245 Version 8.12 15-Jan-2011
246 ------------------------
247
248 1. Fixed some typos in the markup of the man pages, and wrote a script that
249 checks for such things as part of the documentation building process.
250
251 2. On a big-endian 64-bit system, pcregrep did not correctly process the
252 --match-limit and --recursion-limit options (added for 8.11). In
253 particular, this made one of the standard tests fail. (The integer value
254 went into the wrong half of a long int.)
255
256 3. If the --colour option was given to pcregrep with -v (invert match), it
257 did strange things, either producing crazy output, or crashing. It should,
258 of course, ignore a request for colour when reporting lines that do not
259 match.
260
261 4. Another pcregrep bug caused similar problems if --colour was specified with
262 -M (multiline) and the pattern match finished with a line ending.
263
264 5. In pcregrep, when a pattern that ended with a literal newline sequence was
265 matched in multiline mode, the following line was shown as part of the
266 match. This seems wrong, so I have changed it.
267
268 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
269 the check for further matches in the same line (so they could be coloured)
270 to overrun the end of the current line. If another match was found, it was
271 incorrectly shown (and then shown again when found in the next line).
272
273 7. If pcregrep was compiled under Windows, there was a reference to the
274 function pcregrep_exit() before it was defined. I am assuming this was
275 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
276 reported by a user. I've moved the definition above the reference.
277
278
279 Version 8.11 10-Dec-2010
280 ------------------------
281
282 1. (*THEN) was not working properly if there were untried alternatives prior
283 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
284 backtracked to try for "b" instead of moving to the next alternative branch
285 at the same level (in this case, to look for "c"). The Perl documentation
286 is clear that when (*THEN) is backtracked onto, it goes to the "next
287 alternative in the innermost enclosing group".
288
289 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
290 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
291 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
292 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
293 (*THEN).
294
295 3. If \s appeared in a character class, it removed the VT character from
296 the class, even if it had been included by some previous item, for example
297 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
298 of \s, but is part of the POSIX "space" class.)
299
300 4. A partial match never returns an empty string (because you can always
301 match an empty string at the end of the subject); however the checking for
302 an empty string was starting at the "start of match" point. This has been
303 changed to the "earliest inspected character" point, because the returned
304 data for a partial match starts at this character. This means that, for
305 example, /(?<=abc)def/ gives a partial match for the subject "abc"
306 (previously it gave "no match").
307
308 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
309 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
310 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
311 has an implication that the given string is incomplete (because a partial
312 match is preferred over a full match). For this reason, these items now
313 give a partial match in this situation. [Aside: previously, the one case
314 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
315 match rather than a full match, which was wrong by the old rules, but is
316 now correct.]
317
318 6. There was a bug in the handling of #-introduced comments, recognized when
319 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
320 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
321 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
322 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
323 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
324 places in pcre_compile().
325
326 7. Related to (6) above, when pcre_compile() was skipping #-introduced
327 comments when looking ahead for named forward references to subpatterns,
328 the only newline sequence it recognized was NL. It now handles newlines
329 according to the set newline convention.
330
331 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
332 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
333 cater for a lack of strerror(). These oversights have been fixed.
334
335 9. Added --match-limit and --recursion-limit to pcregrep.
336
337 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
338
339 11. When the -o option was used, pcregrep was setting a return code of 1, even
340 when matches were found, and --line-buffered was not being honoured.
341
342 12. Added an optional parentheses number to the -o and --only-matching options
343 of pcregrep.
344
345 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
346 can match an empty string. The code to do it in pcretest and pcredemo
347 needed fixing:
348
349 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
350 only one byte after an empty string match just before CRLF (this case
351 just got forgotten; "any" and "anycrlf" were OK).
352
353 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
354 mode when an empty string match preceded an ASCII character followed by
355 a non-ASCII character. (The code for advancing by one character rather
356 than one byte was nonsense.)
357
358 (c) The pcredemo.c sample program did not have any code at all to handle
359 the cases when CRLF is a valid newline sequence.
360
361 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
362 as a starting offset was within the subject string. There is now a new
363 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
364 negative or greater than the length of the string. In order to test this,
365 pcretest is extended to allow the setting of negative starting offsets.
366
367 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
368 starting offset points to the beginning of a UTF-8 character was
369 unnecessarily clumsy. I tidied it up.
370
371 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
372 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
373
374 17. Nobody had reported that the --include_dir option, which was added in
375 release 7.7 should have been called --include-dir (hyphen, not underscore)
376 for compatibility with GNU grep. I have changed it to --include-dir, but
377 left --include_dir as an undocumented synonym, and the same for
378 --exclude-dir, though that is not available in GNU grep, at least as of
379 release 2.5.4.
380
381 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
382 characters from a string of bytes) have been redefined so as not to use
383 loops, in order to improve performance in some environments. At the same
384 time, I abstracted some of the common code into auxiliary macros to save
385 repetition (this should not affect the compiled code).
386
387 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
388 compile-time error is now given if \c is not followed by an ASCII
389 character, that is, a byte less than 128. (In EBCDIC mode, the code is
390 different, and any byte value is allowed.)
391
392 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
393 START_OPTIMIZE option, which is now allowed at compile time - but just
394 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
395 to pcregrep and other applications that have no direct access to PCRE
396 options. The new /Y option in pcretest sets this option when calling
397 pcre_compile().
398
399 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
400 back references. Groups containing recursive back references were forced to
401 be atomic by that change, but in the case of named groups, the amount of
402 memory required was incorrectly computed, leading to "Failed: internal
403 error: code overflow". This has been fixed.
404
405 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
406 pcretest.c, to avoid build problems in some Borland environments.
407
408
409 Version 8.10 25-Jun-2010
410 ------------------------
411
412 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
413 THEN.
414
415 2. (*ACCEPT) was not working when inside an atomic group.
416
417 3. Inside a character class, \B is treated as a literal by default, but
418 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
419 causes the error). The code is unchanged, but I tidied the documentation.
420
421 4. Inside a character class, PCRE always treated \R and \X as literals,
422 whereas Perl faults them if its -w option is set. I have changed PCRE so
423 that it faults them when PCRE_EXTRA is set.
424
425 5. Added support for \N, which always matches any character other than
426 newline. (It is the same as "." when PCRE_DOTALL is not set.)
427
428 6. When compiling pcregrep with newer versions of gcc which may have
429 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
430 declared with attribute warn_unused_result" were given. Just casting the
431 result to (void) does not stop the warnings; a more elaborate fudge is
432 needed. I've used a macro to implement this.
433
434 7. Minor change to pcretest.c to avoid a compiler warning.
435
436 8. Added four artifical Unicode properties to help with an option to make
437 \s etc use properties (see next item). The new properties are: Xan
438 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
439
440 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
441 use Unicode properties. (*UCP) at the start of a pattern can be used to set
442 this option. Modified pcretest to add /W to test this facility. Added
443 REG_UCP to make it available via the POSIX interface.
444
445 10. Added --line-buffered to pcregrep.
446
447 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
448 studied, and the match started with a letter with a code point greater than
449 127 whose first byte was different to the first byte of the other case of
450 the letter, the other case of this starting letter was not recognized
451 (#976).
452
453 12. If a pattern that was studied started with a repeated Unicode property
454 test, for example, \p{Nd}+, there was the theoretical possibility of
455 setting up an incorrect bitmap of starting bytes, but fortunately it could
456 not have actually happened in practice until change 8 above was made (it
457 added property types that matched character-matching opcodes).
458
459 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
460 possible starting bytes for non-anchored patterns.
461
462 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
463 \R, and also a number of cases that involve Unicode properties, both
464 explicit and implicit when PCRE_UCP is set.
465
466 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
467 input, it could crash or give wrong results if characters with values
468 greater than 0xc0 were present in the subject string. (Detail: it assumed
469 UTF-8 input when processing these items.)
470
471 16. Added a lot of (int) casts to avoid compiler warnings in systems where
472 size_t is 64-bit (#991).
473
474 17. Added a check for running out of memory when PCRE is compiled with
475 --disable-stack-for-recursion (#990).
476
477 18. If the last data line in a file for pcretest does not have a newline on
478 the end, a newline was missing in the output.
479
480 19. The default pcre_chartables.c file recognizes only ASCII characters (values
481 less than 128) in its various bitmaps. However, there is a facility for
482 generating tables according to the current locale when PCRE is compiled. It
483 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
484 characters, are recognized by isspace() and therefore were getting set in
485 these tables, and indeed these tables seem to approximate to ISO 8859. This
486 caused a problem in UTF-8 mode when pcre_study() was used to create a list
487 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
488 which of course cannot start UTF-8 characters. I have changed the code so
489 that only real ASCII characters (less than 128) and the correct starting
490 bytes for UTF-8 encodings are set for characters greater than 127 when in
491 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
492 altogether.)
493
494 20. Added the /T option to pcretest so as to be able to run tests with non-
495 standard character tables, thus making it possible to include the tests
496 used for 19 above in the standard set of tests.
497
498 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
499 reference to a subpattern the other side of a comment that contains an
500 opening parenthesis caused either an internal compiling error, or a
501 reference to the wrong subpattern.
502
503
504 Version 8.02 19-Mar-2010
505 ------------------------
506
507 1. The Unicode data tables have been updated to Unicode 5.2.0.
508
509 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
510 configured.
511
512 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
513 original author of that file, following a query about its status.
514
515 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
516 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
517
518 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
519 quantifier applied to a forward-referencing subroutine call, could compile
520 incorrect code or give the error "internal error: previously-checked
521 referenced subpattern not found".
522
523 6. Both MS Visual Studio and Symbian OS have problems with initializing
524 variables to point to external functions. For these systems, therefore,
525 pcre_malloc etc. are now initialized to local functions that call the
526 relevant global functions.
527
528 7. There were two entries missing in the vectors called coptable and poptable
529 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
530 I've fixed the data, and added a kludgy way of testing at compile time that
531 the lengths are correct (equal to the number of opcodes).
532
533 8. Following on from 7, I added a similar kludge to check the length of the
534 eint vector in pcreposix.c.
535
536 9. Error texts for pcre_compile() are held as one long string to avoid too
537 much relocation at load time. To find a text, the string is searched,
538 counting zeros. There was no check for running off the end of the string,
539 which could happen if a new error number was added without updating the
540 string.
541
542 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
543
544 11. \K was not working if it appeared in an atomic group or in a group that
545 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
546 \K is "not well defined" if used in an assertion. PCRE now accepts it if
547 the assertion is positive, but not if it is negative.
548
549 12. Change 11 fortuitously reduced the size of the stack frame used in the
550 "match()" function of pcre_exec.c by one pointer. Forthcoming
551 implementation of support for (*MARK) will need an extra pointer on the
552 stack; I have reserved it now, so that the stack frame size does not
553 decrease.
554
555 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
556 item in branch that calls a recursion is a subroutine call - as in the
557 second branch in the above example - was incorrectly given the compile-
558 time error "recursive call could loop indefinitely" because pcre_compile()
559 was not correctly checking the subroutine for matching a non-empty string.
560
561 14. The checks for overrunning compiling workspace could trigger after an
562 overrun had occurred. This is a "should never occur" error, but it can be
563 triggered by pathological patterns such as hundreds of nested parentheses.
564 The checks now trigger 100 bytes before the end of the workspace.
565
566 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
567
568
569 Version 8.01 19-Jan-2010
570 ------------------------
571
572 1. If a pattern contained a conditional subpattern with only one branch (in
573 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
574 computed the wrong minimum data length (which is of course zero for such
575 subpatterns). This could cause incorrect "no match" results.
576
577 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
578 the pattern is reset in the first branch, pcre_compile() failed with
579 "internal error: code overflow at offset...". This happened only when
580 the reset was to the original external option setting. (An optimization
581 abstracts leading options settings into an external setting, which was the
582 cause of this.)
583
584 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
585 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
586 assertion pattern did not match (meaning that the assertion was true), it
587 was incorrectly treated as false if the SKIP had been reached during the
588 matching. This also applied to assertions used as conditions.
589
590 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
591 assertion subpattern, including such a pattern used as a condition,
592 unpredictable results occurred, instead of the error return
593 PCRE_ERROR_DFA_UITEM.
594
595 5. The C++ GlobalReplace function was not working like Perl for the special
596 situation when an empty string is matched. It now does the fancy magic
597 stuff that is necessary.
598
599 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
600 removed. (These were left over from very, very early versions of PCRE.)
601
602 7. Some cosmetic changes to the code to make life easier when compiling it
603 as part of something else:
604
605 (a) Change DEBUG to PCRE_DEBUG.
606
607 (b) In pcre_compile(), rename the member of the "branch_chain" structure
608 called "current" as "current_branch", to prevent a collision with the
609 Linux macro when compiled as a kernel module.
610
611 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
612 prevent a collision with the Linux macro when compiled as a kernel
613 module.
614
615 8. In pcre_compile() there are some checks for integer overflows that used to
616 cast potentially large values to (double). This has been changed to that
617 when building, a check for int64_t is made, and if it is found, it is used
618 instead, thus avoiding the use of floating point arithmetic. (There is no
619 other use of FP in PCRE.) If int64_t is not found, the fallback is to
620 double.
621
622 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
623 2005 (difference between two addresses compared to an unsigned value).
624
625 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
626 custom one, because of the following reported problem in Windows:
627
628 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
629 under Win32.
630 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
631 therefore missing the function definition.
632 - The compiler thus generates a "C" signature for the test function.
633 - The linker fails to find the "C" function.
634 - PCRE fails to configure if asked to do so against libbz2.
635
636 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
637 messages were output:
638
639 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
640 rerunning libtoolize, to keep the correct libtool macros in-tree.
641 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
642
643 I have done both of these things.
644
645 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
646 most of the time, it *can* run out if it is given a pattern that contains a
647 runaway infinite recursion. I updated the discussion in the pcrestack man
648 page.
649
650 13. Now that we have gone to the x.xx style of version numbers, the minor
651 version may start with zero. Using 08 or 09 is a bad idea because users
652 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
653 interpreted as invalid octal numbers. I've updated the previous comment in
654 configure.ac, and also added a check that gives an error if 08 or 09 are
655 used.
656
657 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
658 causing partial matching to fail when the end of the subject matched \W
659 in a UTF-8 pattern where \W was quantified with a minimum of 3.
660
661 15. There were some discrepancies between the declarations in pcre_internal.h
662 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
663 their definitions. The declarations used "const uschar *" and the
664 definitions used USPTR. Even though USPTR is normally defined as "const
665 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
666 reported that: "This difference in casting confuses some C++ compilers, for
667 example, SunCC recognizes above declarations as different functions and
668 generates broken code for hbpcre." I have changed the declarations to use
669 USPTR.
670
671 16. GNU libtool is named differently on some systems. The autogen.sh script now
672 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
673 (FreeBSD).
674
675 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
676 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
677 comment: "Figure out how to create a longlong from a string: strtoll and
678 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
679 instance, but it only takes 2 args instead of 3!"
680
681 18. A subtle bug concerned with back references has been fixed by a change of
682 specification, with a corresponding code fix. A pattern such as
683 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
684 refers, was giving matches when it shouldn't. For example, xa=xaaa would
685 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
686 same bug. Such groups have to be quantified to be useful, or contained
687 inside another quantified group. (If there's no repetition, the reference
688 can never match.) The problem arises because, having left the group and
689 moved on to the rest of the pattern, a later failure that backtracks into
690 the group uses the captured value from the final iteration of the group
691 rather than the correct earlier one. I have fixed this in PCRE by forcing
692 any group that contains a reference to itself to be an atomic group; that
693 is, there cannot be any backtracking into it once it has completed. This is
694 similar to recursive and subroutine calls.
695
696
697 Version 8.00 19-Oct-09
698 ----------------------
699
700 1. The table for translating pcre_compile() error codes into POSIX error codes
701 was out-of-date, and there was no check on the pcre_compile() error code
702 being within the table. This could lead to an OK return being given in
703 error.
704
705 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
706 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
707 in a Windows environment.
708
709 3. The pcregrep --count option prints the count for each file even when it is
710 zero, as does GNU grep. However, pcregrep was also printing all files when
711 --files-with-matches was added. Now, when both options are given, it prints
712 counts only for those files that have at least one match. (GNU grep just
713 prints the file name in this circumstance, but including the count seems
714 more useful - otherwise, why use --count?) Also ensured that the
715 combination -clh just lists non-zero counts, with no names.
716
717 4. The long form of the pcregrep -F option was incorrectly implemented as
718 --fixed_strings instead of --fixed-strings. This is an incompatible change,
719 but it seems right to fix it, and I didn't think it was worth preserving
720 the old behaviour.
721
722 5. The command line items --regex=pattern and --regexp=pattern were not
723 recognized by pcregrep, which required --regex pattern or --regexp pattern
724 (with a space rather than an '='). The man page documented the '=' forms,
725 which are compatible with GNU grep; these now work.
726
727 6. No libpcreposix.pc file was created for pkg-config; there was just
728 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
729
730 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
731 when UCP support is not needed, by modifying the Python script that
732 generates it from Unicode data files. This should not matter if the module
733 is correctly used as a library, but I received one complaint about 50K of
734 unwanted data. My guess is that the person linked everything into his
735 program rather than using a library. Anyway, it does no harm.
736
737 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
738 was a minimum greater than 1 for a wide character in a possessive
739 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
740 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
741 character. Chaos in the form of incorrect output or a compiling loop could
742 result.
743
744 9. The restrictions on what a pattern can contain when partial matching is
745 requested for pcre_exec() have been removed. All patterns can now be
746 partially matched by this function. In addition, if there are at least two
747 slots in the offset vector, the offset of the earliest inspected character
748 for the match and the offset of the end of the subject are set in them when
749 PCRE_ERROR_PARTIAL is returned.
750
751 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
752 synonymous with PCRE_PARTIAL, for backwards compatibility, and
753 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
754 and may be more useful for multi-segment matching.
755
756 11. Partial matching with pcre_exec() is now more intuitive. A partial match
757 used to be given if ever the end of the subject was reached; now it is
758 given only if matching could not proceed because another character was
759 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
760 string "Z", which now yields "no match" instead of "partial match". In the
761 case of pcre_dfa_exec(), "no match" is given if every matching path for the
762 final character ended with (*FAIL).
763
764 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
765 if the pattern had a "must contain" character that was already found in the
766 earlier partial match, unless partial matching was again requested. For
767 example, with the pattern /dog.(body)?/, the "must contain" character is
768 "g". If the first part-match was for the string "dog", restarting with
769 "sbody" failed. This bug has been fixed.
770
771 13. The string returned by pcre_dfa_exec() after a partial match has been
772 changed so that it starts at the first inspected character rather than the
773 first character of the match. This makes a difference only if the pattern
774 starts with a lookbehind assertion or \b or \B (\K is not supported by
775 pcre_dfa_exec()). It's an incompatible change, but it makes the two
776 matching functions compatible, and I think it's the right thing to do.
777
778 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
779 so that the demonstration program is easily available in environments where
780 PCRE has not been installed from source.
781
782 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
783 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
784 library.
785
786 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
787 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
788 is not the first non-POSIX option to be added. Clearly some people find
789 these options useful.
790
791 17. If a caller to the POSIX matching function regexec() passes a non-zero
792 value for nmatch with a NULL value for pmatch, the value of
793 nmatch is forced to zero.
794
795 18. RunGrepTest did not have a test for the availability of the -u option of
796 the diff command, as RunTest does. It now checks in the same way as
797 RunTest, and also checks for the -b option.
798
799 19. If an odd number of negated classes containing just a single character
800 interposed, within parentheses, between a forward reference to a named
801 subpattern and the definition of the subpattern, compilation crashed with
802 an internal error, complaining that it could not find the referenced
803 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
804 [The bug was that it was starting one character too far in when skipping
805 over the character class, thus treating the ] as data rather than
806 terminating the class. This meant it could skip too much.]
807
808 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
809 /g option in pcretest when the pattern contains \K, which makes it possible
810 to have an empty string match not at the start, even when the pattern is
811 anchored. Updated pcretest and pcredemo to use this option.
812
813 21. If the maximum number of capturing subpatterns in a recursion was greater
814 than the maximum at the outer level, the higher number was returned, but
815 with unset values at the outer level. The correct (outer level) value is
816 now given.
817
818 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
819 PCRE did not set those parentheses (unlike Perl). I have now found a way to
820 make it do so. The string so far is captured, making this feature
821 compatible with Perl.
822
823 23. The tests have been re-organized, adding tests 11 and 12, to make it
824 possible to check the Perl 5.10 features against Perl 5.10.
825
826 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
827 pattern matches a fixed length string. PCRE did not allow this; now it
828 does. Neither allows recursion.
829
830 25. I finally figured out how to implement a request to provide the minimum
831 length of subject string that was needed in order to match a given pattern.
832 (It was back references and recursion that I had previously got hung up
833 on.) This code has now been added to pcre_study(); it finds a lower bound
834 to the length of subject needed. It is not necessarily the greatest lower
835 bound, but using it to avoid searching strings that are too short does give
836 some useful speed-ups. The value is available to calling programs via
837 pcre_fullinfo().
838
839 26. While implementing 25, I discovered to my embarrassment that pcretest had
840 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
841 study optimizations had never been tested with that matching function.
842 Oops. What is worse, even when it was passed study data, there was a bug in
843 pcre_dfa_exec() that meant it never actually used it. Double oops. There
844 were also very few tests of studied patterns with pcre_dfa_exec().
845
846 27. If (?| is used to create subpatterns with duplicate numbers, they are now
847 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
848 on the other side of the coin, they are no longer allowed to have different
849 names, because these cannot be distinguished in PCRE, and this has caused
850 confusion. (This is a difference from Perl.)
851
852 28. When duplicate subpattern names are present (necessarily with different
853 numbers, as required by 27 above), and a test is made by name in a
854 conditional pattern, either for a subpattern having been matched, or for
855 recursion in such a pattern, all the associated numbered subpatterns are
856 tested, and the overall condition is true if the condition is true for any
857 one of them. This is the way Perl works, and is also more like the way
858 testing by number works.
859
860
861 Version 7.9 11-Apr-09
862 ---------------------
863
864 1. When building with support for bzlib/zlib (pcregrep) and/or readline
865 (pcretest), all targets were linked against these libraries. This included
866 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
867 libraries. This caused unwanted dependencies to be created. This problem
868 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
869 pcretest is linked with readline.
870
871 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
872 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
873 moved outside it again, because FALSE and TRUE are already defined in AIX,
874 but BOOL is not.
875
876 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
877 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
878
879 4. The pcregrep documentation said spaces were inserted as well as colons (or
880 hyphens) following file names and line numbers when outputting matching
881 lines. This is not true; no spaces are inserted. I have also clarified the
882 wording for the --colour (or --color) option.
883
884 5. In pcregrep, when --colour was used with -o, the list of matching strings
885 was not coloured; this is different to GNU grep, so I have changed it to be
886 the same.
887
888 6. When --colo(u)r was used in pcregrep, only the first matching substring in
889 each matching line was coloured. Now it goes on to look for further matches
890 of any of the test patterns, which is the same behaviour as GNU grep.
891
892 7. A pattern that could match an empty string could cause pcregrep to loop; it
893 doesn't make sense to accept an empty string match in pcregrep, so I have
894 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
895 seems to be how GNU grep behaves.
896
897 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
898 start or after a newline", because the conditional assertion was not being
899 correctly handled. The rule now is that both the assertion and what follows
900 in the first alternative must satisfy the test.
901
902 9. If auto-callout was enabled in a pattern with a conditional group whose
903 condition was an assertion, PCRE could crash during matching, both with
904 pcre_exec() and pcre_dfa_exec().
905
906 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
907 used for matching.
908
909 11. Unicode property support in character classes was not working for
910 characters (bytes) greater than 127 when not in UTF-8 mode.
911
912 12. Added the -M command line option to pcretest.
913
914 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
915
916 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
917
918 16. Added comments and documentation about mis-use of no_arg in the C++
919 wrapper.
920
921 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
922 from Martin Jerabek that uses macro names for all relevant character and
923 string constants.
924
925 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
926 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
927 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
928 these, but not everybody uses configure.
929
930 19. A conditional group that had only one branch was not being correctly
931 recognized as an item that could match an empty string. This meant that an
932 enclosing group might also not be so recognized, causing infinite looping
933 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
934 with the subject "ab", where knowledge that the repeated group can match
935 nothing is needed in order to break the loop.
936
937 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
938 exec(), but without supplying a callout function, matching went wrong.
939
940 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
941 leak if the size of the offset vector was greater than 30. When the vector
942 is smaller, the saved offsets during recursion go onto a local stack
943 vector, but for larger vectors malloc() is used. It was failing to free
944 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
945 error, in fact).
946
947 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
948 heapframe that is used only when UTF-8 support is enabled. This caused no
949 problem, but was untidy.
950
951 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
952 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
953 included within another project.
954
955 24. Steven Van Ingelgem's patches to add more options to the CMake support,
956 slightly modified by me:
957
958 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
959 not building pcregrep.
960
961 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
962 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
963
964 25. Forward references, both numeric and by name, in patterns that made use of
965 duplicate group numbers, could behave incorrectly or give incorrect errors,
966 because when scanning forward to find the reference group, PCRE was not
967 taking into account the duplicate group numbers. A pattern such as
968 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
969
970 26. Changed a few more instances of "const unsigned char *" to USPTR, making
971 the feature of a custom pointer more persuasive (as requested by a user).
972
973 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
974 pcretest.c, inside #ifndefs, because it seems they are sometimes already
975 pre-defined.
976
977 28. Added support for (*UTF8) at the start of a pattern.
978
979 29. Arrange for flags added by the "release type" setting in CMake to be shown
980 in the configuration summary.
981
982
983 Version 7.8 05-Sep-08
984 ---------------------
985
986 1. Replaced UCP searching code with optimized version as implemented for Ad
987 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
988 stage table and inline lookup instead of a function, giving speed ups of 2
989 to 5 times on some simple patterns that I tested. Permission was given to
990 distribute the MultiStage2.py script that generates the tables (it's not in
991 the tarball, but is in the Subversion repository).
992
993 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
994 scripts.
995
996 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
997 a group with a zero qualifier. The result of the study could be incorrect,
998 or the function might crash, depending on the pattern.
999
1000 4. Caseless matching was not working for non-ASCII characters in back
1001 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1002 It now works when Unicode Property Support is available.
1003
1004 5. In pcretest, an escape such as \x{de} in the data was always generating
1005 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1006 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1007 truncation.
1008
1009 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1010
1011 7. Added two (int) casts to pcregrep when printing the difference of two
1012 pointers, in case they are 64-bit values.
1013
1014 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
1015 test 2 if it fails.
1016
1017 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1018 and a #define of that name to empty if it is not externally set. This is to
1019 allow users of MSVC to set it if necessary.
1020
1021 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1022 the convenience functions in the pcre_get.c source file.
1023
1024 11. An option change at the start of a pattern that had top-level alternatives
1025 could cause overwriting and/or a crash. This command provoked a crash in
1026 some environments:
1027
1028 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1029
1030 This potential security problem was recorded as CVE-2008-2371.
1031
1032 12. For a pattern where the match had to start at the beginning or immediately
1033 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1034 pcre_dfa_exec() could read past the end of the passed subject if there was
1035 no match. To help with detecting such bugs (e.g. with valgrind), I modified
1036 pcretest so that it places the subject at the end of its malloc-ed buffer.
1037
1038 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1039 exec() might read past the end of the data buffer in UTF-8 mode.
1040
1041 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1042 the data contained the byte 0x85 as part of a UTF-8 character within its
1043 first line. This applied both to normal and DFA matching.
1044
1045 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1046 /^[^d]*?$/8 failed to match "abc".
1047
1048 16. Added a missing copyright notice to pcrecpp_internal.h.
1049
1050 17. Make it more clear in the documentation that values returned from
1051 pcre_exec() in ovector are byte offsets, not character counts.
1052
1053 18. Tidied a few places to stop certain compilers from issuing warnings.
1054
1055 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1056 supplied by Stefan Weber. I made a further small update for 7.8 because
1057 there is a change of source arrangements: the pcre_searchfuncs.c module is
1058 replaced by pcre_ucd.c.
1059
1060
1061 Version 7.7 07-May-08
1062 ---------------------
1063
1064 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
1065 a string to a long long, pretend we don't even have a long long." This is
1066 done by checking for the strtoq, strtoll, and _strtoi64 functions.
1067
1068 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1069 pre-7.6 versions, which defined a global no_arg variable instead of putting
1070 it in the RE class. (See also #8 below.)
1071
1072 3. Remove a line of dead code, identified by coverity and reported by Nuno
1073 Lopes.
1074
1075 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
1076
1077 (1) The include/exclude patterns were being applied to the whole pathnames
1078 of files, instead of just to the final components.
1079
1080 (2) If there was more than one level of directory, the subdirectories were
1081 skipped unless they satisfied the include/exclude conditions. This is
1082 inconsistent with GNU grep (and could even be seen as contrary to the
1083 pcregrep specification - which I improved to make it absolutely clear).
1084 The action now is always to scan all levels of directory, and just
1085 apply the include/exclude patterns to regular files.
1086
1087 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1088 --exclude_dir in the tests to avoid scanning .svn directories.
1089
1090 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1091 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1092 doesn't support NULs in patterns.
1093
1094 7. Added some missing "const"s to declarations of static tables in
1095 pcre_compile.c and pcre_dfa_exec.c.
1096
1097 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1098 caused by fix #2 above. (Subsequently also a second patch to fix the
1099 first patch. And a third patch - this was a messy problem.)
1100
1101 9. Applied Craig's patch to remove the use of push_back().
1102
1103 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1104 matching function regexec().
1105
1106 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1107 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1108 references. PCRE supports relative numbers with this syntax (I don't think
1109 Oniguruma does).
1110
1111 12. Previously, a group with a zero repeat such as (...){0} was completely
1112 omitted from the compiled regex. However, this means that if the group
1113 was called as a subroutine from elsewhere in the pattern, things went wrong
1114 (an internal error was given). Such groups are now left in the compiled
1115 pattern, with a new opcode that causes them to be skipped at execution
1116 time.
1117
1118 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1119 to the way PCRE behaves:
1120
1121 (a) A lone ] character is dis-allowed (Perl treats it as data).
1122
1123 (b) A back reference to an unmatched subpattern matches an empty string
1124 (Perl fails the current match path).
1125
1126 (c) A data ] in a character class must be notated as \] because if the
1127 first data character in a class is ], it defines an empty class. (In
1128 Perl it is not possible to have an empty class.) The empty class []
1129 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1130 The negative empty class [^] matches any one character, independently
1131 of the DOTALL setting.
1132
1133 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1134 non-existent subpattern following a character class starting with ']' and
1135 containing () gave an internal compiling error instead of "reference to
1136 non-existent subpattern". Fortunately, when the pattern did exist, the
1137 compiled code was correct. (When scanning forwards to check for the
1138 existencd of the subpattern, it was treating the data ']' as terminating
1139 the class, so got the count wrong. When actually compiling, the reference
1140 was subsequently set up correctly.)
1141
1142 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1143 it was being rejected as not supported by pcre_dfa_exec(), even though
1144 other assertions are supported. I have made pcre_dfa_exec() support
1145 (*FAIL).
1146
1147 16. The implementation of 13c above involved the invention of a new opcode,
1148 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1149 cannot be changed at match time, I realized I could make a small
1150 improvement to matching performance by compiling OP_ALLANY instead of
1151 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1152 on the OP_ANY path.
1153
1154 17. Compiling pcretest on Windows with readline support failed without the
1155 following two fixes: (1) Make the unistd.h include conditional on
1156 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1157
1158 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1159 ncurses library to be included for pcretest when ReadLine support is
1160 requested, but also to allow for it to be overridden. This patch came from
1161 Daniel Bergström.
1162
1163 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1164 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1165 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1166 spotting this.
1167
1168
1169 Version 7.6 28-Jan-08
1170 ---------------------
1171
1172 1. A character class containing a very large number of characters with
1173 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1174 overflow.
1175
1176 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1177 HAVE_LONG_LONG is not defined.
1178
1179 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1180 bring them up to date and include new features. This patch includes:
1181
1182 - Fixed PH's badly added libz and libbz2 support.
1183 - Fixed a problem with static linking.
1184 - Added pcredemo. [But later removed - see 7 below.]
1185 - Fixed dftables problem and added an option.
1186 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1187 HAVE_LONG_LONG.
1188 - Added readline support for pcretest.
1189 - Added an listing of the option settings after cmake has run.
1190
1191 4. A user submitted a patch to Makefile that makes it easy to create
1192 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1193 Makefile.am that cause it to include this special target, without
1194 affecting anything else. Note that the same mingw target plus all
1195 the other distribution libraries and programs are now supported
1196 when configuring with CMake (see 6 below) instead of with
1197 Configure/Make.
1198
1199 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1200 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1201 exported in the Windows port". It has not yet been confirmed that the patch
1202 solves the problem, but it does no harm.
1203
1204 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1205 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1206 with CMake, and also correct the comment about stack recursion.
1207
1208 7. Remove the automatic building of pcredemo from the ./configure system and
1209 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1210 of a program that users should build themselves after PCRE is installed, so
1211 building it automatically is not really right. What is more, it gave
1212 trouble in some build environments.
1213
1214 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1215
1216
1217 Version 7.5 10-Jan-08
1218 ---------------------
1219
1220 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1221 values in parens when parsing an RE using the C++ wrapper."
1222
1223 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1224 Characters greater than 255 were excluded from the class instead of being
1225 included.
1226
1227 3. The same bug as (2) above applied to negated POSIX classes such as
1228 [:^space:].
1229
1230 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1231 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1232 I have changed it.
1233
1234 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1235 first named subpattern) and a construct such as (?&a) would reference the
1236 first named subpattern whose name started with "a" (in other words, the
1237 length check was missing). Both these problems are fixed. "Subpattern name
1238 expected" is now given for (?&) (a zero-length name), and this patch also
1239 makes it give the same error for \k'' (previously it complained that that
1240 was a reference to a non-existent subpattern).
1241
1242 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1243 this is right because (?- can be followed by option settings as well as by
1244 digits. I have, however, made the messages clearer.
1245
1246 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1247 than the number used in the conditional) now cause a compile-time error.
1248 This is actually not compatible with Perl, which accepts such patterns, but
1249 treats the conditional as always being FALSE (as PCRE used to), but it
1250 seems to me that giving a diagnostic is better.
1251
1252 8. Change "alphameric" to the more common word "alphanumeric" in comments
1253 and messages.
1254
1255 9. Fix two occurrences of "backslash" in comments that should have been
1256 "backspace".
1257
1258 10. Remove two redundant lines of code that can never be obeyed (their function
1259 was moved elsewhere).
1260
1261 11. The program that makes PCRE's Unicode character property table had a bug
1262 which caused it to generate incorrect table entries for sequences of
1263 characters that have the same character type, but are in different scripts.
1264 It amalgamated them into a single range, with the script of the first of
1265 them. In other words, some characters were in the wrong script. There were
1266 thirteen such cases, affecting characters in the following ranges:
1267
1268 U+002b0 - U+002c1
1269 U+0060c - U+0060d
1270 U+0061e - U+00612
1271 U+0064b - U+0065e
1272 U+0074d - U+0076d
1273 U+01800 - U+01805
1274 U+01d00 - U+01d77
1275 U+01d9b - U+01dbf
1276 U+0200b - U+0200f
1277 U+030fc - U+030fe
1278 U+03260 - U+0327f
1279 U+0fb46 - U+0fbb1
1280 U+10450 - U+1049d
1281
1282 12. The -o option (show only the matching part of a line) for pcregrep was not
1283 compatible with GNU grep in that, if there was more than one match in a
1284 line, it showed only the first of them. It now behaves in the same way as
1285 GNU grep.
1286
1287 13. If the -o and -v options were combined for pcregrep, it printed a blank
1288 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1289 does the same. The return code can be used to tell if there were any
1290 non-matching lines.
1291
1292 14. Added --file-offsets and --line-offsets to pcregrep.
1293
1294 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1295 infinitely looping recursion. The bug was that positive lookaheads were not
1296 being skipped when checking for a possible empty match (negative lookaheads
1297 and both kinds of lookbehind were skipped).
1298
1299 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1300 inclusion of <windows.h> to before rather than after the definition of
1301 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1302
1303 17. Specifying a possessive quantifier with a specific limit for a Unicode
1304 character property caused pcre_compile() to compile bad code, which led at
1305 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1306 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1307 caused the error; without that there was no problem.
1308
1309 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1310
1311 19. Added --enable-pcretest-libreadline.
1312
1313 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1314 RE::GlobalReplace(). As a result, the number of replacements returned was
1315 double what it should be. I removed one of the increments, but Craig sent a
1316 later patch that removed the other one (the right fix) and added unit tests
1317 that check the return values (which was not done before).
1318
1319 21. Several CMake things:
1320
1321 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1322 the names libpcre and libpcreposix, not just pcre and pcreposix.
1323
1324 (2) The above change means that pcretest and pcregrep are now correctly
1325 linked with the newly-built libraries, not previously installed ones.
1326
1327 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1328
1329 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1330 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1331 UTF-8 newline character). The key issue is that the pattern starts .*;
1332 this means that the match must be either at the beginning, or after a
1333 newline. The bug was in the code for advancing after a failed match and
1334 checking that the new position followed a newline. It was not taking
1335 account of UTF-8 characters correctly.
1336
1337 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1338 character classes. PCRE was not treating the sequence [:...:] as a
1339 character class unless the ... were all letters. Perl, however, seems to
1340 allow any characters between [: and :], though of course it rejects as
1341 unknown any "names" that contain non-letters, because all the known class
1342 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1343 for example, whereas PCRE did not - it did not recognize a POSIX character
1344 class. This seemed a bit dangerous, so the code has been changed to be
1345 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1346 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1347 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1348 Perl does, and where it didn't before.
1349
1350 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1351 Windows environments %n is disabled by default.
1352
1353
1354 Version 7.4 21-Sep-07
1355 ---------------------
1356
1357 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1358 means that a class such as [\s] counted as "explicit reference to CR or
1359 LF". That isn't really right - the whole point of the change was to try to
1360 help when there was an actual mention of one of the two characters. So now
1361 the change happens only if \r or \n (or a literal CR or LF) character is
1362 encountered.
1363
1364 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1365 of both had grown to the point where there were only 3 bits left.
1366 Fortunately, there was spare space in the data structure, and so I have
1367 moved the internal flags into a new 16-bit field to free up more option
1368 bits.
1369
1370 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1371 but did not set the internal JCHANGED flag - either of these is enough to
1372 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1373 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1374 start sets both bits.
1375
1376 4. Added options (at build time, compile time, exec time) to change \R from
1377 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1378
1379 5. doc/pcresyntax.html was missing from the distribution.
1380
1381 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1382 compatibility, even though it is no longer used.
1383
1384 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1385 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1386 windows.h file is present (where different names are used). [This was
1387 reversed later after testing - see 16 below.]
1388
1389 8. Changed all #include <config.h> to #include "config.h". There were also
1390 some further <pcre.h> cases that I changed to "pcre.h".
1391
1392 9. When pcregrep was used with the --colour option, it missed the line ending
1393 sequence off the lines that it output.
1394
1395 10. It was pointed out to me that arrays of string pointers cause lots of
1396 relocations when a shared library is dynamically loaded. A technique of
1397 using a single long string with a table of offsets can drastically reduce
1398 these. I have refactored PCRE in four places to do this. The result is
1399 dramatic:
1400
1401 Originally: 290
1402 After changing UCP table: 187
1403 After changing error message table: 43
1404 After changing table of "verbs" 36
1405 After changing table of Posix names 22
1406
1407 Thanks to the folks working on Gregex for glib for this insight.
1408
1409 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1410 unicode-properties was also set.
1411
1412 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1413
1414 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1415 checked only for CRLF.
1416
1417 14. Added casts to pcretest.c to avoid compiler warnings.
1418
1419 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1420
1421 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1422 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1423 entirely. This removes changes made in 7 above.
1424
1425 17. The CMake files have been updated, and there is now more information about
1426 building with CMake in the NON-UNIX-USE document.
1427
1428
1429 Version 7.3 28-Aug-07
1430 ---------------------
1431
1432 1. In the rejigging of the build system that eventually resulted in 7.1, the
1433 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1434 brackets there is not right, since it causes compilers to look for an
1435 installed pcre.h, not the version that is in the source that is being
1436 compiled (which of course may be different). I have changed it back to:
1437
1438 #include "pcre.h"
1439
1440 I have a vague recollection that the change was concerned with compiling in
1441 different directories, but in the new build system, that is taken care of
1442 by the VPATH setting the Makefile.
1443
1444 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1445 when the subject happened to end in the byte 0x85 (e.g. if the last
1446 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1447 characters but of course it shouldn't be taken as a newline when it is part
1448 of another character. The bug was that, for an unlimited repeat of . in
1449 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1450 characters when looking for a newline.
1451
1452 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1453
1454 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1455 in debug output.
1456
1457 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1458 long printing in the pcrecpp unittest when running under MinGW.
1459
1460 6. ESC_K was left out of the EBCDIC table.
1461
1462 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1463 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1464 limit also applies to "virtual nesting" when a pattern is recursive, and in
1465 this case 1000 isn't so big. I have been able to remove this limit at the
1466 expense of backing off one optimization in certain circumstances. Normally,
1467 when pcre_exec() would call its internal match() function recursively and
1468 immediately return the result unconditionally, it uses a "tail recursion"
1469 feature to save stack. However, when a subpattern that can match an empty
1470 string has an unlimited repetition quantifier, it no longer makes this
1471 optimization. That gives it a stack frame in which to save the data for
1472 checking that an empty string has been matched. Previously this was taken
1473 from the 1000-entry workspace that had been reserved. So now there is no
1474 explicit limit, but more stack is used.
1475
1476 8. Applied Daniel's patches to solve problems with the import/export magic
1477 syntax that is required for Windows, and which was going wrong for the
1478 pcreposix and pcrecpp parts of the library. These were overlooked when this
1479 problem was solved for the main library.
1480
1481 9. There were some crude static tests to avoid integer overflow when computing
1482 the size of patterns that contain repeated groups with explicit upper
1483 limits. As the maximum quantifier is 65535, the maximum group length was
1484 set at 30,000 so that the product of these two numbers did not overflow a
1485 32-bit integer. However, it turns out that people want to use groups that
1486 are longer than 30,000 bytes (though not repeat them that many times).
1487 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1488 made it possible to implement the integer overflow checks in a much more
1489 dynamic way, which I have now done. The artificial limitation on group
1490 length has been removed - we now have only the limit on the total length of
1491 the compiled pattern, which depends on the LINK_SIZE setting.
1492
1493 10. Fixed a bug in the documentation for get/copy named substring when
1494 duplicate names are permitted. If none of the named substrings are set, the
1495 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1496 empty string.
1497
1498 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1499 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1500 because the ] is interpreted as the first data character and the
1501 terminating ] is not found. PCRE has been made compatible with Perl in this
1502 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1503 cause memory overwriting.
1504
1505 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1506 string has been matched (to stop an infinite loop). It was not recognizing
1507 a conditional subpattern that could match an empty string if that
1508 subpattern was within another subpattern. For example, it looped when
1509 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1510 condition was not nested. This bug has been fixed.
1511
1512 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1513 past the start of the subject in the presence of bytes with the top bit
1514 set, for example "\x8aBCD".
1515
1516 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1517 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1518
1519 14. Optimized (?!) to (*FAIL).
1520
1521 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1522 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1523 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1524 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1525 does: it's just the validity check that is more restrictive.
1526
1527 16. Inserted checks for integer overflows during escape sequence (backslash)
1528 processing, and also fixed erroneous offset values for syntax errors during
1529 backslash processing.
1530
1531 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1532 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1533
1534 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1535 caused an overrun.
1536
1537 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1538 something other than just ASCII characters) inside a group that had an
1539 unlimited repeat caused a loop at compile time (while checking to see
1540 whether the group could match an empty string).
1541
1542 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1543 [\P{Any}] did so. (Error in the code for printing property names.)
1544
1545 21. An orphan \E inside a character class could cause a crash.
1546
1547 22. A repeated capturing bracket such as (A)? could cause a wild memory
1548 reference during compilation.
1549
1550 23. There are several functions in pcre_compile() that scan along a compiled
1551 expression for various reasons (e.g. to see if it's fixed length for look
1552 behind). There were bugs in these functions when a repeated \p or \P was
1553 present in the pattern. These operators have additional parameters compared
1554 with \d, etc, and these were not being taken into account when moving along
1555 the compiled data. Specifically:
1556
1557 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1558 length.
1559
1560 (b) An item such as \pL+ within a repeated group could cause crashes or
1561 loops.
1562
1563 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1564 "reference to non-existent subpattern" error.
1565
1566 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1567
1568 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1569 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1570
1571 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1572
1573 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1574 character were causing crashes (broken optimization).
1575
1576 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1577 \p or \P) caused a compile-time loop.
1578
1579 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1580 break. For example, the unstudied pattern [\r\n]A does not match the string
1581 "\r\nA" because change 7.0/46 below moves the current point on by two
1582 characters after failing to match at the start. However, the pattern \nA
1583 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1584 the same is true. There doesn't seem any very clean way out of this, but
1585 what I have chosen to do makes the common cases work: PCRE now takes note
1586 of whether there can be an explicit match for \r or \n anywhere in the
1587 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1588 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1589 pattern has explicit CR or LF references.
1590
1591 29. Added (*CR) etc for changing newline setting at start of pattern.
1592
1593
1594 Version 7.2 19-Jun-07
1595 ---------------------
1596
1597 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1598 which is apparently normally available under Windows.
1599
1600 2. Re-jig the pcregrep tests with different newline settings in an attempt
1601 to make them independent of the local environment's newline setting.
1602
1603 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1604
1605 4. Some of the "internals" tests were previously cut out when the link size
1606 was not 2, because the output contained actual offsets. The recent new
1607 "Z" feature of pcretest means that these can be cut out, making the tests
1608 usable with all link sizes.
1609
1610 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1611 stack recursion. This gives a massive performance boost under BSD, but just
1612 a small improvement under Linux. However, it saves one field in the frame
1613 in all cases.
1614
1615 6. Added more features from the forthcoming Perl 5.10:
1616
1617 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1618 recursion call. It refers to the nth most recently opened parentheses.
1619
1620 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1621 to be opened parentheses.
1622
1623 (c) Conditions that refer to capturing parentheses can be specified
1624 relatively, for example, (?(-2)... or (?(+3)...
1625
1626 (d) \K resets the start of the current match so that everything before
1627 is not part of it.
1628
1629 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1630
1631 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1632 reference syntax.
1633
1634 (g) (?| introduces a group in which the numbering of parentheses in each
1635 alternative starts with the same number.
1636
1637 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1638
1639 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1640 PCRE_INFO_JCHANGED.
1641
1642 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1643 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1644 for detecting groups that can match an empty string.
1645
1646 9. A pattern with a very large number of alternatives (more than several
1647 hundred) was running out of internal workspace during the pre-compile
1648 phase, where pcre_compile() figures out how much memory will be needed. A
1649 bit of new cunning has reduced the workspace needed for groups with
1650 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1651 workspace instead of running out of the 4096 that are available.
1652
1653 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1654
1655 11. Applied patch from Google to remove an optimization that didn't quite work.
1656 The report of the bug said:
1657
1658 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1659 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1660 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1661
1662 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1663 it matched the wrong number of bytes.
1664
1665
1666 Version 7.1 24-Apr-07
1667 ---------------------
1668
1669 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1670 that is more "standard", making use of automake and other Autotools. There
1671 is some re-arrangement of the files and adjustment of comments consequent
1672 on this.
1673
1674 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1675 for recursive directory scanning broke on some systems because the files
1676 are not scanned in any specific order and on different systems the order
1677 was different. A call to "sort" has been inserted into RunGrepTest for the
1678 approprate test as a short-term fix. In the longer term there may be an
1679 alternative.
1680
1681 3. I had an email from Eric Raymond about problems translating some of PCRE's
1682 man pages to HTML (despite the fact that I distribute HTML pages, some
1683 people do their own conversions for various reasons). The problems
1684 concerned the use of low-level troff macros .br and .in. I have therefore
1685 removed all such uses from the man pages (some were redundant, some could
1686 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1687 HTML has been updated to handle .nf/.fi and to complain if it encounters
1688 .br or .in.
1689
1690 4. Updated comments in configure.ac that get placed in config.h.in and also
1691 arranged for config.h to be included in the distribution, with the name
1692 config.h.generic, for the benefit of those who have to compile without
1693 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1694
1695 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1696 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1697 makevp.bat for the current PCRE, using the additional files
1698 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1699
1700 6. A Windows user reported a minor discrepancy with test 2, which turned out
1701 to be caused by a trailing space on an input line that had got lost in his
1702 copy. The trailing space was an accident, so I've just removed it.
1703
1704 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1705 that is needed.
1706
1707 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1708 as "const" (a) because they are and (b) because it helps the PHP
1709 maintainers who have recently made a script to detect big data structures
1710 in the php code that should be moved to the .rodata section. I remembered
1711 to update Builducptable as well, so it won't revert if ucptable.h is ever
1712 re-created.
1713
1714 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1715 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1716 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1717 support is not required. This saves 1.5-2K of code, which is important in
1718 some applications.
1719
1720 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1721 so as not to refer to the tables, even though these functions will never be
1722 called when UTF-8 support is disabled. Otherwise there are problems with a
1723 shared library.
1724
1725 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1726
1727 (a) It was defining its arguments as char * instead of void *.
1728
1729 (b) It was assuming that all moves were upwards in memory; this was true
1730 a long time ago when I wrote it, but is no longer the case.
1731
1732 The emulated memove() is provided for those environments that have neither
1733 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1734 is clearly not the case, as these two bugs were recently reported.
1735
1736 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1737 and Detrail to create the HTML documentation, the .txt form of the man
1738 pages, and it removes trailing spaces from listed files. It also creates
1739 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1740 case, it wraps all the #defines with #ifndefs. This script should be run
1741 before "make dist".
1742
1743 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1744 with Unicode property support.
1745
1746 (a) For a maximizing quantifier, if the two different cases of the
1747 character were of different lengths in their UTF-8 codings (there are
1748 some cases like this - I found 11), and the matching function had to
1749 back up over a mixture of the two cases, it incorrectly assumed they
1750 were both the same length.
1751
1752 (b) When PCRE was configured to use the heap rather than the stack for
1753 recursion during matching, it was not correctly preserving the data for
1754 the other case of a UTF-8 character when checking ahead for a match
1755 while processing a minimizing repeat. If the check also involved
1756 matching a wide character, but failed, corruption could cause an
1757 erroneous result when trying to check for a repeat of the original
1758 character.
1759
1760 13. Some tidying changes to the testing mechanism:
1761
1762 (a) The RunTest script now detects the internal link size and whether there
1763 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1764 values substituted by "configure". (The RunGrepTest script already did
1765 this for UTF-8.) The configure.ac script no longer substitutes the
1766 relevant variables.
1767
1768 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1769 with length and offset values. This means that the output is different
1770 for different internal link sizes. Test 2 is skipped for link sizes
1771 other than 2 because of this, bypassing the problem. Unfortunately,
1772 there was also a test in test 3 (the locale tests) that used /B and
1773 failed for link sizes other than 2. Rather than cut the whole test out,
1774 I have added a new /Z option to pcretest that replaces the length and
1775 offset values with spaces. This is now used to make test 3 independent
1776 of link size. (Test 2 will be tidied up later.)
1777
1778 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1779 segmentation fault instead of returning the appropriate error message.
1780
1781 15. In multiline mode when the newline sequence was set to "any", the pattern
1782 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1783 This doesn't seem right; it now treats the CRLF combination as the line
1784 ending, and so does not match in that case. It's only a pattern such as ^$
1785 that would hit this one: something like ^ABC$ would have failed after \r
1786 and then tried again after \r\n.
1787
1788 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1789 in an attempt to make files that differ only in their line terminators
1790 compare equal. This works on Linux.
1791
1792 17. Under certain error circumstances pcregrep might try to free random memory
1793 as it exited. This is now fixed, thanks to valgrind.
1794
1795 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1796 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1797 was because its rules for how to advance for /g after matching an empty
1798 string at the end of a line did not allow for this case. They now check for
1799 it specially.
1800
1801 20. pcretest is supposed to handle patterns and data of any length, by
1802 extending its buffers when necessary. It was getting this wrong when the
1803 buffer for a data line had to be extended.
1804
1805 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1806 CRLF as a newline sequence.
1807
1808 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1809 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1810 I have nevertheless tidied it up.
1811
1812 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1813
1814 24. Added a man page for pcre-config.
1815
1816
1817 Version 7.0 19-Dec-06
1818 ---------------------
1819
1820 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1821 moving to gcc 4.1.1.
1822
1823 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1824 sys/time.h, which is documented as needed for this function. It doesn't
1825 seem to matter on Linux, but it showed up on some releases of OS X.
1826
1827 3. It seems that there are systems where bytes whose values are greater than
1828 127 match isprint() in the "C" locale. The "C" locale should be the
1829 default when a C program starts up. In most systems, only ASCII printing
1830 characters match isprint(). This difference caused the output from pcretest
1831 to vary, making some of the tests fail. I have changed pcretest so that:
1832
1833 (a) When it is outputting text in the compiled version of a pattern, bytes
1834 other than 32-126 are always shown as hex escapes.
1835
1836 (b) When it is outputting text that is a matched part of a subject string,
1837 it does the same, unless a different locale has been set for the match
1838 (using the /L modifier). In this case, it uses isprint() to decide.
1839
1840 4. Fixed a major bug that caused incorrect computation of the amount of memory
1841 required for a compiled pattern when options that changed within the
1842 pattern affected the logic of the preliminary scan that determines the
1843 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1844 that the computed length was too small. The symptoms of this bug were
1845 either the PCRE error "internal error: code overflow" from pcre_compile(),
1846 or a glibc crash with a message such as "pcretest: free(): invalid next
1847 size (fast)". Examples of patterns that provoked this bug (shown in
1848 pcretest format) are:
1849
1850 /(?-x: )/x
1851 /(?x)(?-x: \s*#\s*)/
1852 /((?i)[\x{c0}])/8
1853 /(?i:[\x{c0}])/8
1854
1855 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1856 is now done differently.
1857
1858 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1859 wrapper classes; (b) implement a new function in the C++ scanner that is
1860 more efficient than the old way of doing things because it avoids levels of
1861 recursion in the regex matching; (c) add a paragraph to the documentation
1862 for the FullMatch() function.
1863
1864 6. The escape sequence \n was being treated as whatever was defined as
1865 "newline". Not only was this contrary to the documentation, which states
1866 that \n is character 10 (hex 0A), but it also went horribly wrong when
1867 "newline" was defined as CRLF. This has been fixed.
1868
1869 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1870 was being set to -1 for the "end of line" case (supposedly a value that no
1871 character can have). Though this value is never used (the check for end of
1872 line is "zero bytes in current character"), it caused compiler complaints.
1873 I've changed it to 0xffffffff.
1874
1875 8. In pcre_version.c, the version string was being built by a sequence of
1876 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1877 string (as it is for production releases) called a macro with an empty
1878 argument. The C standard says the result of this is undefined. The gcc
1879 compiler treats it as an empty string (which was what was wanted) but it is
1880 reported that Visual C gives an error. The source has been hacked around to
1881 avoid this problem.
1882
1883 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1884 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1885 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1886 of them did).
1887
1888 10. Originally, pcretest opened its input and output without "b"; then I was
1889 told that "b" was needed in some environments, so it was added for release
1890 5.0 to both the input and output. (It makes no difference on Unix-like
1891 systems.) Later I was told that it is wrong for the input on Windows. I've
1892 now abstracted the modes into two macros, to make it easier to fiddle with
1893 them, and removed "b" from the input mode under Windows.
1894
1895 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1896
1897 12. Added -help and --help to pcretest as an official way of being reminded
1898 of the options.
1899
1900 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1901 and pcrecpp.cc because they annoy compilers at high warning levels.
1902
1903 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1904
1905 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1906 scripts are not C programs :-) and which was not noticed because it works
1907 on Linux.
1908
1909 16. pcretest is supposed to handle any length of pattern and data line (as one
1910 line or as a continued sequence of lines) by extending its input buffer if
1911 necessary. This feature was broken for very long pattern lines, leading to
1912 a string of junk being passed to pcre_compile() if the pattern was longer
1913 than about 50K.
1914
1915 17. I have done a major re-factoring of the way pcre_compile() computes the
1916 amount of memory needed for a compiled pattern. Previously, there was code
1917 that made a preliminary scan of the pattern in order to do this. That was
1918 OK when PCRE was new, but as the facilities have expanded, it has become
1919 harder and harder to keep it in step with the real compile phase, and there
1920 have been a number of bugs (see for example, 4 above). I have now found a
1921 cunning way of running the real compile function in a "fake" mode that
1922 enables it to compute how much memory it would need, while actually only
1923 ever using a few hundred bytes of working memory and without too many
1924 tests of the mode. This should make future maintenance and development
1925 easier. A side effect of this work is that the limit of 200 on the nesting
1926 depth of parentheses has been removed (though this was never a serious
1927 limitation, I suspect). However, there is a downside: pcre_compile() now
1928 runs more slowly than before (30% or more, depending on the pattern). I
1929 hope this isn't a big issue. There is no effect on runtime performance.
1930
1931 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1932 newline (only possible for the last line of a file) and it was a
1933 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1934
1935 19. Added additional timing features to pcretest. (1) The -tm option now times
1936 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1937 separate command line item, by a number that specifies the number of
1938 repeats to use when timing. The default is 50000; this gives better
1939 precision, but takes uncomfortably long for very large patterns.
1940
1941 20. Extended pcre_study() to be more clever in cases where a branch of a
1942 subpattern has no definite first character. For example, (a*|b*)[cd] would
1943 previously give no result from pcre_study(). Now it recognizes that the
1944 first character must be a, b, c, or d.
1945
1946 21. There was an incorrect error "recursive call could loop indefinitely" if
1947 a subpattern (or the entire pattern) that was being tested for matching an
1948 empty string contained only one non-empty item after a nested subpattern.
1949 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1950 incorrectly, because the \d was being skipped in the check.
1951
1952 22. The pcretest program now has a new pattern option /B and a command line
1953 option -b, which is equivalent to adding /B to every pattern. This causes
1954 it to show the compiled bytecode, without the additional information that
1955 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1956 is the same as /B/I).
1957
1958 23. A new optimization is now able automatically to treat some sequences such
1959 as a*b as a*+b. More specifically, if something simple (such as a character
1960 or a simple class like \d) has an unlimited quantifier, and is followed by
1961 something that cannot possibly match the quantified thing, the quantifier
1962 is automatically "possessified".
1963
1964 24. A recursive reference to a subpattern whose number was greater than 39
1965 went wrong under certain circumstances in UTF-8 mode. This bug could also
1966 have affected the operation of pcre_study().
1967
1968 25. Realized that a little bit of performance could be had by replacing
1969 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1970
1971 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1972
1973 27. Possessive quantifiers such as a++ were previously implemented by turning
1974 them into atomic groups such as ($>a+). Now they have their own opcodes,
1975 which improves performance. This includes the automatically created ones
1976 from 23 above.
1977
1978 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1979 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1980 the first matched character to be a colon. This applied both to named and
1981 numbered groups.
1982
1983 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1984
1985 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1986 building PCRE on Windows easier, so I have included it in the distribution.
1987
1988 31. There is now a check in pcretest against a ridiculously large number being
1989 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1990 loop, the loop is abandoned.
1991
1992 32. Forward references to subpatterns in conditions such as (?(2)...) where
1993 subpattern 2 is defined later cause pcre_compile() to search forwards in
1994 the pattern for the relevant set of parentheses. This search went wrong
1995 when there were unescaped parentheses in a character class, parentheses
1996 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1997
1998 33. "Subroutine" calls and backreferences were previously restricted to
1999 referencing subpatterns earlier in the regex. This restriction has now
2000 been removed.
2001
2002 34. Added a number of extra features that are going to be in Perl 5.10. On the
2003 whole, these are just syntactic alternatives for features that PCRE had
2004 previously implemented using the Python syntax or my own invention. The
2005 other formats are all retained for compatibility.
2006
2007 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2008 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2009 also .NET compatible.
2010
2011 (b) A recursion or subroutine call to a named group can now be defined as
2012 (?&name) as well as (?P>name).
2013
2014 (c) A backreference to a named group can now be defined as \k<name> or
2015 \k'name' as well as (?P=name). The new forms, as well as being in Perl
2016 5.10, are also .NET compatible.
2017
2018 (d) A conditional reference to a named group can now use the syntax
2019 (?(<name>) or (?('name') as well as (?(name).
2020
2021 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2022 groups (named and numbered) that are never evaluated inline, but can be
2023 called as "subroutines" from elsewhere. In effect, the DEFINE condition
2024 is always false. There may be only one alternative in such a group.
2025
2026 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2027 as the simple (?(R). The condition is true only if the most recent
2028 recursion is that of the given number or name. It does not search out
2029 through the entire recursion stack.
2030
2031 (g) The escape \gN or \g{N} has been added, where N is a positive or
2032 negative number, specifying an absolute or relative reference.
2033
2034 35. Tidied to get rid of some further signed/unsigned compiler warnings and
2035 some "unreachable code" warnings.
2036
2037 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2038 things, this adds five new scripts.
2039
2040 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2041 There were also incompatibilities regarding the handling of \Q..\E inside
2042 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2043 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2044
2045 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2046 matches an empty string, and forcibly breaks the loop. There were bugs in
2047 this code in non-simple cases. For a pattern such as ^(a()*)* matched
2048 against aaaa the result was just "a" rather than "aaaa", for example. Two
2049 separate and independent bugs (that affected different cases) have been
2050 fixed.
2051
2052 39. Refactored the code to abolish the use of different opcodes for small
2053 capturing bracket numbers. This is a tidy that I avoided doing when I
2054 removed the limit on the number of capturing brackets for 3.5 back in 2001.
2055 The new approach is not only tidier, it makes it possible to reduce the
2056 memory needed to fix the previous bug (38).
2057
2058 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2059 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2060 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2061 mode.
2062
2063 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2064 report.
2065
2066 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2067 copy construction and assignment in the C++ wrapper.
2068
2069 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2070 couple of bugs that could have given wrong results in the "--newline=crlf"
2071 case.
2072
2073 44. Added a number of casts and did some reorganization of signed/unsigned int
2074 variables following suggestions from Dair Grant. Also renamed the variable
2075 "this" as "item" because it is a C++ keyword.
2076
2077 45. Arranged for dftables to add
2078
2079 #include "pcre_internal.h"
2080
2081 to pcre_chartables.c because without it, gcc 4.x may remove the array
2082 definition from the final binary if PCRE is built into a static library and
2083 dead code stripping is activated.
2084
2085 46. For an unanchored pattern, if a match attempt fails at the start of a
2086 newline sequence, and the newline setting is CRLF or ANY, and the next two
2087 characters are CRLF, advance by two characters instead of one.
2088
2089
2090 Version 6.7 04-Jul-06
2091 ---------------------
2092
2093 1. In order to handle tests when input lines are enormously long, pcretest has
2094 been re-factored so that it automatically extends its buffers when
2095 necessary. The code is crude, but this _is_ just a test program. The
2096 default size has been increased from 32K to 50K.
2097
2098 2. The code in pcre_study() was using the value of the re argument before
2099 testing it for NULL. (Of course, in any sensible call of the function, it
2100 won't be NULL.)
2101
2102 3. The memmove() emulation function in pcre_internal.h, which is used on
2103 systems that lack both memmove() and bcopy() - that is, hardly ever -
2104 was missing a "static" storage class specifier.
2105
2106 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2107 containing an extended class (one that cannot be represented by a bitmap
2108 because it contains high-valued characters or Unicode property items, e.g.
2109 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2110 pattern, but PCRE should not loop if you do not (it no longer does).
2111 [Detail: two cases were found: (a) a repeated subpattern containing an
2112 extended class; (b) a recursive reference to a subpattern that followed a
2113 previous extended class. It wasn't skipping over the extended class
2114 correctly when UTF-8 mode was not set.]
2115
2116 5. A negated single-character class was not being recognized as fixed-length
2117 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2118 compile error "lookbehind assertion is not fixed length".
2119
2120 6. The RunPerlTest auxiliary script was showing an unexpected difference
2121 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2122 write a Perl script that can interpret lines of an input file either as
2123 byte characters or as UTF-8, which is what "perltest" was being required to
2124 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2125 can't do is switch easily at run time between having the "use utf8;" pragma
2126 or not. In the end, I fudged it by using the RunPerlTest script to insert
2127 "use utf8;" explicitly for the UTF-8 tests.
2128
2129 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2130 the end of the subject string, contrary to the documentation and to what
2131 Perl does. This was true of both matching functions. Now it matches only at
2132 the start of the subject and immediately after *internal* newlines.
2133
2134 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2135 a pointer to an int instead of a pointer to an unsigned long int. This
2136 caused problems on 64-bit systems.
2137
2138 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2139 instance of the 'standard' template library not being so standard".
2140
2141 10. There was no check on the number of named subpatterns nor the maximum
2142 length of a subpattern name. The product of these values is used to compute
2143 the size of the memory block for a compiled pattern. By supplying a very
2144 long subpattern name and a large number of named subpatterns, the size
2145 computation could be caused to overflow. This is now prevented by limiting
2146 the length of names to 32 characters, and the number of named subpatterns
2147 to 10,000.
2148
2149 11. Subpatterns that are repeated with specific counts have to be replicated in
2150 the compiled pattern. The size of memory for this was computed from the
2151 length of the subpattern and the repeat count. The latter is limited to
2152 65535, but there was no limit on the former, meaning that integer overflow
2153 could in principle occur. The compiled length of a repeated subpattern is
2154 now limited to 30,000 bytes in order to prevent this.
2155
2156 12. Added the optional facility to have named substrings with the same name.
2157
2158 13. Added the ability to use a named substring as a condition, using the
2159 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2160 are numbers (not recommended). Forward references are permitted.
2161
2162 14. Added forward references in named backreferences (if you see what I mean).
2163
2164 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2165 pattern could run off the end of the subject. For example, the pattern
2166 "(?s)(.{1,5})"8 did this with the subject "ab".
2167
2168 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2169 PCRE_CASELESS was set when matching characters that were quantified with ?
2170 or *.
2171
2172 17. A character class other than a single negated character that had a minimum
2173 but no maximum quantifier - for example [ab]{6,} - was not handled
2174 correctly by pce_dfa_exec(). It would match only one character.
2175
2176 18. A valid (though odd) pattern that looked like a POSIX character
2177 class but used an invalid character after [ (for example [[,abc,]]) caused
2178 pcre_compile() to give the error "Failed: internal error: code overflow" or
2179 in some cases to crash with a glibc free() error. This could even happen if
2180 the pattern terminated after [[ but there just happened to be a sequence of
2181 letters, a binary zero, and a closing ] in the memory that followed.
2182
2183 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2184 over the years. Originally (before any Unicode support), just the bottom 8
2185 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2186 output from "man perlunicode" includes this:
2187
2188 The regular expression compiler produces polymorphic opcodes. That
2189 is, the pattern adapts to the data and automatically switches to
2190 the Unicode character scheme when presented with Unicode data--or
2191 instead uses a traditional byte scheme when presented with byte
2192 data.
2193
2194 Sadly, a wide octal escape does not cause a switch, and in a string with
2195 no other multibyte characters, these octal escapes are treated as before.
2196 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2197 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2198 Unicode string.
2199
2200 I have not perpetrated such confusion in PCRE. Up till now, it took just
2201 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2202 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2203 translate to the appropriate multibyte character.
2204
2205 29. Applied some refactoring to reduce the number of warnings from Microsoft
2206 and Borland compilers. This has included removing the fudge introduced
2207 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2208 a warning about an unused variable.
2209
2210 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2211 characters since release 4.0, because Perl (from release 5.004) does not.
2212 [Or at least, is documented not to: some releases seem to be in conflict
2213 with the documentation.] However, when a pattern was studied with
2214 pcre_study() and all its branches started with \s, PCRE still included VT
2215 as a possible starting character. Of course, this did no harm; it just
2216 caused an unnecessary match attempt.
2217
2218 22. Removed a now-redundant internal flag bit that recorded the fact that case
2219 dependency changed within the pattern. This was once needed for "required
2220 byte" processing, but is no longer used. This recovers a now-scarce options
2221 bit. Also moved the least significant internal flag bit to the most-
2222 significant bit of the word, which was not previously used (hangover from
2223 the days when it was an int rather than a uint) to free up another bit for
2224 the future.
2225
2226 23. Added support for CRLF line endings as well as CR and LF. As well as the
2227 default being selectable at build time, it can now be changed at runtime
2228 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2229 specify that it is scanning data with non-default line endings.
2230
2231 24. Changed the definition of CXXLINK to make it agree with the definition of
2232 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2233
2234 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2235 recursions. This makes a big different to stack usage for some patterns.
2236
2237 26. If a subpattern containing a named recursion or subroutine reference such
2238 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2239 the space required for the compiled pattern went wrong and gave too small a
2240 value. Depending on the environment, this could lead to "Failed: internal
2241 error: code overflow at offset 49" or "glibc detected double free or
2242 corruption" errors.
2243
2244 27. Applied patches from Google (a) to support the new newline modes and (b) to
2245 advance over multibyte UTF-8 characters in GlobalReplace.
2246
2247 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2248 difference for some implementation of PCRE in some Windows version.
2249
2250 29. Added some extra testing facilities to pcretest:
2251
2252 \q<number> in a data line sets the "match limit" value
2253 \Q<number> in a data line sets the "match recursion limt" value
2254 -S <number> sets the stack size, where <number> is in megabytes
2255
2256 The -S option isn't available for Windows.
2257
2258
2259 Version 6.6 06-Feb-06
2260 ---------------------
2261
2262 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2263 in pcreposix.h. I have copied the definition from pcre.h.
2264
2265 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2266 because pcre.h is no longer a built file.
2267
2268 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2269 not normally included in the compiled code.
2270
2271
2272 Version 6.5 01-Feb-06
2273 ---------------------
2274
2275 1. When using the partial match feature with pcre_dfa_exec(), it was not
2276 anchoring the second and subsequent partial matches at the new starting
2277 point. This could lead to incorrect results. For example, with the pattern
2278 /1234/, partially matching against "123" and then "a4" gave a match.
2279
2280 2. Changes to pcregrep:
2281
2282 (a) All non-match returns from pcre_exec() were being treated as failures
2283 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2284 error message is output. Some extra information is given for the
2285 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2286 probably the only errors that are likely to be caused by users (by
2287 specifying a regex that has nested indefinite repeats, for instance).
2288 If there are more than 20 of these errors, pcregrep is abandoned.
2289
2290 (b) A binary zero was treated as data while matching, but terminated the
2291 output line if it was written out. This has been fixed: binary zeroes
2292 are now no different to any other data bytes.
2293
2294 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2295 used to set a locale for matching. The --locale=xxxx long option has
2296 been added (no short equivalent) to specify a locale explicitly on the
2297 pcregrep command, overriding the environment variables.
2298
2299 (d) When -B was used with -n, some line numbers in the output were one less
2300 than they should have been.
2301
2302 (e) Added the -o (--only-matching) option.
2303
2304 (f) If -A or -C was used with -c (count only), some lines of context were
2305 accidentally printed for the final match.
2306
2307 (g) Added the -H (--with-filename) option.
2308
2309 (h) The combination of options -rh failed to suppress file names for files
2310 that were found from directory arguments.
2311
2312 (i) Added the -D (--devices) and -d (--directories) options.
2313
2314 (j) Added the -F (--fixed-strings) option.
2315
2316 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2317
2318 (l) Added the --colo(u)r option.
2319
2320 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2321 is not present by default.
2322
2323 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2324 items such as (?R) or (?1), when the recursion could match a number of
2325 alternatives. If it matched one of the alternatives, but subsequently,
2326 outside the recursion, there was a failure, the code tried to back up into
2327 the recursion. However, because of the way PCRE is implemented, this is not
2328 possible, and the result was an incorrect result from the match.
2329
2330 In order to prevent this happening, the specification of recursion has
2331 been changed so that all such subpatterns are automatically treated as
2332 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2333
2334 4. I had overlooked the fact that, in some locales, there are characters for
2335 which isalpha() is true but neither isupper() nor islower() are true. In
2336 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2337 and ordfeminine) are like this. This affected the treatment of \w and \W
2338 when they appeared in character classes, but not when they appeared outside
2339 a character class. The bit map for "word" characters is now created
2340 separately from the results of isalnum() instead of just taking it from the
2341 upper, lower, and digit maps. (Plus the underscore character, of course.)
2342
2343 5. The above bug also affected the handling of POSIX character classes such as
2344 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2345 permanent tables. Instead, the bit maps for such a class were previously
2346 created as the appropriate unions of the upper, lower, and digit bitmaps.
2347 Now they are created by subtraction from the [[:word:]] class, which has
2348 its own bitmap.
2349
2350 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2351 It is created by subtracting the vertical space characters (\x09, \x0a,
2352 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2353 subtraction was done in the overall bitmap for a character class, meaning
2354 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2355 be recognized. This bug has been fixed.
2356
2357 7. Patches from the folks at Google:
2358
2359 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2360 real life, but is still worth protecting against".
2361
2362 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2363 regular expressions".
2364
2365 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2366 have it.
2367
2368 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2369 "configure" and the latter not, in order to fix a problem somebody had
2370 with compiling the Arg class on HP-UX.
2371
2372 (e) Improve the error-handling of the C++ wrapper a little bit.
2373
2374 (f) New tests for checking recursion limiting.
2375
2376 8. The pcre_memmove() function, which is used only if the environment does not
2377 have a standard memmove() function (and is therefore rarely compiled),
2378 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2379 returning a result (though PCRE never actually uses the result).
2380
2381 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2382 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2383 returned instead of calling malloc() with an overflowing number that would
2384 most likely cause subsequent chaos.
2385
2386 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2387
2388 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2389 with this option is matched, the nmatch and pmatch options of regexec() are
2390 ignored.
2391
2392 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2393 provided in case anyone wants to the the POSIX interface with UTF-8
2394 strings.
2395
2396 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2397 C++ linking (needed for some HP-UX environments).
2398
2399 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2400 (unused parameter) and in the pcre_printint() function (omitted "default"
2401 switch label when the default is to do nothing).
2402
2403 15. Added some code to make it possible, when PCRE is compiled as a C++
2404 library, to replace subject pointers for pcre_exec() with a smart pointer
2405 class, thus making it possible to process discontinuous strings.
2406
2407 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2408 much the same function. They were added by different people who were trying
2409 to make PCRE easy to compile on non-Unix systems. It has been suggested
2410 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2411 for compiling on Windows systems. I have therefore replaced it with
2412 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2413 defaults to "extern" for C or "extern C" for C++, which works fine on
2414 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2415 SCOPE with something explicit in config.h. In addition:
2416
2417 (a) pcreposix.h still had just "extern" instead of either of these macros;
2418 I have replaced it with PCRE_DATA_SCOPE.
2419
2420 (b) Functions such as _pcre_xclass(), which are internal to the library,
2421 but external in the C sense, all had PCRE_EXPORT in their definitions.
2422 This is apparently wrong for the Windows case, so I have removed it.
2423 (It makes no difference on Unix-like systems.)
2424
2425 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2426 of recursive calls to match(). This is different to MATCH_LIMIT because
2427 that limits the total number of calls to match(), not all of which increase
2428 the depth of recursion. Limiting the recursion depth limits the amount of
2429 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2430 when PCRE is compiled, and changed at run time. A patch from Google adds
2431 this functionality to the C++ interface.
2432
2433 18. Changes to the handling of Unicode character properties:
2434
2435 (a) Updated the table to Unicode 4.1.0.
2436
2437 (b) Recognize characters that are not in the table as "Cn" (undefined).
2438
2439 (c) I revised the way the table is implemented to a much improved format
2440 which includes recognition of ranges. It now supports the ranges that
2441 are defined in UnicodeData.txt, and it also amalgamates other
2442 characters into ranges. This has reduced the number of entries in the
2443 table from around 16,000 to around 3,000, thus reducing its size
2444 considerably. I realized I did not need to use a tree structure after
2445 all - a binary chop search is just as efficient. Having reduced the
2446 number of entries, I extended their size from 6 bytes to 8 bytes to
2447 allow for more data.
2448
2449 (d) Added support for Unicode script names via properties such as \p{Han}.
2450
2451 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2452 matching that character.
2453
2454 20. When matching a repeated Unicode property with a minimum greater than zero,
2455 (for example \pL{2,}), PCRE could look past the end of the subject if it
2456 reached it while seeking the minimum number of characters. This could
2457 happen only if some of the characters were more than one byte long, because
2458 there is a check for at least the minimum number of bytes.
2459
2460 21. Refactored the implementation of \p and \P so as to be more general, to
2461 allow for more different types of property in future. This has changed the
2462 compiled form incompatibly. Anybody with saved compiled patterns that use
2463 \p or \P will have to recompile them.
2464
2465 22. Added "Any" and "L&" to the supported property types.
2466
2467 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2468 but give a compile time error if the value is greater than 0xff.
2469
2470 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2471 accidentally not being installed or uninstalled.
2472
2473 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2474 made were to insert the current release number. This seemed silly, because
2475 it made things harder for people building PCRE on systems that don't run
2476 "configure". I have turned pcre.h into a distributed file, no longer built
2477 by "configure", with the version identification directly included. There is
2478 no longer a pcre.h.in file.
2479
2480 However, this change necessitated a change to the pcre-config script as
2481 well. It is built from pcre-config.in, and one of the substitutions was the
2482 release number. I have updated configure.ac so that ./configure now finds
2483 the release number by grepping pcre.h.
2484
2485 26. Added the ability to run the tests under valgrind.
2486
2487
2488 Version 6.4 05-Sep-05
2489 ---------------------
2490
2491 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2492 "--" to be printed when multiple files were scanned, even when none of the
2493 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2494 consider it to be a bug, and have restored the previous behaviour.
2495
2496 2. A couple of code tidies to get rid of compiler warnings.
2497
2498 3. The pcretest program used to cheat by referring to symbols in the library
2499 whose names begin with _pcre_. These are internal symbols that are not
2500 really supposed to be visible externally, and in some environments it is
2501 possible to suppress them. The cheating is now confined to including
2502 certain files from the library's source, which is a bit cleaner.
2503
2504 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2505 file's purpose clearer.
2506
2507 5. Reorganized pcre_ucp_findchar().
2508
2509
2510 Version 6.3 15-Aug-05
2511 ---------------------
2512
2513 1. The file libpcre.pc.in did not have general read permission in the tarball.
2514
2515 2. There were some problems when building without C++ support:
2516
2517 (a) If C++ support was not built, "make install" and "make test" still
2518 tried to test it.
2519
2520 (b) There were problems when the value of CXX was explicitly set. Some
2521 changes have been made to try to fix these, and ...
2522
2523 (c) --disable-cpp can now be used to explicitly disable C++ support.
2524
2525 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2526 backslash in a target when C++ was disabled. This confuses some
2527 versions of "make", apparently. Using an intermediate variable solves
2528 this. (Same for CPP_LOBJ.)
2529
2530 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2531 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2532 necessary on certain architectures.
2533
2534 4. Added a setting of -export-symbols-regex to the link command to remove
2535 those symbols that are exported in the C sense, but actually are local
2536 within the library, and not documented. Their names all begin with
2537 "_pcre_". This is not a perfect job, because (a) we have to except some
2538 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2539 available (and never for static libraries). I have made a note to try to
2540 find a way round (a) in the future.
2541
2542
2543 Version 6.2 01-Aug-05
2544 ---------------------
2545
2546 1. There was no test for integer overflow of quantifier values. A construction
2547 such as {1111111111111111} would give undefined results. What is worse, if
2548 a minimum quantifier for a parenthesized subpattern overflowed and became
2549 negative, the calculation of the memory size went wrong. This could have
2550 led to memory overwriting.
2551
2552 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2553
2554 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2555 operating environments where this matters.
2556
2557 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2558 PCRE options from within the C++ wrapper.
2559
2560 5. Named capturing subpatterns were not being correctly counted when a pattern
2561 was compiled. This caused two problems: (a) If there were more than 100
2562 such subpatterns, the calculation of the memory needed for the whole
2563 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2564 back references of the form \12, where the number was greater than 9, were
2565 not recognized as back references, even though there were sufficient
2566 previous subpatterns.
2567
2568 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2569 versions of gcc, e.g. 2.95.4.
2570
2571
2572 Version 6.1 21-Jun-05
2573 ---------------------
2574
2575 1. There was one reference to the variable "posix" in pcretest.c that was not
2576 surrounded by "#if !defined NOPOSIX".
2577
2578 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2579 the cross-check on the old pcre_info() function, for the benefit of the
2580 cut-down version of PCRE that is currently imported into Exim.
2581
2582 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2583 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2584 patterns that start (?i) but I don't think that matters. The use of (?i) is
2585 just an example; this all applies to the other options as well.
2586
2587 4. Since libtool seems to echo the compile commands it is issuing, the output
2588 from "make" can be reduced a bit by putting "@" in front of each libtool
2589 compile command.
2590
2591 5. Patch from the folks at Google for configure.in to be a bit more thorough
2592 in checking for a suitable C++ installation before trying to compile the
2593 C++ stuff. This should fix a reported problem when a compiler was present,
2594 but no suitable headers.
2595
2596 6. The man pages all had just "PCRE" as their title. I have changed them to
2597 be the relevant file name. I have also arranged that these names are
2598 retained in the file doc/pcre.txt, which is a concatenation in text format
2599 of all the man pages except the little individual ones for each function.
2600
2601 7. The NON-UNIX-USE file had not been updated for the different set of source
2602 files that come with release 6. I also added a few comments about the C++
2603 wrapper.
2604
2605
2606 Version 6.0 07-Jun-05
2607 ---------------------
2608
2609 1. Some minor internal re-organization to help with my DFA experiments.
2610
2611 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2612 didn't matter for the library itself when fully configured, but did matter
2613 when compiling without UCP support, or within Exim, where the ucp files are
2614 not imported.
2615
2616 3. Refactoring of the library code to split up the various functions into
2617 different source modules. The addition of the new DFA matching code (see
2618 below) to a single monolithic source would have made it really too
2619 unwieldy, quite apart from causing all the code to be include in a
2620 statically linked application, when only some functions are used. This is
2621 relevant even without the DFA addition now that patterns can be compiled in
2622 one application and matched in another.
2623
2624 The downside of splitting up is that there have to be some external
2625 functions and data tables that are used internally in different modules of
2626 the library but which are not part of the API. These have all had their
2627 names changed to start with "_pcre_" so that they are unlikely to clash
2628 with other external names.
2629
2630 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2631 a different (DFA) algorithm. Although it is slower than the original
2632 function, it does have some advantages for certain types of matching
2633 problem.
2634
2635 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2636 including restarting after a partial match.
2637
2638 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2639 defined when compiling for Windows was sent to me. I have put it into the
2640 code, though I have no means of testing or verifying it.
2641
2642 7. Added the pcre_refcount() auxiliary function.
2643
2644 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2645 match before or at the first newline in the subject string. In pcretest,
2646 the /f option on a pattern can be used to set this.
2647
2648 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2649 would behave wrongly. This has been present in PCRE since release 4.0.
2650
2651 10. A number of changes to the pcregrep command:
2652
2653 (a) Refactored how -x works; insert ^(...)$ instead of setting
2654 PCRE_ANCHORED and checking the length, in preparation for adding
2655 something similar for -w.
2656
2657 (b) Added the -w (match as a word) option.
2658
2659 (c) Refactored the way lines are read and buffered so as to have more
2660 than one at a time available.
2661
2662 (d) Implemented a pcregrep test script.
2663
2664 (e) Added the -M (multiline match) option. This allows patterns to match
2665 over several lines of the subject. The buffering ensures that at least
2666 8K, or the rest of the document (whichever is the shorter) is available
2667 for matching (and similarly the previous 8K for lookbehind assertions).
2668
2669 (f) Changed the --help output so that it now says
2670
2671 -w, --word-regex(p)
2672
2673 instead of two lines, one with "regex" and the other with "regexp"
2674 because that confused at least one person since the short forms are the
2675 same. (This required a bit of code, as the output is generated
2676 automatically from a table. It wasn't just a text change.)
2677
2678 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2679 option but starts with a hyphen. Could be a pattern or a path name
2680 starting with a hyphen, for instance.
2681
2682 (h) "-" can be given as a file name to represent stdin.
2683
2684 (i) When file names are being printed, "(standard input)" is used for
2685 the standard input, for compatibility with GNU grep. Previously
2686 "<stdin>" was used.
2687
2688 (j) The option --label=xxx can be used to supply a name to be used for
2689 stdin when file names are being printed. There is no short form.
2690
2691 (k) Re-factored the options decoding logic because we are going to add
2692 two more options that take data. Such options can now be given in four
2693 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2694
2695 (l) Added the -A, -B, and -C options for requesting that lines of context
2696 around matches be printed.
2697
2698 (m) Added the -L option to print the names of files that do not contain
2699 any matching lines, that is, the complement of -l.
2700
2701 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2702 continue to scan other files.
2703
2704 (o) The -s option was incorrectly implemented. For compatibility with other
2705 greps, it now suppresses the error message for a non-existent or non-
2706 accessible file (but not the return code). There is a new option called
2707 -q that suppresses the output of matching lines, which was what -s was
2708 previously doing.
2709
2710 (p) Added --include and --exclude options to specify files for inclusion
2711 and exclusion when recursing.
2712
2713 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2714 Hopefully, it now does.
2715
2716 12. Missing cast in pcre_study().
2717
2718 13. Added an "uninstall" target to the makefile.
2719
2720 14. Replaced "extern" in the function prototypes in Makefile.in with
2721 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2722 world, but is set differently for Windows.
2723
2724 15. Added a second compiling function called pcre_compile2(). The only
2725 difference is that it has an extra argument, which is a pointer to an
2726 integer error code. When there is a compile-time failure, this is set
2727 non-zero, in addition to the error test pointer being set to point to an
2728 error message. The new argument may be NULL if no error number is required
2729 (but then you may as well call pcre_compile(), which is now just a
2730 wrapper). This facility is provided because some applications need a
2731 numeric error indication, but it has also enabled me to tidy up the way
2732 compile-time errors are handled in the POSIX wrapper.
2733
2734 16. Added VPATH=.libs to the makefile; this should help when building with one
2735 prefix path and installing with another. (Or so I'm told by someone who
2736 knows more about this stuff than I do.)
2737
2738 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2739 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2740 match everything, including newlines. This is not POSIX-compatible, but
2741 somebody wanted the feature. From pcretest it can be activated by using
2742 both the P and the s flags.
2743
2744 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2745
2746 19. libpcre.pc was being incorrectly installed as executable.
2747
2748 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2749 it now also looks for '\r' so that it will work unmodified on Windows.
2750
2751 21. Added Google's contributed C++ wrapper to the distribution.
2752
2753 22. Added some untidy missing memory free() calls in pcretest, to keep
2754 Electric Fence happy when testing.
2755
2756
2757
2758 Version 5.0 13-Sep-04
2759 ---------------------
2760
2761 1. Internal change: literal characters are no longer packed up into items
2762 containing multiple characters in a single byte-string. Each character
2763 is now matched using a separate opcode. However, there may be more than one
2764 byte in the character in UTF-8 mode.
2765
2766 2. The pcre_callout_block structure has two new fields: pattern_position and
2767 next_item_length. These contain the offset in the pattern to the next match
2768 item, and its length, respectively.
2769
2770 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2771 insertion of callouts before each pattern item. Added the /C option to
2772 pcretest to make use of this.
2773
2774 4. On the advice of a Windows user, the lines
2775
2776 #if defined(_WIN32) || defined(WIN32)
2777 _setmode( _fileno( stdout ), 0x8000 );
2778 #endif /* defined(_WIN32) || defined(WIN32) */
2779
2780 have been added to the source of pcretest. This apparently does useful
2781 magic in relation to line terminators.
2782
2783 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2784 for the benefit of those environments where the "b" makes a difference.
2785
2786 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2787 to know about it. I have put a hack into configure.in that adds in code
2788 to set GCC=yes if CC=icc. This seems to end up at a point in the
2789 generated configure script that is early enough to affect the setting of
2790 compiler options, which is what is needed, but I have no means of testing
2791 whether it really works. (The user who reported this had patched the
2792 generated configure script, which of course I cannot do.)
2793
2794 LATER: After change 22 below (new libtool files), the configure script
2795 seems to know about icc (and also ecc). Therefore, I have commented out
2796 this hack in configure.in.
2797
2798 7. Added support for pkg-config (2 patches were sent in).
2799
2800 8. Negated POSIX character classes that used a combination of internal tables
2801 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2802 [[:^ascii]]. Typically, they would match almost any characters. The other
2803 POSIX classes were not broken in this way.
2804
2805 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2806 to find the match, as PCRE was deluded into thinking that the match had to
2807 start at the start point or following a newline. The same bug applied to
2808 patterns with negative forward assertions or any backward assertions
2809 preceding ".*" at the start, unless the pattern required a fixed first
2810 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2811
2812 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2813 starting at the last subject character, bytes beyond the end of the subject
2814 string were read.
2815
2816 11. Renamed the variable "class" as "classbits" to make life easier for C++
2817 users. (Previously there was a macro definition, but it apparently wasn't
2818 enough.)
2819
2820 12. Added the new field "tables" to the extra data so that tables can be passed
2821 in at exec time, or the internal tables can be re-selected. This allows
2822 a compiled regex to be saved and re-used at a later time by a different
2823 program that might have everything at different addresses.
2824
2825 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2826 -R library as well as a -L library.
2827
2828 14. The debugging options of pcretest (-d on the command line or D on a
2829 pattern) showed incorrect output for anything following an extended class
2830 that contained multibyte characters and which was followed by a quantifier.
2831
2832 15. Added optional support for general category Unicode character properties
2833 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2834 support. It adds about 90K to the size of the library. The meanings of the
2835 inbuilt class escapes such as \d and \s have NOT been changed.
2836
2837 16. Updated pcredemo.c to include calls to free() to release the memory for the
2838 compiled pattern.
2839
2840 17. The generated file chartables.c was being created in the source directory
2841 instead of in the building directory. This caused the build to fail if the
2842 source directory was different from the building directory, and was
2843 read-only.
2844
2845 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2846 file. No doubt somebody will tell me if they don't make sense... Also added
2847 Dan Mooney's comments about building on OpenVMS.
2848
2849 19. Added support for partial matching via the PCRE_PARTIAL option for
2850 pcre_exec() and the \P data escape in pcretest.
2851
2852 20. Extended pcretest with 3 new pattern features:
2853
2854 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2855 write the compiled pattern to the file whose name is "rest-of-line".
2856 This is a straight binary dump of the data, with the saved pointer to
2857 the character tables forced to be NULL. The study data, if any, is
2858 written too. After writing, pcretest reads a new pattern.
2859
2860 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2861 compiled pattern from the given file. There must not be any
2862 occurrences of "<" in the file name (pretty unlikely); if there are,
2863 pcretest will instead treat the initial "<" as a pattern delimiter.
2864 After reading in the pattern, pcretest goes on to read data lines as
2865 usual.
2866
2867 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2868 and 16-bit fields in a compiled pattern, to simulate a pattern that
2869 was compiled on a host of opposite endianness.
2870
2871 21. The pcre-exec() function can now cope with patterns that were compiled on
2872 hosts of opposite endianness, with this restriction:
2873
2874 As for any compiled expression that is saved and used later, the tables
2875 pointer field cannot be preserved; the extra_data field in the arguments
2876 to pcre_exec() should be used to pass in a tables address if a value
2877 other than the default internal tables were used at compile time.
2878
2879 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2880 now diagnosed as an error. Previously, most of the time, a negative number
2881 would have been treated as zero, but if in addition "ovector" was passed as
2882 NULL, a crash could occur.
2883
2884 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2885 new versions from the libtool 1.5 distribution (the last one is a copy of
2886 a file called libtool.m4). This seems to have fixed the need to patch
2887 "configure" to support Darwin 1.3 (which I used to do). However, I still
2888 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2889 workstation).
2890
2891 24. Changed the PCRE licence to be the more standard "BSD" licence.
2892
2893
2894 Version 4.5 01-Dec-03
2895 ---------------------
2896
2897 1. There has been some re-arrangement of the code for the match() function so
2898 that it can be compiled in a version that does not call itself recursively.
2899 Instead, it keeps those local variables that need separate instances for
2900 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2901 needs to "recurse". Keeping track of where control must go is done by means
2902 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2903 hide most of the details from the main code, and operates only if
2904 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2905 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2906 operating.
2907
2908 To make it easier for callers to provide specially tailored get/free
2909 functions for this usage, two new functions, pcre_stack_malloc, and
2910 pcre_stack_free, are used. They are always called in strict stacking order,
2911 and the size of block requested is always the same.
2912
2913 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2914 PCRE has been compiled to use the stack or the heap for recursion. The
2915 -C option of pcretest uses this to show which version is compiled.
2916
2917 A new data escape \S, is added to pcretest; it causes the amounts of store
2918 obtained and freed by both kinds of malloc/free at match time to be added
2919 to the output.
2920
2921 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2922 what's available on my current Linux desktop machine.
2923
2924 3. When matching a UTF-8 string, the test for a valid string at the start has
2925 been extended. If start_offset is not zero, PCRE now checks that it points
2926 to a byte that is the start of a UTF-8 character. If not, it returns
2927 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2928 this is necessary because there may be backward assertions in the pattern.
2929 When matching the same subject several times, it may save resources to use
2930 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2931
2932 4. The code for checking the validity of UTF-8 strings has been tightened so
2933 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2934 containing "overlong sequences".
2935
2936 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2937 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2938 should have been "&", but it just so happened that all the cases this let
2939 through by mistake were picked up later in the function.
2940
2941 6. I had used a variable called "isblank" - this is a C99 function, causing
2942 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2943
2944 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2945 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2946 so the version is shown; (c) stop "make test" echoing "./RunTest".
2947
2948 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2949
2950 9. The prototype for memmove() for systems that don't have it was using
2951 size_t, but the inclusion of the header that defines size_t was later. I've
2952 moved the #includes for the C headers earlier to avoid this.
2953
2954 10. Added some adjustments to the code to make it easier to compiler on certain
2955 special systems:
2956
2957 (a) Some "const" qualifiers were missing.
2958 (b) Added the macro EXPORT before all exported functions; by default this
2959 is defined to be empty.
2960 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2961 that it reads its output file name as an argument instead of writing
2962 to the standard output and assuming this can be redirected.
2963
2964 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2965 class containing characters with values greater than 255, PCRE compilation
2966 went into a loop.
2967
2968 12. A recursive reference to a subpattern that was within another subpattern
2969 that had a minimum quantifier of zero caused PCRE to crash. For example,
2970 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2971 recursion. If the recursively-called subpattern itself had a zero repeat,
2972 that was OK.
2973
2974 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2975 buffer into which it was copied (for escape processing) was still set at
2976 1024, so long lines caused crashes.
2977
2978 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2979 "internal error: code overflow...". This applied to any character class
2980 that was followed by a possessive quantifier.
2981
2982 15. Modified the Makefile to add libpcre.la as a prerequisite for
2983 libpcreposix.la because I was told this is needed for a parallel build to
2984 work.
2985
2986 16. If a pattern that contained .* following optional items at the start was
2987 studied, the wrong optimizing data was generated, leading to matching
2988 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2989 matching string must start with a or b or c. The correct conclusion for
2990 this pattern is that a match can start with any character.
2991
2992
2993 Version 4.4 13-Aug-03
2994 ---------------------
2995
2996 1. In UTF-8 mode, a character class containing characters with values between
2997 127 and 255 was not handled correctly if the compiled pattern was studied.
2998 In fixing this, I have also improved the studying algorithm for such
2999 classes (slightly).
3000
3001 2. Three internal functions had redundant arguments passed to them. Removal
3002 might give a very teeny performance improvement.
3003
3004 3. Documentation bug: the value of the capture_top field in a callout is *one
3005 more than* the number of the hightest numbered captured substring.
3006
3007 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3008 in incorrectly linking with a previously installed version. They now link
3009 explicitly with libpcre.la.
3010
3011 5. configure.in no longer needs to recognize Cygwin specially.
3012
3013 6. A problem in pcre.in for Windows platforms is fixed.
3014
3015 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3016 pcretest, it used to include the size of the study block as part of its
3017 output. Unfortunately, the structure contains a field that has a different
3018 size on different hardware architectures. This meant that the tests that
3019 showed this size failed. As the block is currently always of a fixed size,
3020 this information isn't actually particularly useful in pcretest output, so
3021 I have just removed it.
3022
3023 8. Three pre-processor statements accidentally did not start in column 1.
3024 Sadly, there are *still* compilers around that complain, even though
3025 standard C has not required this for well over a decade. Sigh.
3026
3027 9. In pcretest, the code for checking callouts passed small integers in the
3028 callout_data field, which is a void * field. However, some picky compilers
3029 complained about the casts involved for this on 64-bit systems. Now
3030 pcretest passes the address of the small integer instead, which should get
3031 rid of the warnings.
3032
3033 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3034 both compile and run time, and gives an error if an invalid UTF-8 sequence
3035 is found. There is a option for disabling this check in cases where the
3036 string is known to be correct and/or the maximum performance is wanted.
3037
3038 11. In response to a bug report, I changed one line in Makefile.in from
3039
3040 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3041 to
3042 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3043
3044 to look similar to other lines, but I have no way of telling whether this
3045 is the right thing to do, as I do not use Windows. No doubt I'll get told
3046 if it's wrong...
3047
3048
3049 Version 4.3 21-May-03
3050 ---------------------
3051
3052 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3053 Makefile.
3054
3055 2. Some refactoring to improve the quality of the code:
3056
3057 (i) The utf8_table... variables are now declared "const".
3058
3059 (ii) The code for \cx, which used the "case flipping" table to upper case
3060 lower case letters, now just substracts 32. This is ASCII-specific,
3061 but the whole concept of \cx is ASCII-specific, so it seems
3062 reasonable.
3063
3064 (iii) PCRE was using its character types table to recognize decimal and
3065 hexadecimal digits in the pattern. This is silly, because it handles
3066 only 0-9, a-f, and A-F, but the character types table is locale-
3067 specific, which means strange things might happen. A private
3068 table is now used for this - though it costs 256 bytes, a table is
3069 much faster than multiple explicit tests. Of course, the standard
3070 character types table is still used for matching digits in subject
3071 strings against \d.
3072
3073 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3074 ending in _t are). So I've renamed it as ESC_tee.
3075
3076 3. The first argument for regexec() in the POSIX wrapper should have been
3077 defined as "const".
3078
3079 4. Changed pcretest to use malloc() for its buffers so that they can be
3080 Electric Fenced for debugging.
3081
3082 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3083 to read one or more bytes before the start of the subject string. Often this
3084 had no effect on PCRE's behaviour, but in some circumstances it could
3085 provoke a segmentation fault.
3086
3087 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3088 to try to read one or more bytes before the start of the subject string.
3089
3090 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3091 UTF-8 support could misbehave in various ways if the subject string
3092 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3093 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3094 back over UTF-8 characters.)
3095
3096
3097 Version 4.2 14-Apr-03
3098 ---------------------
3099
3100 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3101
3102 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3103 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3104 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3105 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3106 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3107 and BUILD_EXEEXT
3108 Note: automatic setting of the BUILD variables is not yet working
3109 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3110 compile-time but not at link-time
3111 [LINK]: use for linking executables only
3112 make different versions for Windows and non-Windows
3113 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3114 libraries
3115 [LINK_FOR_BUILD]: new variable
3116 [OBJEXT]: use throughout
3117 [EXEEXT]: use throughout
3118 <winshared>: new target
3119 <wininstall>: new target
3120 <dftables.o>: use native compiler
3121 <dftables>: use native linker
3122 <install>: handle Windows platform correctly
3123 <clean>: ditto
3124 <check>: ditto
3125 copy DLL to top builddir before testing
3126
3127 As part of these changes, -no-undefined was removed again. This was reported
3128 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3129 in any case.
3130
3131 3. Some tidies to get rid of compiler warnings:
3132
3133 . In the match_data structure, match_limit was an unsigned long int, whereas
3134 match_call_count was an int. I've made them both unsigned long ints.
3135
3136 . In pcretest the fact that a const uschar * doesn't automatically cast to
3137 a void * provoked a warning.
3138
3139 . Turning on some more compiler warnings threw up some "shadow" variables
3140 and a few more missing casts.
3141
3142 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3143 option, a class that contained a single character with a value between 128
3144 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3145
3146 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3147 option, a class that contained several characters, but with at least one
3148 whose value was between 128 and 255 caused PCRE to crash.
3149
3150
3151 Version 4.1 12-Mar-03
3152 ---------------------
3153
3154 1. Compiling with gcc -pedantic found a couple of places where casts were
3155 needed, and a string in dftables.c that was longer than standard compilers are
3156 required to support.
3157
3158 2. Compiling with Sun's compiler found a few more places where the code could
3159 be tidied up in order to avoid warnings.
3160
3161 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3162 first of these names is deprecated in the latest Autoconf in favour of the name
3163 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3164 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3165 analogy I have changed it to CFLAGS_FOR_BUILD.
3166
3167 4. Added -no-undefined to the linking command in the Makefile, because this is
3168 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3169 linking step for the pcreposix library.
3170
3171 5. PCRE was failing to diagnose the case of two named groups with the same
3172 name.
3173
3174 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3175 literal character that is needed in the subject for a match, and scans along to
3176 ensure that it is present before embarking on the full matching process. This
3177 saves time in cases of nested unlimited repeats that are never going to match.
3178 Problem: the scan can take a lot of time if the subject is very long (e.g.
3179 megabytes), thus penalizing straightforward matches. It is now done only if the
3180 amount of subject to be scanned is less than 1000 bytes.
3181
3182 7. A lesser problem with the same optimization is that it was recording the
3183 first character of an anchored pattern as "needed", thus provoking a search
3184 right along the subject, even when the first match of the pattern was going to
3185 fail. The "needed" character is now not set for anchored patterns, unless it
3186 follows something in the pattern that is of non-fixed length. Thus, it still
3187 fulfils its original purpose of finding quick non-matches in cases of nested
3188 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3189
3190
3191 Version 4.0 17-Feb-03
3192 ---------------------
3193
3194 1. If a comment in an extended regex that started immediately after a meta-item
3195 extended to the end of string, PCRE compiled incorrect data. This could lead to
3196 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3197
3198 2. Moved to autoconf 2.53 and libtool 1.4.2.
3199
3200 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3201 the special perltest8 script is no longer needed - all the tests can be run
3202 from a single perltest script.
3203
3204 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3205 by \s. It has now been removed in PCRE. This means it isn't recognized as
3206 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3207 class [:space:] *does* include VT, thereby creating a mess.
3208
3209 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3210 space and tab.
3211
3212 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3213 its new features into the main test script, reducing the number of scripts.
3214
3215 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3216 were backward compatible, and made the (?i) apply to the whole pattern, as if
3217 /i were given. Now it behaves more logically, and applies the option setting
3218 only to what follows. PCRE has been changed to follow suit. However, if it
3219 finds options settings right at the start of the pattern, it extracts them into
3220 the global options, as before. Thus, they show up in the info data.
3221
3222 8. Added support for the \Q...\E escape sequence. Characters in between are
3223 treated as literals. This is slightly different from Perl in that $ and @ are
3224 also handled as literals inside the quotes. In Perl, they will cause variable
3225 interpolation. Note the following examples:
3226
3227 Pattern PCRE matches Perl matches
3228
3229 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3230 \Qabc\$xyz\E abc\$xyz abc\$xyz
3231 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3232
3233 For compatibility with Perl, \Q...\E sequences are recognized inside character
3234 classes as well as outside them.
3235
3236 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3237 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3238 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3239 signed/unsigned warnings.
3240
3241 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3242 option for pcretest, so I've replaced it by a simple function that does just
3243 that job.
3244
3245 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3246 "pcregrep -".
3247
3248 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3249 Java package. This provides some syntactic sugar for simple cases of what my
3250 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3251 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3252 item, you can use this simplified notation. Note that only makes sense with
3253 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3254 greediness, whatever the setting of the PCRE_UNGREEDY option.
3255
3256 13. A change of greediness default within a pattern was not taking effect at
3257 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3258 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3259 was abstracted outside.
3260
3261 14. PCRE now supports the \G assertion. It is true when the current matching
3262 position is at the start point of the match. This differs from \A when the
3263 starting offset is non-zero. Used with the /g option of pcretest (or similar
3264 code), it works in the same way as it does for Perl's /g option. If all
3265 alternatives of a regex begin with \G, the expression is anchored to the start
3266 match position, and the "anchored" flag is set in the compiled expression.
3267
3268 15. Some bugs concerning the handling of certain option changes within patterns
3269 have been fixed. These applied to options other than (?ims). For example,
3270 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3271 been the other way round. Some of this was related to change 7 above.
3272
3273 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3274 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3275 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3276 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3277
3278 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3279 mode. Unlike ".", it always matches newline, whatever the setting of
3280 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3281 assertions. Perl allows it, but it doesn't (in general) work because it can't
3282 calculate the length of the lookbehind. At least, that's the case for Perl
3283 5.8.0 - I've been told they are going to document that it doesn't work in
3284 future.
3285
3286 18. Added an error diagnosis for escapes that PCRE does not support: these are
3287 \L, \l, \N, \P, \p, \U, \u, and \X.
3288
3289 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3290 reading past the end of the pattern in cases such as /[abcd/.
3291
3292 20. PCRE was getting more memory than necessary for patterns with classes that
3293 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3294
3295 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3296 compiling PCRE for use with Virtual Pascal.
3297
3298 22. Small fix to the Makefile to make it work properly if the build is done
3299 outside the source tree.
3300
3301 23. Added a new extension: a condition to go with recursion. If a conditional
3302 subpattern starts with (?(R) the "true" branch is used if recursion has
3303 happened, whereas the "false" branch is used only at the top level.
3304
3305 24. When there was a very long string of literal characters (over 255 bytes
3306 without UTF support, over 250 bytes with UTF support), the computation of how
3307 much memory was required could be incorrect, leading to segfaults or other
3308 strange effects.
3309
3310 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3311 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3312 there was a subsequent back reference to those brackets. This meant that, for
3313 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3314 possible to check for precisely this case. All we can do is abandon the
3315 optimization if .* occurs inside capturing brackets when there are any back
3316 references whatsoever. (See below for a better fix that came later.)
3317
3318 26. The handling of the optimization for finding the first character of a
3319 non-anchored pattern, and for finding a character that is required later in the
3320 match were failing in some cases. This didn't break the matching; it just
3321 failed to optimize when it could. The way this is done has been re-implemented.
3322
3323 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3324
3325 28. Added a new feature that provides some of the functionality that Perl
3326 provides with (?{...}). The facility is termed a "callout". The way it is done
3327 in PCRE is for the caller to provide an optional function, by setting
3328 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3329 global variable. By default it is unset, which disables all calling out. To get
3330 the function called, the regex must include (?C) at appropriate points. This
3331 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3332 This provides a means of identifying different callout points. When PCRE
3333 reaches such a point in the regex, if pcre_callout has been set, the external
3334 function is called. It is provided with data in a structure called
3335 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3336 matching continues; if it returns a non-zero value, the match at the current
3337 point fails. However, backtracking will occur if possible. [This was changed
3338 later and other features added - see item 49 below.]
3339
3340 29. pcretest is upgraded to test the callout functionality. It provides a
3341 callout function that displays information. By default, it shows the start of
3342 the match and the current position in the text. There are some new data escapes
3343 to vary what happens:
3344
3345 \C+ in addition, show current contents of captured substrings
3346 \C- do not supply a callout function
3347 \C!n return 1 when callout number n is reached
3348 \C!n!m return 1 when callout number n is reached for the mth time
3349
3350 30. If pcregrep was called with the -l option and just a single file name, it
3351 output "<stdin>" if a match was found, instead of the file name.
3352
3353 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3354 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3355 pcre_exec(). This saves a malloc/free per call. The default value of
3356 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3357 when configuring.
3358
3359 32. The default maximum size of a compiled pattern is 64K. There have been a
3360 few cases of people hitting this limit. The code now uses macros to handle the
3361 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3362 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3363 configuring. Tests 2 and 5 work only with 2-byte links because they output
3364 debugging information about compiled patterns.
3365
3366 33. Internal code re-arrangements:
3367
3368 (a) Moved the debugging function for printing out a compiled regex into
3369 its own source file (printint.c) and used #include to pull it into
3370 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3371 separate copies.
3372
3373 (b) Defined the list of op-code names for debugging as a macro in
3374 internal.h so that it is next to the definition of the opcodes.
3375
3376 (c) Defined a table of op-code lengths for simpler skipping along compiled
3377 code. This is again a macro in internal.h so that it is next to the
3378 definition of the opcodes.
3379
3380 34. Added support for recursive calls to individual subpatterns, along the
3381 lines of Robin Houston's patch (but implemented somewhat differently).
3382
3383 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3384 allow it to read and process whole directories in Win32. This code was
3385 contributed by Lionel Fourquaux; it has not been tested by me.
3386
3387 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3388 used to name a group. Names consist of alphanumerics and underscores, and must
3389 be unique. Back references use the syntax (?P=name) and recursive calls use
3390 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3391 numbers. The function pcre_fullinfo() can be used after compilation to extract
3392 a name/number map. There are three relevant calls:
3393
3394 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3395 PCRE_INFO_NAMECOUNT yields the number of entries
3396 PCRE_INFO_NAMETABLE yields a pointer to the map.
3397
3398 The map is a vector of fixed-size entries. The size of each entry depends on
3399 the length of the longest name used. The first two bytes of each entry are the
3400 group number, most significant byte first. There follows the corresponding
3401 name, zero terminated. The names are in alphabetical order.
3402
3403 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3404 case instead of 255. Making it the same both with and without UTF-8 support
3405 means that the same test output works with both.
3406
3407 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3408 calling malloc() with a zero argument.
3409
3410 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3411 optimization. I've improved things by keeping a bitmap of backreferences with
3412 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3413 fact referenced, the optimization can be applied. It is unlikely that a
3414 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3415 the match to follow \n) will appear inside brackets with a number greater than
3416 31, but if it does, any back reference > 31 suppresses the optimization.
3417
3418 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3419 of disabling numbered capturing parentheses. Any opening parenthesis that is
3420 not followed by ? behaves as if it were followed by ?: but named parentheses
3421 can still be used for capturing (and they will acquire numbers in the usual
3422 way).
3423
3424 41. Redesigned the return codes from the match() function into yes/no/error so
3425 that errors can be passed back from deep inside the nested calls. A malloc
3426 failure while inside a recursive subpattern call now causes the
3427 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3428
3429 42. It is now possible to set a limit on the number of times the match()
3430 function is called in a call to pcre_exec(). This facility makes it possible to
3431 limit the amount of recursion and backtracking, though not in a directly
3432 obvious way, because the match() function is used in a number of different
3433 circumstances. The count starts from zero for each position in the subject
3434 string (for non-anchored patterns). The default limit is, for compatibility, a
3435 large number, namely 10 000 000. You can change this in two ways:
3436
3437 (a) When configuring PCRE before making, you can use --with-match-limit=n
3438 to set a default value for the compiled library.
3439
3440 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3441 a different value is set. See 45 below.
3442
3443 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3444
3445 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3446 of things that can be changed at compile time. The first argument specifies
3447 what is wanted and the second points to where the information is to be placed.
3448 The current list of available information is:
3449
3450 PCRE_CONFIG_UTF8
3451
3452 The output is an integer that is set to one if UTF-8 support is available;
3453 otherwise it is set to zero.
3454
3455 PCRE_CONFIG_NEWLINE
3456
3457 The output is an integer that it set to the value of the code that is used for
3458 newline. It is either LF (10) or CR (13).
3459
3460 PCRE_CONFIG_LINK_SIZE
3461
3462 The output is an integer that contains the number of bytes used for internal
3463 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3464
3465 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3466
3467 The output is an integer that contains the threshold above which the POSIX
3468 interface uses malloc() for output vectors. See item 31 above.
3469
3470 PCRE_CONFIG_MATCH_LIMIT
3471
3472 The output is an unsigned integer that contains the default limit of the number
3473 of match() calls in a pcre_exec() execution. See 42 above.
3474
3475 44. pcretest has been upgraded by the addition of the -C option. This causes it
3476 to extract all the available output from the new pcre_config() function, and to
3477 output it. The program then exits immediately.
3478
3479 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3480 order to support additional features. One way would have been to define
3481 pcre_exec2() (for example) with extra arguments, but this would not have been
3482 extensible, and would also have required all calls to the original function to
3483 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3484 is used for passing in "extra" data from pcre_study().
3485
3486 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3487 contains the following fields:
3488
3489 flags a bitmap indicating which of the following fields are set
3490 study_data opaque data from pcre_study()
3491 match_limit a way of specifying a limit on match() calls for a specific
3492 call to pcre_exec()
3493 callout_data data for callouts (see 49 below)
3494
3495 The flag bits are also defined in pcre.h, and are
3496
3497 PCRE_EXTRA_STUDY_DATA
3498 PCRE_EXTRA_MATCH_LIMIT
3499 PCRE_EXTRA_CALLOUT_DATA
3500
3501 The pcre_study() function now returns one of these new pcre_extra blocks, with
3502 the actual study data pointed to by the study_data field, and the
3503 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3504 before. That is, this change is entirely upwards-compatible and requires no
3505 change to existing code.
3506
3507 If you want to pass in additional data to pcre_exec(), you can either place it
3508 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3509 block.
3510
3511 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3512 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3513 times with different match limits, until it finds the minimum value needed for
3514 pcre_exec() to complete. The value is then output. This can be instructive; for
3515 most simple matches the number is quite small, but for pathological cases it
3516 gets very large very quickly.
3517
3518 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3519 returns the size of the data block pointed to by the study_data field in a
3520 pcre_extra block, that is, the value that was passed as the argument to
3521 pcre_malloc() when PCRE was getting memory in which to place the information
3522 created by pcre_study(). The fourth argument should point to a size_t variable.
3523 pcretest has been extended so that this information is shown after a successful
3524 pcre_study() call when information about the compiled regex is being displayed.
3525
3526 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3527 because what follows is always an absolute path. (Later: it turns out that this
3528 is more than cosmetic for MinGW, because it doesn't like empty path
3529 components.)
3530
3531 49. Some changes have been made to the callout feature (see 28 above):
3532
3533 (i) A callout function now has three choices for what it returns:
3534
3535 0 => success, carry on matching
3536 > 0 => failure at this point, but backtrack if possible
3537 < 0 => serious error, return this value from pcre_exec()
3538
3539 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3540 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3541 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3542 use by callout functions. It will never be used by PCRE itself.
3543
3544 (ii) The pcre_extra structure (see 45 above) has a void * field called
3545 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3546 pcre_callout_block structure has a field of the same name. The contents of
3547 the field passed in the pcre_extra structure are passed to the callout
3548 function in the corresponding field in the callout block. This makes it
3549 easier to use the same callout-containing regex from multiple threads. For
3550 testing, the pcretest program has a new data escape
3551
3552 \C*n pass the number n (may be negative) as callout_data
3553
3554 If the callout function in pcretest receives a non-zero value as
3555 callout_data, it returns that value.
3556
3557 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3558 there were some redundant $(CFLAGS) in commands that are now specified as
3559 $(LINK), which already includes $(CFLAGS).
3560
3561 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3562 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3563 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3564 one-byte characters throughout. Note that case-insensitive matching applies
3565 only to characters whose values are less than 256. PCRE doesn't support the
3566 notion of cases for higher-valued characters.
3567
3568 (i) A character class whose characters are all within 0-255 is handled as
3569 a bit map, and the map is inverted for negative classes. Previously, a
3570 character > 255 always failed to match such a class; however it should
3571 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3572
3573 (ii) A negated character class with a single character < 255 is coded as
3574 "not this character" (OP_NOT). This wasn't working properly when the test
3575 character was multibyte, either singly or repeated.
3576
3577 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3578 mode, for example: \x{100}{2,3}.
3579
3580 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3581 singly or repeated) now correctly test multibyte characters. However,
3582 PCRE doesn't recognize any characters with values greater than 255 as
3583 digits, spaces, or word characters. Such characters always match \D, \S,
3584 and \W, and never match \d, \s, or \w.
3585
3586 (v) Classes may now contain characters and character ranges with values
3587 greater than 255. For example: [ab\x{100}-\x{400}].
3588
3589 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3590 PCRE in UTF-8 mode.
3591
3592 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3593 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3594 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3595 value.)
3596
3597 53. The single man page has become too large. I have therefore split it up into
3598 a number of separate man pages. These also give rise to individual HTML pages;
3599 these are now put in a separate directory, and there is an index.html page that
3600 lists them all. Some hyperlinking between the pages has been installed.
3601
3602 54. Added convenience functions for handling named capturing parentheses.
3603
3604 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3605 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3606 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3607 are faulted.
3608
3609 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3610 calling configure. These values are used when compiling the dftables.c program
3611 which is run to generate the source of the default character tables. They
3612 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3613 you will need to set these values.
3614
3615 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3616
3617
3618 Version 3.9 02-Jan-02
3619 ---------------------
3620
3621 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3622
3623 2. If --disable-static was given, the building process failed when trying to
3624 build pcretest and pcregrep. (For some reason it was using libtool to compile
3625 them, which is not right, as they aren't part of the library.)
3626
3627
3628 Version 3.8 18-Dec-01
3629 ---------------------
3630
3631 1. The experimental UTF-8 code was completely screwed up. It was packing the
3632 bytes in the wrong order. How dumb can you get?
3633
3634
3635 Version 3.7 29-Oct-01
3636 ---------------------
3637
3638 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3639 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3640 this didn't happen under Solaris 8, where I normally test things.
3641
3642 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3643 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3644 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3645 this fix an hour or so after the initial 3.7 release.)
3646
3647
3648 Version 3.6 23-Oct-01
3649 ---------------------
3650
3651 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3652 offsets passed as NULL with zero offset count.
3653
3654 2. The config.guess and config.sub files had not been updated when I moved to
3655 the latest autoconf.
3656
3657
3658 Version 3.5 15-Aug-01
3659 ---------------------
3660
3661 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3662 had been forgotten.
3663
3664 2. By using declared but undefined structures, we can avoid using "void"
3665 definitions in pcre.h while keeping the internal definitions of the structures
3666 private.
3667
3668 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3669 user point of view, this means that both static and shared libraries are built
3670 by default, but this can be individually controlled. More of the work of
3671 handling this static/shared cases is now inside libtool instead of PCRE's make
3672 file.
3673
3674 4. The pcretest utility is now installed along with pcregrep because it is
3675 useful for users (to test regexs) and by doing this, it automatically gets
3676 relinked by libtool. The documentation has been turned into a man page, so
3677 there are now .1, .txt, and .html versions in /doc.
3678
3679 5. Upgrades to pcregrep:
3680 (i) Added long-form option names like gnu grep.
3681 (ii) Added --help to list all options with an explanatory phrase.
3682 (iii) Added -r, --recursive to recurse into sub-directories.
3683 (iv) Added -f, --file to read patterns from a file.
3684
3685 6. pcre_exec() was referring to its "code" argument before testing that
3686 argument for NULL (and giving an error if it was NULL).
3687
3688 7. Upgraded Makefile.in to allow for compiling in a different directory from
3689 the source directory.
3690
3691 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3692 options bits, the pointer it was passed was to an int instead of to an unsigned
3693 long int. This mattered only on 64-bit systems.
3694
3695 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3696 generated) instead of pcre.in, which it its source. Also made the same change
3697 in several of the .c files.
3698
3699 10. A new release of gcc defines printf() as a macro, which broke pcretest
3700 because it had an ifdef in the middle of a string argument for printf(). Fixed
3701 by using separate calls to printf().
3702
3703 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3704 script, to force use of CR or LF instead of \n in the source. On non-Unix
3705 systems, the value can be set in config.h.
3706
3707 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3708 absolute limit. Changed the text of the error message to make this clear, and
3709 likewise updated the man page.
3710
3711 13. The limit of 99 on the number of capturing subpatterns has been removed.
3712 The new limit is 65535, which I hope will not be a "real" limit.
3713
3714
3715 Version 3.4 22-Aug-00
3716 ---------------------
3717
3718 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3719
3720 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3721
3722
3723 Version 3.3 01-Aug-00
3724 ---------------------
3725
3726 1. If an octal character was given, but the value was greater than \377, it
3727 was not getting masked to the least significant bits, as documented. This could
3728 lead to crashes in some systems.
3729
3730 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3731 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3732
3733 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3734 These just pass their arguments on to (pcre_free)(), but they are provided
3735 because some uses of PCRE bind it to non-C systems that can call its functions,
3736 but cannot call free() or pcre_free() directly.
3737
3738 4. Add "make test" as a synonym for "make check". Corrected some comments in
3739 the Makefile.
3740
3741 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3742 Makefile.
3743
3744 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3745 command called pgrep for grepping around the active processes.
3746
3747 7. Added the beginnings of support for UTF-8 character strings.
3748
3749 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3750 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3751 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3752 out for the ar command.)
3753
3754
3755 Version 3.2 12-May-00
3756 ---------------------
3757
3758 This is purely a bug fixing release.
3759
3760 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3761 of ZA. This was just one example of several cases that could provoke this bug,
3762 which was introduced by change 9 of version 2.00. The code for breaking
3763 infinite loops after an iteration that matches an empty string was't working
3764 correctly.
3765
3766 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3767 when matched against abbab (for example). After matching an empty string, it
3768 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3769 caused it to match further down the string than it should.
3770
3771 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3772 was there because it doesn't seem to be needed, and it causes trouble on some
3773 systems, as it is not a Standard C header. It has been removed.
3774
3775 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3776 were reported on the Macintosh. The changes were from
3777
3778 while ((c = *(++ptr)) != 0 && c != '\n');
3779 to
3780 while ((c = *(++ptr)) != 0 && c != '\n') ;
3781
3782 Totally extraordinary, but if that's what it takes...
3783
3784 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3785 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3786 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3787 assumes the way PCRE uses memmove() (always moving upwards).
3788
3789 6. PCRE is being used in one environment where strchr() is not available. There
3790 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3791 faster code anyway.
3792
3793
3794 Version 3.1 09-Feb-00
3795 ---------------------
3796
3797 The only change in this release is the fixing of some bugs in Makefile.in for
3798 the "install" target:
3799
3800 (1) It was failing to install pcreposix.h.
3801
3802 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3803
3804
3805 Version 3.0 01-Feb-00
3806 ---------------------
3807
3808 1. Add support for the /+ modifier to perltest (to output $` like it does in
3809 pcretest).
3810
3811 2. Add support for the /g modifier to perltest.
3812
3813 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3814 matches null strings.
3815
3816 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3817 pattern. Perl treats empty patterns specially - it reuses the most recent
3818 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3819 effect.
3820
3821 5. The POSIX interface was broken in that it was just handing over the POSIX
3822 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3823 required a bigger vector, with some working space on the end. This means that
3824 the POSIX wrapper now has to get and free some memory, and copy the results.
3825
3826 6. Added some simple autoconf support, placing the test data and the
3827 documentation in separate directories, re-organizing some of the
3828 information files, and making it build pcre-config (a GNU standard). Also added
3829 libtool support for building PCRE as a shared library, which is now the
3830 default.
3831
3832 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3833 09 are not valid octal constants. Single digits will be used for minor values
3834 less than 10.
3835
3836 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3837 existing programs that set these in the POSIX interface can use PCRE without
3838 modification.
3839
3840 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3841 return all that pcre_info() returns, plus additional data. The pcre_info()
3842 function is retained for compatibility, but is considered to be obsolete.
3843
3844 10. Added experimental recursion feature (?R) to handle one common case that
3845 Perl 5.6 will be able to do with (?p{...}).
3846
3847 11. Added support for POSIX character classes like [:alpha:], which Perl is
3848 adopting.
3849
3850
3851 Version 2.08 31-Aug-99
3852 ----------------------
3853
3854 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3855 trying to match at the startoffset position, but instead was moving forward to
3856 the next newline as if a previous match had failed.
3857
3858 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3859 and could get into a loop if a null string was matched other than at the start
3860 of the subject.
3861
3862 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3863 be distinguished at compile time, and for completeness also added PCRE_DATE.
3864
3865 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3866 in GnuWin32 environments.
3867
3868
3869 Version 2.07 29-Jul-99
3870 ----------------------
3871
3872 1. The documentation is now supplied in plain text form and HTML as well as in
3873 the form of man page sources.
3874
3875 2. C++ compilers don't like assigning (void *) values to other pointer types.
3876 In particular this affects malloc(). Although there is no problem in Standard
3877 C, I've put in casts to keep C++ compilers happy.
3878
3879 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3880 should be (const char *).
3881
3882 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3883 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3884 However, I haven't made this a standard facility. The documentation doesn't
3885 mention it, and the Makefile doesn't support it.
3886
3887 5. The Makefile now contains an "install" target, with editable destinations at
3888 the top of the file. The pcretest program is not installed.
3889
3890 6. pgrep -V now gives the PCRE version number and date.
3891
3892 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3893 causing the entire string to be ignored, instead of just the last character.
3894
3895 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3896 non-matching string, it can take a very, very long time, even for strings of
3897 quite modest length, because of the nested recursion. PCRE now does better in
3898 some of these cases. It does this by remembering the last required literal
3899 character in the pattern, and pre-searching the subject to ensure it is present
3900 before running the real match. In other words, it applies a heuristic to detect
3901 some types of certain failure quickly, and in the above example, if presented
3902 with a string that has no trailing " it gives "no match" very quickly.
3903
3904 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3905 other alternatives are tried instead.
3906
3907
3908 Version 2.06 09-Jun-99
3909 ----------------------
3910
3911 1. Change pcretest's output for amount of store used to show just the code
3912 space, because the remainder (the data block) varies in size between 32-bit and
3913 64-bit systems.
3914
3915 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3916 start matching at. This allows lookbehinds to work when searching for multiple
3917 occurrences in a string.
3918
3919 3. Added additional options to pcretest for testing multiple occurrences:
3920
3921 /+ outputs the rest of the string that follows a match
3922 /g loops for multiple occurrences, using the new startoffset argument
3923 /G loops for multiple occurrences by passing an incremented pointer
3924
3925 4. PCRE wasn't doing the "first character" optimization for patterns starting
3926 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3927 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3928 the letter 'x'. On long subject strings, this gives a significant speed-up.
3929
3930
3931 Version 2.05 21-Apr-99
3932 ----------------------
3933
3934 1. Changed the type of magic_number from int to long int so that it works
3935 properly on 16-bit systems.
3936
3937 2. Fixed a bug which caused patterns starting with .* not to work correctly
3938 when the subject string contained newline characters. PCRE was assuming
3939 anchoring for such patterns in all cases, which is not correct because .* will
3940 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3941 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3942 must be retried after every newline in the subject.
3943
3944
3945 Version 2.04 18-Feb-99
3946 ----------------------
3947
3948 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3949 computation of the store needed to hold the pattern was incorrect (too large).
3950 If such patterns were nested a few deep, this could multiply and become a real
3951 problem.
3952
3953 2. Added /M option to pcretest to show the memory requirement of a specific
3954 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3955
3956 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3957 compiled in such a way that the backtracking after subsequent failure was
3958 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3959 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3960
3961
3962 Version 2.03 02-Feb-99
3963 ----------------------
3964
3965 1. Fixed typo and small mistake in man page.
3966
3967 2. Added 4th condition (GPL supersedes if conflict) and created separate
3968 LICENCE file containing the conditions.
3969
3970 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3971 Perl, that is the internal \ allows the delimiter to be included in the
3972 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3973 the final delimiter, add \ to the end of the pattern (to test the error).
3974
3975 4. Added the convenience functions for extracting substrings after a successful
3976 match. Updated pcretest to make it able to test these functions.
3977
3978
3979 Version 2.02 14-Jan-99
3980 ----------------------
3981
3982 1. Initialized the working variables associated with each extraction so that
3983 their saving and restoring doesn't refer to uninitialized store.
3984
3985 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3986 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3987 fix the problem.
3988
3989 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3990 calls, and wasn't printing the correct value for compiling calls. Increased the
3991 default value of LOOPREPEAT, and the number of significant figures in the
3992 times.
3993
3994 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3995
3996 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3997 a building problem on Windows NT with a FAT file system.
3998
3999
4000 Version 2.01 21-Oct-98
4001 ----------------------
4002
4003 1. Changed the API for pcre_compile() to allow for the provision of a pointer
4004 to character tables built by pcre_maketables() in the current locale. If NULL
4005 is passed, the default tables are used.
4006
4007
4008 Version 2.00 24-Sep-98
4009 ----------------------
4010
4011 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
4012 it any more.
4013
4014 2. Allow quantification of (?>) groups, and make it work correctly.
4015
4016 3. The first character computation wasn't working for (?>) groups.
4017
4018 4. Correct the implementation of \Z (it is permitted to match on the \n at the
4019 end of the subject) and add 5.005's \z, which really does match only at the
4020 very end of the subject.
4021
4022 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
4023
4024 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
4025 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
4026 localized options. All options to pcre_study() were also removed.
4027
4028 7. Add other new features from 5.005:
4029
4030 $(?<= positive lookbehind
4031 $(?<! negative lookbehind
4032 (?imsx-imsx) added the unsetting capability
4033 such a setting is global if at outer level; local otherwise
4034 (?imsx-imsx:) non-capturing groups with option setting
4035 (?(cond)re|re) conditional pattern matching
4036
4037 A backreference to itself in a repeated group matches the previous
4038 captured string.
4039
4040 8. General tidying up of studying (both automatic and via "study")
4041 consequential on the addition of new assertions.
4042
4043 9. As in 5.005, unlimited repeated groups that could match an empty substring
4044 are no longer faulted at compile time. Instead, the loop is forcibly broken at
4045 runtime if any iteration does actually match an empty substring.
4046
4047 10. Include the RunTest script in the distribution.
4048
4049 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
4050 discrepancies, some of which were old and were also with respect to 5.004. They
4051 have now been fixed.
4052
4053
4054 Version 1.09 28-Apr-98
4055 ----------------------
4056
4057 1. A negated single character class followed by a quantifier with a minimum
4058 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
4059 program crashes, or just wrong answers. This did not apply to negated classes
4060 containing more than one character, or to minima other than one.
4061
4062
4063 Version 1.08 27-Mar-98
4064 ----------------------
4065
4066 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
4067
4068 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
4069 latter must appear before anything that relies on it in the pattern.
4070
4071
4072 Version 1.07 16-Feb-98
4073 ----------------------
4074
4075 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
4076 repeat of a potentially empty string).
4077
4078
4079 Version 1.06 23-Jan-98
4080 ----------------------
4081
4082 1. Added Markus Oberhumer's little patches for C++.
4083
4084 2. Literal strings longer than 255 characters were broken.
4085
4086
4087 Version 1.05 23-Dec-97
4088 ----------------------
4089
4090 1. Negated character classes containing more than one character were failing if
4091 PCRE_CASELESS was set at run time.
4092
4093
4094 Version 1.04 19-Dec-97
4095 ----------------------
4096
4097 1. Corrected the man page, where some "const" qualifiers had been omitted.
4098
4099 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4100 input syntax.
4101
4102 3. Fixed memory leak which occurred when a regex with back references was
4103 matched with an offsets vector that wasn't big enough. The temporary memory
4104 that is used in this case wasn't being freed if the match failed.
4105
4106 4. Tidied pcretest to ensure it frees memory that it gets.
4107
4108 5. Temporary memory was being obtained in the case where the passed offsets
4109 vector was exactly big enough.
4110
4111 6. Corrected definition of offsetof() from change 5 below.
4112
4113 7. I had screwed up change 6 below and broken the rules for the use of
4114 setjmp(). Now fixed.
4115
4116
4117 Version 1.03 18-Dec-97
4118 ----------------------
4119
4120 1. A erroneous regex with a missing opening parenthesis was correctly
4121 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4122 on some systems.
4123
4124 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4125 it was reported that one broken compiler failed on the former because "code" is
4126 also an independent variable.
4127
4128 3. The erroneous regex a[]b caused an array overrun reference.
4129
4130 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4131 fail on data ending with that character. (It was going on too far, and checking
4132 the next character, typically a binary zero.) This was specific to the
4133 optimized code for single-character negative classes.
4134
4135 5. Added a contributed patch from the TIN world which does the following:
4136
4137 + Add an undef for memmove, in case the the system defines a macro for it.
4138
4139 + Add a definition of offsetof(), in case there isn't one. (I don't know
4140 the reason behind this - offsetof() is part of the ANSI standard - but
4141 it does no harm).
4142
4143 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4144 most of the places where whitespace preceded '#'. I have given up and
4145 allowed the remaining 2 cases to be at the margin.
4146
4147 + Rename some variables in pcre to eliminate shadowing. This seems very
4148 pedantic, but does no harm, of course.
4149
4150 6. Moved the call to setjmp() into its own function, to get rid of warnings
4151 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4152
4153 7. Constructs such as \d{8,} were compiling into the equivalent of
4154 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4155 outcome, but in this particular case used more store than had been allocated,
4156 which caused the bug to be discovered because it threw up an internal error.
4157
4158 8. The debugging code in both pcre and pcretest for outputting the compiled
4159 form of a regex was going wrong in the case of back references followed by
4160 curly-bracketed repeats.
4161
4162
4163 Version 1.02 12-Dec-97
4164 ----------------------
4165
4166 1. Typos in pcre.3 and comments in the source fixed.
4167
4168 2. Applied a contributed patch to get rid of places where it used to remove
4169 'const' from variables, and fixed some signed/unsigned and uninitialized
4170 variable warnings.
4171
4172 3. Added the "runtest" target to Makefile.
4173
4174 4. Set default compiler flag to -O2 rather than just -O.
4175
4176
4177 Version 1.01 19-Nov-97
4178 ----------------------
4179
4180 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4181 like /([ab]*)*/, that is, for classes with more than one character in them.
4182
4183 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4184 as /((?>a*))*/ (a PCRE_EXTRA facility).
4185
4186
4187 Version 1.00 18-Nov-97
4188 ----------------------
4189
4190 1. Added compile-time macros to support systems such as SunOS4 which don't have
4191 memmove() or strerror() but have other things that can be used instead.
4192
4193 2. Arranged that "make clean" removes the executables.
4194
4195
4196 Version 0.99 27-Oct-97
4197 ----------------------
4198
4199 1. Fixed bug in code for optimizing classes with only one character. It was
4200 initializing a 32-byte map regardless, which could cause it to run off the end
4201 of the memory it had got.
4202
4203 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4204
4205
4206 Version 0.98 22-Oct-97
4207 ----------------------
4208
4209 1. Fixed bug in code for handling temporary memory usage when there are more
4210 back references than supplied space in the ovector. This could cause segfaults.
4211
4212
4213 Version 0.97 21-Oct-97
4214 ----------------------
4215
4216 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4217
4218 2. Optimized negated single characters not to use a bit map.
4219
4220 3. Brought error texts together as macro definitions; clarified some of them;
4221 fixed one that was wrong - it said "range out of order" when it meant "invalid
4222 escape sequence".
4223
4224 4. Changed some char * arguments to const char *.
4225
4226 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4227
4228 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4229 pcretest.
4230
4231
4232 Version 0.96 16-Oct-97
4233 ----------------------
4234
4235 1. Added a simple "pgrep" utility to the distribution.
4236
4237 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4238 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4239 where "ddd" means "one or more decimal digits".
4240
4241 3. Fixed serious bug. If a pattern had a back reference, but the call to
4242 pcre_exec() didn't supply a large enough ovector to record the related
4243 identifying subpattern, the match always failed. PCRE now remembers the number
4244 of the largest back reference, and gets some temporary memory in which to save
4245 the offsets during matching if necessary, in order to ensure that
4246 backreferences always work.
4247
4248 4. Increased the compatibility with Perl in a number of ways:
4249
4250 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4251 to request this handling. The option can be set at compile or exec time.
4252
4253 (b) $ matches before a terminating newline by default; an option
4254 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4255 mode). The option can be set at compile or exec time.
4256
4257 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4258 the same as Perl's. If the decimal number it represents is less than 10
4259 or there aren't that many previous left capturing parentheses, an octal
4260 escape is read. Inside a character class, it's always an octal escape,
4261 even if it is a single digit.
4262
4263 (d) An escaped but undefined alphabetic character is taken as a literal,
4264 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4265 escapes.
4266
4267 (e) {0} is now permitted. (The previous item is removed from the compiled
4268 pattern).
4269
4270 5. Changed all the names of code files so that the basic parts are no longer
4271 than 10 characters, and abolished the teeny "globals.c" file.
4272
4273 6. Changed the handling of character classes; they are now done with a 32-byte
4274 bit map always.
4275
4276 7. Added the -d and /D options to pcretest to make it possible to look at the
4277 internals of compilation without having to recompile pcre.
4278
4279
4280 Version 0.95 23-Sep-97
4281 ----------------------
4282
4283 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4284 \x20 at the start of a run of normal characters. These were being treated as
4285 real characters, instead of the source characters being re-checked.
4286
4287
4288 Version 0.94 18-Sep-97
4289 ----------------------
4290
4291 1. The functions are now thread-safe, with the caveat that the global variables
4292 containing pointers to malloc() and free() or alternative functions are the
4293 same for all threads.
4294
4295 2. Get pcre_study() to generate a bitmap of initial characters for non-
4296 anchored patterns when this is possible, and use it if passed to pcre_exec().
4297
4298
4299 Version 0.93 15-Sep-97
4300 ----------------------
4301
4302 1. /(b)|(:+)/ was computing an incorrect first character.
4303
4304 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4305 but not actually doing anything yet.
4306
4307 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4308 as Perl does (e.g. [-az] or [az-]).
4309
4310 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4311 all possible positions.
4312
4313 5. Split up into different modules to avoid including unneeded functions in a
4314 compiled binary. However, compile and exec are still in one module. The "study"
4315 function is split off.
4316
4317 6. The character tables are now in a separate module whose source is generated
4318 by an auxiliary program - but can then be edited by hand if required. There are
4319 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4320 toupper() in the code.
4321
4322 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4323 make them global. Abolish the function for setting them, as the caller can now
4324 set them directly.
4325
4326
4327 Version 0.92 11-Sep-97
4328 ----------------------
4329
4330 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4331 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4332
4333 2. Caseless matching was not working in character classes if the characters in
4334 the pattern were in upper case.
4335
4336 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4337
4338 4. Make PCRE_ANCHORED public and accept as a compile option.
4339
4340 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4341 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4342 pass them.
4343
4344 6. Give an error if bad option bits passed at compile or run time.
4345
4346 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4347 pcretest to cause it to pass that flag.
4348
4349 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4350 options, and the first character, if set.
4351
4352 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4353
4354
4355 Version 0.91 10-Sep-97
4356 ----------------------
4357
4358 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4359 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4360
4361 2. PCRE was looping on encountering an indefinitely repeated back reference to
4362 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4363 Perl does - treats the match as successful.
4364
4365 ****

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5