ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 637 - (show annotations)
Sun Jul 24 17:44:12 2011 UTC (9 years, 9 months ago) by ph10
File size: 208057 byte(s)
Error occurred while calculating annotation data.
Permit quantifiers on parenthesized assertions.
1 ChangeLog for PCRE
2 ------------------
4 Version 8.13 30-Apr-2011
5 ------------------------
7 1. The Unicode data tables have been updated to Unicode 6.0.0.
9 2. Two minor typos in pcre_internal.h have been fixed.
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).
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).
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.
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().
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.
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.
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.
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().
60 11. While implementing 10, a number of bugs in the handling of groups were
61 discovered and fixed:
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.
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.
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.
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).
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.
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.
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.
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.
107 19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108 was incorrectly computed.
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.
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.
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.
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).
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.
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.
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.
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.
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.
155 28. Some minor code refactoring concerning Unicode properties and scripts
156 should reduce the stack requirement of match() slightly.
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.
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).
166 31. A *MARK encountered during the processing of a positive assertion is now
167 recorded and passed back (compatible with Perl).
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.
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.
182 Version 8.12 15-Jan-2011
183 ------------------------
185 1. Fixed some typos in the markup of the man pages, and wrote a script that
186 checks for such things as part of the documentation building process.
188 2. On a big-endian 64-bit system, pcregrep did not correctly process the
189 --match-limit and --recursion-limit options (added for 8.11). In
190 particular, this made one of the standard tests fail. (The integer value
191 went into the wrong half of a long int.)
193 3. If the --colour option was given to pcregrep with -v (invert match), it
194 did strange things, either producing crazy output, or crashing. It should,
195 of course, ignore a request for colour when reporting lines that do not
196 match.
198 4. Another pcregrep bug caused similar problems if --colour was specified with
199 -M (multiline) and the pattern match finished with a line ending.
201 5. In pcregrep, when a pattern that ended with a literal newline sequence was
202 matched in multiline mode, the following line was shown as part of the
203 match. This seems wrong, so I have changed it.
205 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
206 the check for further matches in the same line (so they could be coloured)
207 to overrun the end of the current line. If another match was found, it was
208 incorrectly shown (and then shown again when found in the next line).
210 7. If pcregrep was compiled under Windows, there was a reference to the
211 function pcregrep_exit() before it was defined. I am assuming this was
212 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
213 reported by a user. I've moved the definition above the reference.
216 Version 8.11 10-Dec-2010
217 ------------------------
219 1. (*THEN) was not working properly if there were untried alternatives prior
220 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
221 backtracked to try for "b" instead of moving to the next alternative branch
222 at the same level (in this case, to look for "c"). The Perl documentation
223 is clear that when (*THEN) is backtracked onto, it goes to the "next
224 alternative in the innermost enclosing group".
226 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
227 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
228 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
229 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
230 (*THEN).
232 3. If \s appeared in a character class, it removed the VT character from
233 the class, even if it had been included by some previous item, for example
234 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
235 of \s, but is part of the POSIX "space" class.)
237 4. A partial match never returns an empty string (because you can always
238 match an empty string at the end of the subject); however the checking for
239 an empty string was starting at the "start of match" point. This has been
240 changed to the "earliest inspected character" point, because the returned
241 data for a partial match starts at this character. This means that, for
242 example, /(?<=abc)def/ gives a partial match for the subject "abc"
243 (previously it gave "no match").
245 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
246 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
247 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
248 has an implication that the given string is incomplete (because a partial
249 match is preferred over a full match). For this reason, these items now
250 give a partial match in this situation. [Aside: previously, the one case
251 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
252 match rather than a full match, which was wrong by the old rules, but is
253 now correct.]
255 6. There was a bug in the handling of #-introduced comments, recognized when
256 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
257 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
258 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
259 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
260 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
261 places in pcre_compile().
263 7. Related to (6) above, when pcre_compile() was skipping #-introduced
264 comments when looking ahead for named forward references to subpatterns,
265 the only newline sequence it recognized was NL. It now handles newlines
266 according to the set newline convention.
268 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
269 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
270 cater for a lack of strerror(). These oversights have been fixed.
272 9. Added --match-limit and --recursion-limit to pcregrep.
274 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
276 11. When the -o option was used, pcregrep was setting a return code of 1, even
277 when matches were found, and --line-buffered was not being honoured.
279 12. Added an optional parentheses number to the -o and --only-matching options
280 of pcregrep.
282 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
283 can match an empty string. The code to do it in pcretest and pcredemo
284 needed fixing:
286 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
287 only one byte after an empty string match just before CRLF (this case
288 just got forgotten; "any" and "anycrlf" were OK).
290 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
291 mode when an empty string match preceded an ASCII character followed by
292 a non-ASCII character. (The code for advancing by one character rather
293 than one byte was nonsense.)
295 (c) The pcredemo.c sample program did not have any code at all to handle
296 the cases when CRLF is a valid newline sequence.
298 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
299 as a starting offset was within the subject string. There is now a new
300 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
301 negative or greater than the length of the string. In order to test this,
302 pcretest is extended to allow the setting of negative starting offsets.
304 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
305 starting offset points to the beginning of a UTF-8 character was
306 unnecessarily clumsy. I tidied it up.
308 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
309 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
311 17. Nobody had reported that the --include_dir option, which was added in
312 release 7.7 should have been called --include-dir (hyphen, not underscore)
313 for compatibility with GNU grep. I have changed it to --include-dir, but
314 left --include_dir as an undocumented synonym, and the same for
315 --exclude-dir, though that is not available in GNU grep, at least as of
316 release 2.5.4.
318 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
319 characters from a string of bytes) have been redefined so as not to use
320 loops, in order to improve performance in some environments. At the same
321 time, I abstracted some of the common code into auxiliary macros to save
322 repetition (this should not affect the compiled code).
324 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
325 compile-time error is now given if \c is not followed by an ASCII
326 character, that is, a byte less than 128. (In EBCDIC mode, the code is
327 different, and any byte value is allowed.)
329 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
330 START_OPTIMIZE option, which is now allowed at compile time - but just
331 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
332 to pcregrep and other applications that have no direct access to PCRE
333 options. The new /Y option in pcretest sets this option when calling
334 pcre_compile().
336 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
337 back references. Groups containing recursive back references were forced to
338 be atomic by that change, but in the case of named groups, the amount of
339 memory required was incorrectly computed, leading to "Failed: internal
340 error: code overflow". This has been fixed.
342 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
343 pcretest.c, to avoid build problems in some Borland environments.
346 Version 8.10 25-Jun-2010
347 ------------------------
349 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
350 THEN.
352 2. (*ACCEPT) was not working when inside an atomic group.
354 3. Inside a character class, \B is treated as a literal by default, but
355 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
356 causes the error). The code is unchanged, but I tidied the documentation.
358 4. Inside a character class, PCRE always treated \R and \X as literals,
359 whereas Perl faults them if its -w option is set. I have changed PCRE so
360 that it faults them when PCRE_EXTRA is set.
362 5. Added support for \N, which always matches any character other than
363 newline. (It is the same as "." when PCRE_DOTALL is not set.)
365 6. When compiling pcregrep with newer versions of gcc which may have
366 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
367 declared with attribute warn_unused_result" were given. Just casting the
368 result to (void) does not stop the warnings; a more elaborate fudge is
369 needed. I've used a macro to implement this.
371 7. Minor change to pcretest.c to avoid a compiler warning.
373 8. Added four artifical Unicode properties to help with an option to make
374 \s etc use properties (see next item). The new properties are: Xan
375 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
377 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
378 use Unicode properties. (*UCP) at the start of a pattern can be used to set
379 this option. Modified pcretest to add /W to test this facility. Added
380 REG_UCP to make it available via the POSIX interface.
382 10. Added --line-buffered to pcregrep.
384 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
385 studied, and the match started with a letter with a code point greater than
386 127 whose first byte was different to the first byte of the other case of
387 the letter, the other case of this starting letter was not recognized
388 (#976).
390 12. If a pattern that was studied started with a repeated Unicode property
391 test, for example, \p{Nd}+, there was the theoretical possibility of
392 setting up an incorrect bitmap of starting bytes, but fortunately it could
393 not have actually happened in practice until change 8 above was made (it
394 added property types that matched character-matching opcodes).
396 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
397 possible starting bytes for non-anchored patterns.
399 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
400 \R, and also a number of cases that involve Unicode properties, both
401 explicit and implicit when PCRE_UCP is set.
403 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
404 input, it could crash or give wrong results if characters with values
405 greater than 0xc0 were present in the subject string. (Detail: it assumed
406 UTF-8 input when processing these items.)
408 16. Added a lot of (int) casts to avoid compiler warnings in systems where
409 size_t is 64-bit (#991).
411 17. Added a check for running out of memory when PCRE is compiled with
412 --disable-stack-for-recursion (#990).
414 18. If the last data line in a file for pcretest does not have a newline on
415 the end, a newline was missing in the output.
417 19. The default pcre_chartables.c file recognizes only ASCII characters (values
418 less than 128) in its various bitmaps. However, there is a facility for
419 generating tables according to the current locale when PCRE is compiled. It
420 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
421 characters, are recognized by isspace() and therefore were getting set in
422 these tables, and indeed these tables seem to approximate to ISO 8859. This
423 caused a problem in UTF-8 mode when pcre_study() was used to create a list
424 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
425 which of course cannot start UTF-8 characters. I have changed the code so
426 that only real ASCII characters (less than 128) and the correct starting
427 bytes for UTF-8 encodings are set for characters greater than 127 when in
428 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
429 altogether.)
431 20. Added the /T option to pcretest so as to be able to run tests with non-
432 standard character tables, thus making it possible to include the tests
433 used for 19 above in the standard set of tests.
435 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
436 reference to a subpattern the other side of a comment that contains an
437 opening parenthesis caused either an internal compiling error, or a
438 reference to the wrong subpattern.
441 Version 8.02 19-Mar-2010
442 ------------------------
444 1. The Unicode data tables have been updated to Unicode 5.2.0.
446 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
447 configured.
449 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
450 original author of that file, following a query about its status.
452 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
453 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
455 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
456 quantifier applied to a forward-referencing subroutine call, could compile
457 incorrect code or give the error "internal error: previously-checked
458 referenced subpattern not found".
460 6. Both MS Visual Studio and Symbian OS have problems with initializing
461 variables to point to external functions. For these systems, therefore,
462 pcre_malloc etc. are now initialized to local functions that call the
463 relevant global functions.
465 7. There were two entries missing in the vectors called coptable and poptable
466 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
467 I've fixed the data, and added a kludgy way of testing at compile time that
468 the lengths are correct (equal to the number of opcodes).
470 8. Following on from 7, I added a similar kludge to check the length of the
471 eint vector in pcreposix.c.
473 9. Error texts for pcre_compile() are held as one long string to avoid too
474 much relocation at load time. To find a text, the string is searched,
475 counting zeros. There was no check for running off the end of the string,
476 which could happen if a new error number was added without updating the
477 string.
479 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
481 11. \K was not working if it appeared in an atomic group or in a group that
482 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
483 \K is "not well defined" if used in an assertion. PCRE now accepts it if
484 the assertion is positive, but not if it is negative.
486 12. Change 11 fortuitously reduced the size of the stack frame used in the
487 "match()" function of pcre_exec.c by one pointer. Forthcoming
488 implementation of support for (*MARK) will need an extra pointer on the
489 stack; I have reserved it now, so that the stack frame size does not
490 decrease.
492 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
493 item in branch that calls a recursion is a subroutine call - as in the
494 second branch in the above example - was incorrectly given the compile-
495 time error "recursive call could loop indefinitely" because pcre_compile()
496 was not correctly checking the subroutine for matching a non-empty string.
498 14. The checks for overrunning compiling workspace could trigger after an
499 overrun had occurred. This is a "should never occur" error, but it can be
500 triggered by pathological patterns such as hundreds of nested parentheses.
501 The checks now trigger 100 bytes before the end of the workspace.
503 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
506 Version 8.01 19-Jan-2010
507 ------------------------
509 1. If a pattern contained a conditional subpattern with only one branch (in
510 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
511 computed the wrong minimum data length (which is of course zero for such
512 subpatterns). This could cause incorrect "no match" results.
514 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
515 the pattern is reset in the first branch, pcre_compile() failed with
516 "internal error: code overflow at offset...". This happened only when
517 the reset was to the original external option setting. (An optimization
518 abstracts leading options settings into an external setting, which was the
519 cause of this.)
521 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
522 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
523 assertion pattern did not match (meaning that the assertion was true), it
524 was incorrectly treated as false if the SKIP had been reached during the
525 matching. This also applied to assertions used as conditions.
527 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
528 assertion subpattern, including such a pattern used as a condition,
529 unpredictable results occurred, instead of the error return
532 5. The C++ GlobalReplace function was not working like Perl for the special
533 situation when an empty string is matched. It now does the fancy magic
534 stuff that is necessary.
536 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
537 removed. (These were left over from very, very early versions of PCRE.)
539 7. Some cosmetic changes to the code to make life easier when compiling it
540 as part of something else:
542 (a) Change DEBUG to PCRE_DEBUG.
544 (b) In pcre_compile(), rename the member of the "branch_chain" structure
545 called "current" as "current_branch", to prevent a collision with the
546 Linux macro when compiled as a kernel module.
548 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
549 prevent a collision with the Linux macro when compiled as a kernel
550 module.
552 8. In pcre_compile() there are some checks for integer overflows that used to
553 cast potentially large values to (double). This has been changed to that
554 when building, a check for int64_t is made, and if it is found, it is used
555 instead, thus avoiding the use of floating point arithmetic. (There is no
556 other use of FP in PCRE.) If int64_t is not found, the fallback is to
557 double.
559 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
560 2005 (difference between two addresses compared to an unsigned value).
562 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
563 custom one, because of the following reported problem in Windows:
565 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
566 under Win32.
567 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
568 therefore missing the function definition.
569 - The compiler thus generates a "C" signature for the test function.
570 - The linker fails to find the "C" function.
571 - PCRE fails to configure if asked to do so against libbz2.
573 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
574 messages were output:
576 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
577 rerunning libtoolize, to keep the correct libtool macros in-tree.
578 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
580 I have done both of these things.
582 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
583 most of the time, it *can* run out if it is given a pattern that contains a
584 runaway infinite recursion. I updated the discussion in the pcrestack man
585 page.
587 13. Now that we have gone to the x.xx style of version numbers, the minor
588 version may start with zero. Using 08 or 09 is a bad idea because users
589 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
590 interpreted as invalid octal numbers. I've updated the previous comment in
591 configure.ac, and also added a check that gives an error if 08 or 09 are
592 used.
594 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
595 causing partial matching to fail when the end of the subject matched \W
596 in a UTF-8 pattern where \W was quantified with a minimum of 3.
598 15. There were some discrepancies between the declarations in pcre_internal.h
599 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
600 their definitions. The declarations used "const uschar *" and the
601 definitions used USPTR. Even though USPTR is normally defined as "const
602 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
603 reported that: "This difference in casting confuses some C++ compilers, for
604 example, SunCC recognizes above declarations as different functions and
605 generates broken code for hbpcre." I have changed the declarations to use
606 USPTR.
608 16. GNU libtool is named differently on some systems. The autogen.sh script now
609 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
610 (FreeBSD).
612 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
613 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
614 comment: "Figure out how to create a longlong from a string: strtoll and
615 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
616 instance, but it only takes 2 args instead of 3!"
618 18. A subtle bug concerned with back references has been fixed by a change of
619 specification, with a corresponding code fix. A pattern such as
620 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
621 refers, was giving matches when it shouldn't. For example, xa=xaaa would
622 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
623 same bug. Such groups have to be quantified to be useful, or contained
624 inside another quantified group. (If there's no repetition, the reference
625 can never match.) The problem arises because, having left the group and
626 moved on to the rest of the pattern, a later failure that backtracks into
627 the group uses the captured value from the final iteration of the group
628 rather than the correct earlier one. I have fixed this in PCRE by forcing
629 any group that contains a reference to itself to be an atomic group; that
630 is, there cannot be any backtracking into it once it has completed. This is
631 similar to recursive and subroutine calls.
634 Version 8.00 19-Oct-09
635 ----------------------
637 1. The table for translating pcre_compile() error codes into POSIX error codes
638 was out-of-date, and there was no check on the pcre_compile() error code
639 being within the table. This could lead to an OK return being given in
640 error.
642 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
643 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
644 in a Windows environment.
646 3. The pcregrep --count option prints the count for each file even when it is
647 zero, as does GNU grep. However, pcregrep was also printing all files when
648 --files-with-matches was added. Now, when both options are given, it prints
649 counts only for those files that have at least one match. (GNU grep just
650 prints the file name in this circumstance, but including the count seems
651 more useful - otherwise, why use --count?) Also ensured that the
652 combination -clh just lists non-zero counts, with no names.
654 4. The long form of the pcregrep -F option was incorrectly implemented as
655 --fixed_strings instead of --fixed-strings. This is an incompatible change,
656 but it seems right to fix it, and I didn't think it was worth preserving
657 the old behaviour.
659 5. The command line items --regex=pattern and --regexp=pattern were not
660 recognized by pcregrep, which required --regex pattern or --regexp pattern
661 (with a space rather than an '='). The man page documented the '=' forms,
662 which are compatible with GNU grep; these now work.
664 6. No libpcreposix.pc file was created for pkg-config; there was just
665 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
667 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
668 when UCP support is not needed, by modifying the Python script that
669 generates it from Unicode data files. This should not matter if the module
670 is correctly used as a library, but I received one complaint about 50K of
671 unwanted data. My guess is that the person linked everything into his
672 program rather than using a library. Anyway, it does no harm.
674 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
675 was a minimum greater than 1 for a wide character in a possessive
676 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
677 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
678 character. Chaos in the form of incorrect output or a compiling loop could
679 result.
681 9. The restrictions on what a pattern can contain when partial matching is
682 requested for pcre_exec() have been removed. All patterns can now be
683 partially matched by this function. In addition, if there are at least two
684 slots in the offset vector, the offset of the earliest inspected character
685 for the match and the offset of the end of the subject are set in them when
686 PCRE_ERROR_PARTIAL is returned.
688 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
689 synonymous with PCRE_PARTIAL, for backwards compatibility, and
690 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
691 and may be more useful for multi-segment matching.
693 11. Partial matching with pcre_exec() is now more intuitive. A partial match
694 used to be given if ever the end of the subject was reached; now it is
695 given only if matching could not proceed because another character was
696 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
697 string "Z", which now yields "no match" instead of "partial match". In the
698 case of pcre_dfa_exec(), "no match" is given if every matching path for the
699 final character ended with (*FAIL).
701 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
702 if the pattern had a "must contain" character that was already found in the
703 earlier partial match, unless partial matching was again requested. For
704 example, with the pattern /dog.(body)?/, the "must contain" character is
705 "g". If the first part-match was for the string "dog", restarting with
706 "sbody" failed. This bug has been fixed.
708 13. The string returned by pcre_dfa_exec() after a partial match has been
709 changed so that it starts at the first inspected character rather than the
710 first character of the match. This makes a difference only if the pattern
711 starts with a lookbehind assertion or \b or \B (\K is not supported by
712 pcre_dfa_exec()). It's an incompatible change, but it makes the two
713 matching functions compatible, and I think it's the right thing to do.
715 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
716 so that the demonstration program is easily available in environments where
717 PCRE has not been installed from source.
719 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
720 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
721 library.
723 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
724 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
725 is not the first non-POSIX option to be added. Clearly some people find
726 these options useful.
728 17. If a caller to the POSIX matching function regexec() passes a non-zero
729 value for nmatch with a NULL value for pmatch, the value of
730 nmatch is forced to zero.
732 18. RunGrepTest did not have a test for the availability of the -u option of
733 the diff command, as RunTest does. It now checks in the same way as
734 RunTest, and also checks for the -b option.
736 19. If an odd number of negated classes containing just a single character
737 interposed, within parentheses, between a forward reference to a named
738 subpattern and the definition of the subpattern, compilation crashed with
739 an internal error, complaining that it could not find the referenced
740 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
741 [The bug was that it was starting one character too far in when skipping
742 over the character class, thus treating the ] as data rather than
743 terminating the class. This meant it could skip too much.]
745 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
746 /g option in pcretest when the pattern contains \K, which makes it possible
747 to have an empty string match not at the start, even when the pattern is
748 anchored. Updated pcretest and pcredemo to use this option.
750 21. If the maximum number of capturing subpatterns in a recursion was greater
751 than the maximum at the outer level, the higher number was returned, but
752 with unset values at the outer level. The correct (outer level) value is
753 now given.
755 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
756 PCRE did not set those parentheses (unlike Perl). I have now found a way to
757 make it do so. The string so far is captured, making this feature
758 compatible with Perl.
760 23. The tests have been re-organized, adding tests 11 and 12, to make it
761 possible to check the Perl 5.10 features against Perl 5.10.
763 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
764 pattern matches a fixed length string. PCRE did not allow this; now it
765 does. Neither allows recursion.
767 25. I finally figured out how to implement a request to provide the minimum
768 length of subject string that was needed in order to match a given pattern.
769 (It was back references and recursion that I had previously got hung up
770 on.) This code has now been added to pcre_study(); it finds a lower bound
771 to the length of subject needed. It is not necessarily the greatest lower
772 bound, but using it to avoid searching strings that are too short does give
773 some useful speed-ups. The value is available to calling programs via
774 pcre_fullinfo().
776 26. While implementing 25, I discovered to my embarrassment that pcretest had
777 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
778 study optimizations had never been tested with that matching function.
779 Oops. What is worse, even when it was passed study data, there was a bug in
780 pcre_dfa_exec() that meant it never actually used it. Double oops. There
781 were also very few tests of studied patterns with pcre_dfa_exec().
783 27. If (?| is used to create subpatterns with duplicate numbers, they are now
784 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
785 on the other side of the coin, they are no longer allowed to have different
786 names, because these cannot be distinguished in PCRE, and this has caused
787 confusion. (This is a difference from Perl.)
789 28. When duplicate subpattern names are present (necessarily with different
790 numbers, as required by 27 above), and a test is made by name in a
791 conditional pattern, either for a subpattern having been matched, or for
792 recursion in such a pattern, all the associated numbered subpatterns are
793 tested, and the overall condition is true if the condition is true for any
794 one of them. This is the way Perl works, and is also more like the way
795 testing by number works.
798 Version 7.9 11-Apr-09
799 ---------------------
801 1. When building with support for bzlib/zlib (pcregrep) and/or readline
802 (pcretest), all targets were linked against these libraries. This included
803 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
804 libraries. This caused unwanted dependencies to be created. This problem
805 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
806 pcretest is linked with readline.
808 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
809 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
810 moved outside it again, because FALSE and TRUE are already defined in AIX,
811 but BOOL is not.
813 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
814 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
816 4. The pcregrep documentation said spaces were inserted as well as colons (or
817 hyphens) following file names and line numbers when outputting matching
818 lines. This is not true; no spaces are inserted. I have also clarified the
819 wording for the --colour (or --color) option.
821 5. In pcregrep, when --colour was used with -o, the list of matching strings
822 was not coloured; this is different to GNU grep, so I have changed it to be
823 the same.
825 6. When --colo(u)r was used in pcregrep, only the first matching substring in
826 each matching line was coloured. Now it goes on to look for further matches
827 of any of the test patterns, which is the same behaviour as GNU grep.
829 7. A pattern that could match an empty string could cause pcregrep to loop; it
830 doesn't make sense to accept an empty string match in pcregrep, so I have
831 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
832 seems to be how GNU grep behaves.
834 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
835 start or after a newline", because the conditional assertion was not being
836 correctly handled. The rule now is that both the assertion and what follows
837 in the first alternative must satisfy the test.
839 9. If auto-callout was enabled in a pattern with a conditional group whose
840 condition was an assertion, PCRE could crash during matching, both with
841 pcre_exec() and pcre_dfa_exec().
843 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
844 used for matching.
846 11. Unicode property support in character classes was not working for
847 characters (bytes) greater than 127 when not in UTF-8 mode.
849 12. Added the -M command line option to pcretest.
851 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
853 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
855 16. Added comments and documentation about mis-use of no_arg in the C++
856 wrapper.
858 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
859 from Martin Jerabek that uses macro names for all relevant character and
860 string constants.
862 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
863 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
864 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
865 these, but not everybody uses configure.
867 19. A conditional group that had only one branch was not being correctly
868 recognized as an item that could match an empty string. This meant that an
869 enclosing group might also not be so recognized, causing infinite looping
870 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
871 with the subject "ab", where knowledge that the repeated group can match
872 nothing is needed in order to break the loop.
874 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
875 exec(), but without supplying a callout function, matching went wrong.
877 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
878 leak if the size of the offset vector was greater than 30. When the vector
879 is smaller, the saved offsets during recursion go onto a local stack
880 vector, but for larger vectors malloc() is used. It was failing to free
881 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
882 error, in fact).
884 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
885 heapframe that is used only when UTF-8 support is enabled. This caused no
886 problem, but was untidy.
888 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
889 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
890 included within another project.
892 24. Steven Van Ingelgem's patches to add more options to the CMake support,
893 slightly modified by me:
895 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
896 not building pcregrep.
898 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
899 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
901 25. Forward references, both numeric and by name, in patterns that made use of
902 duplicate group numbers, could behave incorrectly or give incorrect errors,
903 because when scanning forward to find the reference group, PCRE was not
904 taking into account the duplicate group numbers. A pattern such as
905 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
907 26. Changed a few more instances of "const unsigned char *" to USPTR, making
908 the feature of a custom pointer more persuasive (as requested by a user).
910 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
911 pcretest.c, inside #ifndefs, because it seems they are sometimes already
912 pre-defined.
914 28. Added support for (*UTF8) at the start of a pattern.
916 29. Arrange for flags added by the "release type" setting in CMake to be shown
917 in the configuration summary.
920 Version 7.8 05-Sep-08
921 ---------------------
923 1. Replaced UCP searching code with optimized version as implemented for Ad
924 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
925 stage table and inline lookup instead of a function, giving speed ups of 2
926 to 5 times on some simple patterns that I tested. Permission was given to
927 distribute the MultiStage2.py script that generates the tables (it's not in
928 the tarball, but is in the Subversion repository).
930 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
931 scripts.
933 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
934 a group with a zero qualifier. The result of the study could be incorrect,
935 or the function might crash, depending on the pattern.
937 4. Caseless matching was not working for non-ASCII characters in back
938 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
939 It now works when Unicode Property Support is available.
941 5. In pcretest, an escape such as \x{de} in the data was always generating
942 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
943 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
944 truncation.
946 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
948 7. Added two (int) casts to pcregrep when printing the difference of two
949 pointers, in case they are 64-bit values.
951 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
952 test 2 if it fails.
954 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
955 and a #define of that name to empty if it is not externally set. This is to
956 allow users of MSVC to set it if necessary.
958 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
959 the convenience functions in the pcre_get.c source file.
961 11. An option change at the start of a pattern that had top-level alternatives
962 could cause overwriting and/or a crash. This command provoked a crash in
963 some environments:
965 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
967 This potential security problem was recorded as CVE-2008-2371.
969 12. For a pattern where the match had to start at the beginning or immediately
970 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
971 pcre_dfa_exec() could read past the end of the passed subject if there was
972 no match. To help with detecting such bugs (e.g. with valgrind), I modified
973 pcretest so that it places the subject at the end of its malloc-ed buffer.
975 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
976 exec() might read past the end of the data buffer in UTF-8 mode.
978 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
979 the data contained the byte 0x85 as part of a UTF-8 character within its
980 first line. This applied both to normal and DFA matching.
982 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
983 /^[^d]*?$/8 failed to match "abc".
985 16. Added a missing copyright notice to pcrecpp_internal.h.
987 17. Make it more clear in the documentation that values returned from
988 pcre_exec() in ovector are byte offsets, not character counts.
990 18. Tidied a few places to stop certain compilers from issuing warnings.
992 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
993 supplied by Stefan Weber. I made a further small update for 7.8 because
994 there is a change of source arrangements: the pcre_searchfuncs.c module is
995 replaced by pcre_ucd.c.
998 Version 7.7 07-May-08
999 ---------------------
1001 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
1002 a string to a long long, pretend we don't even have a long long." This is
1003 done by checking for the strtoq, strtoll, and _strtoi64 functions.
1005 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1006 pre-7.6 versions, which defined a global no_arg variable instead of putting
1007 it in the RE class. (See also #8 below.)
1009 3. Remove a line of dead code, identified by coverity and reported by Nuno
1010 Lopes.
1012 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
1014 (1) The include/exclude patterns were being applied to the whole pathnames
1015 of files, instead of just to the final components.
1017 (2) If there was more than one level of directory, the subdirectories were
1018 skipped unless they satisfied the include/exclude conditions. This is
1019 inconsistent with GNU grep (and could even be seen as contrary to the
1020 pcregrep specification - which I improved to make it absolutely clear).
1021 The action now is always to scan all levels of directory, and just
1022 apply the include/exclude patterns to regular files.
1024 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1025 --exclude_dir in the tests to avoid scanning .svn directories.
1027 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1028 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1029 doesn't support NULs in patterns.
1031 7. Added some missing "const"s to declarations of static tables in
1032 pcre_compile.c and pcre_dfa_exec.c.
1034 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1035 caused by fix #2 above. (Subsequently also a second patch to fix the
1036 first patch. And a third patch - this was a messy problem.)
1038 9. Applied Craig's patch to remove the use of push_back().
1040 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1041 matching function regexec().
1043 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1044 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1045 references. PCRE supports relative numbers with this syntax (I don't think
1046 Oniguruma does).
1048 12. Previously, a group with a zero repeat such as (...){0} was completely
1049 omitted from the compiled regex. However, this means that if the group
1050 was called as a subroutine from elsewhere in the pattern, things went wrong
1051 (an internal error was given). Such groups are now left in the compiled
1052 pattern, with a new opcode that causes them to be skipped at execution
1053 time.
1055 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1056 to the way PCRE behaves:
1058 (a) A lone ] character is dis-allowed (Perl treats it as data).
1060 (b) A back reference to an unmatched subpattern matches an empty string
1061 (Perl fails the current match path).
1063 (c) A data ] in a character class must be notated as \] because if the
1064 first data character in a class is ], it defines an empty class. (In
1065 Perl it is not possible to have an empty class.) The empty class []
1066 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1067 The negative empty class [^] matches any one character, independently
1068 of the DOTALL setting.
1070 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1071 non-existent subpattern following a character class starting with ']' and
1072 containing () gave an internal compiling error instead of "reference to
1073 non-existent subpattern". Fortunately, when the pattern did exist, the
1074 compiled code was correct. (When scanning forwards to check for the
1075 existencd of the subpattern, it was treating the data ']' as terminating
1076 the class, so got the count wrong. When actually compiling, the reference
1077 was subsequently set up correctly.)
1079 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1080 it was being rejected as not supported by pcre_dfa_exec(), even though
1081 other assertions are supported. I have made pcre_dfa_exec() support
1082 (*FAIL).
1084 16. The implementation of 13c above involved the invention of a new opcode,
1085 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1086 cannot be changed at match time, I realized I could make a small
1087 improvement to matching performance by compiling OP_ALLANY instead of
1088 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1089 on the OP_ANY path.
1091 17. Compiling pcretest on Windows with readline support failed without the
1092 following two fixes: (1) Make the unistd.h include conditional on
1093 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1095 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1096 ncurses library to be included for pcretest when ReadLine support is
1097 requested, but also to allow for it to be overridden. This patch came from
1098 Daniel Bergström.
1100 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1101 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1102 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1103 spotting this.
1106 Version 7.6 28-Jan-08
1107 ---------------------
1109 1. A character class containing a very large number of characters with
1110 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1111 overflow.
1113 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1114 HAVE_LONG_LONG is not defined.
1116 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1117 bring them up to date and include new features. This patch includes:
1119 - Fixed PH's badly added libz and libbz2 support.
1120 - Fixed a problem with static linking.
1121 - Added pcredemo. [But later removed - see 7 below.]
1122 - Fixed dftables problem and added an option.
1123 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1125 - Added readline support for pcretest.
1126 - Added an listing of the option settings after cmake has run.
1128 4. A user submitted a patch to Makefile that makes it easy to create
1129 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1130 Makefile.am that cause it to include this special target, without
1131 affecting anything else. Note that the same mingw target plus all
1132 the other distribution libraries and programs are now supported
1133 when configuring with CMake (see 6 below) instead of with
1134 Configure/Make.
1136 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1137 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1138 exported in the Windows port". It has not yet been confirmed that the patch
1139 solves the problem, but it does no harm.
1141 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1142 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1143 with CMake, and also correct the comment about stack recursion.
1145 7. Remove the automatic building of pcredemo from the ./configure system and
1146 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1147 of a program that users should build themselves after PCRE is installed, so
1148 building it automatically is not really right. What is more, it gave
1149 trouble in some build environments.
1151 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1154 Version 7.5 10-Jan-08
1155 ---------------------
1157 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1158 values in parens when parsing an RE using the C++ wrapper."
1160 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1161 Characters greater than 255 were excluded from the class instead of being
1162 included.
1164 3. The same bug as (2) above applied to negated POSIX classes such as
1165 [:^space:].
1167 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1168 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1169 I have changed it.
1171 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1172 first named subpattern) and a construct such as (?&a) would reference the
1173 first named subpattern whose name started with "a" (in other words, the
1174 length check was missing). Both these problems are fixed. "Subpattern name
1175 expected" is now given for (?&) (a zero-length name), and this patch also
1176 makes it give the same error for \k'' (previously it complained that that
1177 was a reference to a non-existent subpattern).
1179 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1180 this is right because (?- can be followed by option settings as well as by
1181 digits. I have, however, made the messages clearer.
1183 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1184 than the number used in the conditional) now cause a compile-time error.
1185 This is actually not compatible with Perl, which accepts such patterns, but
1186 treats the conditional as always being FALSE (as PCRE used to), but it
1187 seems to me that giving a diagnostic is better.
1189 8. Change "alphameric" to the more common word "alphanumeric" in comments
1190 and messages.
1192 9. Fix two occurrences of "backslash" in comments that should have been
1193 "backspace".
1195 10. Remove two redundant lines of code that can never be obeyed (their function
1196 was moved elsewhere).
1198 11. The program that makes PCRE's Unicode character property table had a bug
1199 which caused it to generate incorrect table entries for sequences of
1200 characters that have the same character type, but are in different scripts.
1201 It amalgamated them into a single range, with the script of the first of
1202 them. In other words, some characters were in the wrong script. There were
1203 thirteen such cases, affecting characters in the following ranges:
1205 U+002b0 - U+002c1
1206 U+0060c - U+0060d
1207 U+0061e - U+00612
1208 U+0064b - U+0065e
1209 U+0074d - U+0076d
1210 U+01800 - U+01805
1211 U+01d00 - U+01d77
1212 U+01d9b - U+01dbf
1213 U+0200b - U+0200f
1214 U+030fc - U+030fe
1215 U+03260 - U+0327f
1216 U+0fb46 - U+0fbb1
1217 U+10450 - U+1049d
1219 12. The -o option (show only the matching part of a line) for pcregrep was not
1220 compatible with GNU grep in that, if there was more than one match in a
1221 line, it showed only the first of them. It now behaves in the same way as
1222 GNU grep.
1224 13. If the -o and -v options were combined for pcregrep, it printed a blank
1225 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1226 does the same. The return code can be used to tell if there were any
1227 non-matching lines.
1229 14. Added --file-offsets and --line-offsets to pcregrep.
1231 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1232 infinitely looping recursion. The bug was that positive lookaheads were not
1233 being skipped when checking for a possible empty match (negative lookaheads
1234 and both kinds of lookbehind were skipped).
1236 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1237 inclusion of <windows.h> to before rather than after the definition of
1238 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1240 17. Specifying a possessive quantifier with a specific limit for a Unicode
1241 character property caused pcre_compile() to compile bad code, which led at
1242 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1243 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1244 caused the error; without that there was no problem.
1246 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1248 19. Added --enable-pcretest-libreadline.
1250 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1251 RE::GlobalReplace(). As a result, the number of replacements returned was
1252 double what it should be. I removed one of the increments, but Craig sent a
1253 later patch that removed the other one (the right fix) and added unit tests
1254 that check the return values (which was not done before).
1256 21. Several CMake things:
1258 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1259 the names libpcre and libpcreposix, not just pcre and pcreposix.
1261 (2) The above change means that pcretest and pcregrep are now correctly
1262 linked with the newly-built libraries, not previously installed ones.
1266 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1267 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1268 UTF-8 newline character). The key issue is that the pattern starts .*;
1269 this means that the match must be either at the beginning, or after a
1270 newline. The bug was in the code for advancing after a failed match and
1271 checking that the new position followed a newline. It was not taking
1272 account of UTF-8 characters correctly.
1274 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1275 character classes. PCRE was not treating the sequence [:...:] as a
1276 character class unless the ... were all letters. Perl, however, seems to
1277 allow any characters between [: and :], though of course it rejects as
1278 unknown any "names" that contain non-letters, because all the known class
1279 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1280 for example, whereas PCRE did not - it did not recognize a POSIX character
1281 class. This seemed a bit dangerous, so the code has been changed to be
1282 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1283 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1284 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1285 Perl does, and where it didn't before.
1287 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1288 Windows environments %n is disabled by default.
1291 Version 7.4 21-Sep-07
1292 ---------------------
1294 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1295 means that a class such as [\s] counted as "explicit reference to CR or
1296 LF". That isn't really right - the whole point of the change was to try to
1297 help when there was an actual mention of one of the two characters. So now
1298 the change happens only if \r or \n (or a literal CR or LF) character is
1299 encountered.
1301 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1302 of both had grown to the point where there were only 3 bits left.
1303 Fortunately, there was spare space in the data structure, and so I have
1304 moved the internal flags into a new 16-bit field to free up more option
1305 bits.
1307 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1308 but did not set the internal JCHANGED flag - either of these is enough to
1309 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1310 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1311 start sets both bits.
1313 4. Added options (at build time, compile time, exec time) to change \R from
1314 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1316 5. doc/pcresyntax.html was missing from the distribution.
1318 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1319 compatibility, even though it is no longer used.
1321 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1322 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1323 windows.h file is present (where different names are used). [This was
1324 reversed later after testing - see 16 below.]
1326 8. Changed all #include <config.h> to #include "config.h". There were also
1327 some further <pcre.h> cases that I changed to "pcre.h".
1329 9. When pcregrep was used with the --colour option, it missed the line ending
1330 sequence off the lines that it output.
1332 10. It was pointed out to me that arrays of string pointers cause lots of
1333 relocations when a shared library is dynamically loaded. A technique of
1334 using a single long string with a table of offsets can drastically reduce
1335 these. I have refactored PCRE in four places to do this. The result is
1336 dramatic:
1338 Originally: 290
1339 After changing UCP table: 187
1340 After changing error message table: 43
1341 After changing table of "verbs" 36
1342 After changing table of Posix names 22
1344 Thanks to the folks working on Gregex for glib for this insight.
1346 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1347 unicode-properties was also set.
1349 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1351 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1352 checked only for CRLF.
1354 14. Added casts to pcretest.c to avoid compiler warnings.
1356 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1358 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1359 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1360 entirely. This removes changes made in 7 above.
1362 17. The CMake files have been updated, and there is now more information about
1363 building with CMake in the NON-UNIX-USE document.
1366 Version 7.3 28-Aug-07
1367 ---------------------
1369 1. In the rejigging of the build system that eventually resulted in 7.1, the
1370 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1371 brackets there is not right, since it causes compilers to look for an
1372 installed pcre.h, not the version that is in the source that is being
1373 compiled (which of course may be different). I have changed it back to:
1375 #include "pcre.h"
1377 I have a vague recollection that the change was concerned with compiling in
1378 different directories, but in the new build system, that is taken care of
1379 by the VPATH setting the Makefile.
1381 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1382 when the subject happened to end in the byte 0x85 (e.g. if the last
1383 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1384 characters but of course it shouldn't be taken as a newline when it is part
1385 of another character. The bug was that, for an unlimited repeat of . in
1386 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1387 characters when looking for a newline.
1389 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1391 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1392 in debug output.
1394 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1395 long printing in the pcrecpp unittest when running under MinGW.
1397 6. ESC_K was left out of the EBCDIC table.
1399 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1400 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1401 limit also applies to "virtual nesting" when a pattern is recursive, and in
1402 this case 1000 isn't so big. I have been able to remove this limit at the
1403 expense of backing off one optimization in certain circumstances. Normally,
1404 when pcre_exec() would call its internal match() function recursively and
1405 immediately return the result unconditionally, it uses a "tail recursion"
1406 feature to save stack. However, when a subpattern that can match an empty
1407 string has an unlimited repetition quantifier, it no longer makes this
1408 optimization. That gives it a stack frame in which to save the data for
1409 checking that an empty string has been matched. Previously this was taken
1410 from the 1000-entry workspace that had been reserved. So now there is no
1411 explicit limit, but more stack is used.
1413 8. Applied Daniel's patches to solve problems with the import/export magic
1414 syntax that is required for Windows, and which was going wrong for the
1415 pcreposix and pcrecpp parts of the library. These were overlooked when this
1416 problem was solved for the main library.
1418 9. There were some crude static tests to avoid integer overflow when computing
1419 the size of patterns that contain repeated groups with explicit upper
1420 limits. As the maximum quantifier is 65535, the maximum group length was
1421 set at 30,000 so that the product of these two numbers did not overflow a
1422 32-bit integer. However, it turns out that people want to use groups that
1423 are longer than 30,000 bytes (though not repeat them that many times).
1424 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1425 made it possible to implement the integer overflow checks in a much more
1426 dynamic way, which I have now done. The artificial limitation on group
1427 length has been removed - we now have only the limit on the total length of
1428 the compiled pattern, which depends on the LINK_SIZE setting.
1430 10. Fixed a bug in the documentation for get/copy named substring when
1431 duplicate names are permitted. If none of the named substrings are set, the
1432 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1433 empty string.
1435 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1436 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1437 because the ] is interpreted as the first data character and the
1438 terminating ] is not found. PCRE has been made compatible with Perl in this
1439 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1440 cause memory overwriting.
1442 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1443 string has been matched (to stop an infinite loop). It was not recognizing
1444 a conditional subpattern that could match an empty string if that
1445 subpattern was within another subpattern. For example, it looped when
1446 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1447 condition was not nested. This bug has been fixed.
1449 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1450 past the start of the subject in the presence of bytes with the top bit
1451 set, for example "\x8aBCD".
1453 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1454 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1456 14. Optimized (?!) to (*FAIL).
1458 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1459 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1460 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1461 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1462 does: it's just the validity check that is more restrictive.
1464 16. Inserted checks for integer overflows during escape sequence (backslash)
1465 processing, and also fixed erroneous offset values for syntax errors during
1466 backslash processing.
1468 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1469 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1471 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1472 caused an overrun.
1474 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1475 something other than just ASCII characters) inside a group that had an
1476 unlimited repeat caused a loop at compile time (while checking to see
1477 whether the group could match an empty string).
1479 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1480 [\P{Any}] did so. (Error in the code for printing property names.)
1482 21. An orphan \E inside a character class could cause a crash.
1484 22. A repeated capturing bracket such as (A)? could cause a wild memory
1485 reference during compilation.
1487 23. There are several functions in pcre_compile() that scan along a compiled
1488 expression for various reasons (e.g. to see if it's fixed length for look
1489 behind). There were bugs in these functions when a repeated \p or \P was
1490 present in the pattern. These operators have additional parameters compared
1491 with \d, etc, and these were not being taken into account when moving along
1492 the compiled data. Specifically:
1494 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1495 length.
1497 (b) An item such as \pL+ within a repeated group could cause crashes or
1498 loops.
1500 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1501 "reference to non-existent subpattern" error.
1503 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1505 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1506 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1508 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1510 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1511 character were causing crashes (broken optimization).
1513 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1514 \p or \P) caused a compile-time loop.
1516 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1517 break. For example, the unstudied pattern [\r\n]A does not match the string
1518 "\r\nA" because change 7.0/46 below moves the current point on by two
1519 characters after failing to match at the start. However, the pattern \nA
1520 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1521 the same is true. There doesn't seem any very clean way out of this, but
1522 what I have chosen to do makes the common cases work: PCRE now takes note
1523 of whether there can be an explicit match for \r or \n anywhere in the
1524 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1525 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1526 pattern has explicit CR or LF references.
1528 29. Added (*CR) etc for changing newline setting at start of pattern.
1531 Version 7.2 19-Jun-07
1532 ---------------------
1534 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1535 which is apparently normally available under Windows.
1537 2. Re-jig the pcregrep tests with different newline settings in an attempt
1538 to make them independent of the local environment's newline setting.
1540 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1542 4. Some of the "internals" tests were previously cut out when the link size
1543 was not 2, because the output contained actual offsets. The recent new
1544 "Z" feature of pcretest means that these can be cut out, making the tests
1545 usable with all link sizes.
1547 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1548 stack recursion. This gives a massive performance boost under BSD, but just
1549 a small improvement under Linux. However, it saves one field in the frame
1550 in all cases.
1552 6. Added more features from the forthcoming Perl 5.10:
1554 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1555 recursion call. It refers to the nth most recently opened parentheses.
1557 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1558 to be opened parentheses.
1560 (c) Conditions that refer to capturing parentheses can be specified
1561 relatively, for example, (?(-2)... or (?(+3)...
1563 (d) \K resets the start of the current match so that everything before
1564 is not part of it.
1566 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1568 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1569 reference syntax.
1571 (g) (?| introduces a group in which the numbering of parentheses in each
1572 alternative starts with the same number.
1574 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1576 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1579 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1580 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1581 for detecting groups that can match an empty string.
1583 9. A pattern with a very large number of alternatives (more than several
1584 hundred) was running out of internal workspace during the pre-compile
1585 phase, where pcre_compile() figures out how much memory will be needed. A
1586 bit of new cunning has reduced the workspace needed for groups with
1587 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1588 workspace instead of running out of the 4096 that are available.
1590 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1592 11. Applied patch from Google to remove an optimization that didn't quite work.
1593 The report of the bug said:
1595 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1596 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1597 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1599 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1600 it matched the wrong number of bytes.
1603 Version 7.1 24-Apr-07
1604 ---------------------
1606 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1607 that is more "standard", making use of automake and other Autotools. There
1608 is some re-arrangement of the files and adjustment of comments consequent
1609 on this.
1611 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1612 for recursive directory scanning broke on some systems because the files
1613 are not scanned in any specific order and on different systems the order
1614 was different. A call to "sort" has been inserted into RunGrepTest for the
1615 approprate test as a short-term fix. In the longer term there may be an
1616 alternative.
1618 3. I had an email from Eric Raymond about problems translating some of PCRE's
1619 man pages to HTML (despite the fact that I distribute HTML pages, some
1620 people do their own conversions for various reasons). The problems
1621 concerned the use of low-level troff macros .br and .in. I have therefore
1622 removed all such uses from the man pages (some were redundant, some could
1623 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1624 HTML has been updated to handle .nf/.fi and to complain if it encounters
1625 .br or .in.
1627 4. Updated comments in configure.ac that get placed in config.h.in and also
1628 arranged for config.h to be included in the distribution, with the name
1629 config.h.generic, for the benefit of those who have to compile without
1630 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1632 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1633 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1634 makevp.bat for the current PCRE, using the additional files
1635 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1637 6. A Windows user reported a minor discrepancy with test 2, which turned out
1638 to be caused by a trailing space on an input line that had got lost in his
1639 copy. The trailing space was an accident, so I've just removed it.
1641 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1642 that is needed.
1644 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1645 as "const" (a) because they are and (b) because it helps the PHP
1646 maintainers who have recently made a script to detect big data structures
1647 in the php code that should be moved to the .rodata section. I remembered
1648 to update Builducptable as well, so it won't revert if ucptable.h is ever
1649 re-created.
1651 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1652 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1653 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1654 support is not required. This saves 1.5-2K of code, which is important in
1655 some applications.
1657 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1658 so as not to refer to the tables, even though these functions will never be
1659 called when UTF-8 support is disabled. Otherwise there are problems with a
1660 shared library.
1662 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1664 (a) It was defining its arguments as char * instead of void *.
1666 (b) It was assuming that all moves were upwards in memory; this was true
1667 a long time ago when I wrote it, but is no longer the case.
1669 The emulated memove() is provided for those environments that have neither
1670 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1671 is clearly not the case, as these two bugs were recently reported.
1673 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1674 and Detrail to create the HTML documentation, the .txt form of the man
1675 pages, and it removes trailing spaces from listed files. It also creates
1676 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1677 case, it wraps all the #defines with #ifndefs. This script should be run
1678 before "make dist".
1680 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1681 with Unicode property support.
1683 (a) For a maximizing quantifier, if the two different cases of the
1684 character were of different lengths in their UTF-8 codings (there are
1685 some cases like this - I found 11), and the matching function had to
1686 back up over a mixture of the two cases, it incorrectly assumed they
1687 were both the same length.
1689 (b) When PCRE was configured to use the heap rather than the stack for
1690 recursion during matching, it was not correctly preserving the data for
1691 the other case of a UTF-8 character when checking ahead for a match
1692 while processing a minimizing repeat. If the check also involved
1693 matching a wide character, but failed, corruption could cause an
1694 erroneous result when trying to check for a repeat of the original
1695 character.
1697 13. Some tidying changes to the testing mechanism:
1699 (a) The RunTest script now detects the internal link size and whether there
1700 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1701 values substituted by "configure". (The RunGrepTest script already did
1702 this for UTF-8.) The configure.ac script no longer substitutes the
1703 relevant variables.
1705 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1706 with length and offset values. This means that the output is different
1707 for different internal link sizes. Test 2 is skipped for link sizes
1708 other than 2 because of this, bypassing the problem. Unfortunately,
1709 there was also a test in test 3 (the locale tests) that used /B and
1710 failed for link sizes other than 2. Rather than cut the whole test out,
1711 I have added a new /Z option to pcretest that replaces the length and
1712 offset values with spaces. This is now used to make test 3 independent
1713 of link size. (Test 2 will be tidied up later.)
1715 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1716 segmentation fault instead of returning the appropriate error message.
1718 15. In multiline mode when the newline sequence was set to "any", the pattern
1719 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1720 This doesn't seem right; it now treats the CRLF combination as the line
1721 ending, and so does not match in that case. It's only a pattern such as ^$
1722 that would hit this one: something like ^ABC$ would have failed after \r
1723 and then tried again after \r\n.
1725 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1726 in an attempt to make files that differ only in their line terminators
1727 compare equal. This works on Linux.
1729 17. Under certain error circumstances pcregrep might try to free random memory
1730 as it exited. This is now fixed, thanks to valgrind.
1732 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1733 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1734 was because its rules for how to advance for /g after matching an empty
1735 string at the end of a line did not allow for this case. They now check for
1736 it specially.
1738 20. pcretest is supposed to handle patterns and data of any length, by
1739 extending its buffers when necessary. It was getting this wrong when the
1740 buffer for a data line had to be extended.
1742 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1743 CRLF as a newline sequence.
1745 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1746 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1747 I have nevertheless tidied it up.
1749 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1751 24. Added a man page for pcre-config.
1754 Version 7.0 19-Dec-06
1755 ---------------------
1757 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1758 moving to gcc 4.1.1.
1760 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1761 sys/time.h, which is documented as needed for this function. It doesn't
1762 seem to matter on Linux, but it showed up on some releases of OS X.
1764 3. It seems that there are systems where bytes whose values are greater than
1765 127 match isprint() in the "C" locale. The "C" locale should be the
1766 default when a C program starts up. In most systems, only ASCII printing
1767 characters match isprint(). This difference caused the output from pcretest
1768 to vary, making some of the tests fail. I have changed pcretest so that:
1770 (a) When it is outputting text in the compiled version of a pattern, bytes
1771 other than 32-126 are always shown as hex escapes.
1773 (b) When it is outputting text that is a matched part of a subject string,
1774 it does the same, unless a different locale has been set for the match
1775 (using the /L modifier). In this case, it uses isprint() to decide.
1777 4. Fixed a major bug that caused incorrect computation of the amount of memory
1778 required for a compiled pattern when options that changed within the
1779 pattern affected the logic of the preliminary scan that determines the
1780 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1781 that the computed length was too small. The symptoms of this bug were
1782 either the PCRE error "internal error: code overflow" from pcre_compile(),
1783 or a glibc crash with a message such as "pcretest: free(): invalid next
1784 size (fast)". Examples of patterns that provoked this bug (shown in
1785 pcretest format) are:
1787 /(?-x: )/x
1788 /(?x)(?-x: \s*#\s*)/
1789 /((?i)[\x{c0}])/8
1790 /(?i:[\x{c0}])/8
1792 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1793 is now done differently.
1795 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1796 wrapper classes; (b) implement a new function in the C++ scanner that is
1797 more efficient than the old way of doing things because it avoids levels of
1798 recursion in the regex matching; (c) add a paragraph to the documentation
1799 for the FullMatch() function.
1801 6. The escape sequence \n was being treated as whatever was defined as
1802 "newline". Not only was this contrary to the documentation, which states
1803 that \n is character 10 (hex 0A), but it also went horribly wrong when
1804 "newline" was defined as CRLF. This has been fixed.
1806 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1807 was being set to -1 for the "end of line" case (supposedly a value that no
1808 character can have). Though this value is never used (the check for end of
1809 line is "zero bytes in current character"), it caused compiler complaints.
1810 I've changed it to 0xffffffff.
1812 8. In pcre_version.c, the version string was being built by a sequence of
1813 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1814 string (as it is for production releases) called a macro with an empty
1815 argument. The C standard says the result of this is undefined. The gcc
1816 compiler treats it as an empty string (which was what was wanted) but it is
1817 reported that Visual C gives an error. The source has been hacked around to
1818 avoid this problem.
1820 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1821 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1822 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1823 of them did).
1825 10. Originally, pcretest opened its input and output without "b"; then I was
1826 told that "b" was needed in some environments, so it was added for release
1827 5.0 to both the input and output. (It makes no difference on Unix-like
1828 systems.) Later I was told that it is wrong for the input on Windows. I've
1829 now abstracted the modes into two macros, to make it easier to fiddle with
1830 them, and removed "b" from the input mode under Windows.
1832 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1834 12. Added -help and --help to pcretest as an official way of being reminded
1835 of the options.
1837 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1838 and pcrecpp.cc because they annoy compilers at high warning levels.
1840 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1842 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1843 scripts are not C programs :-) and which was not noticed because it works
1844 on Linux.
1846 16. pcretest is supposed to handle any length of pattern and data line (as one
1847 line or as a continued sequence of lines) by extending its input buffer if
1848 necessary. This feature was broken for very long pattern lines, leading to
1849 a string of junk being passed to pcre_compile() if the pattern was longer
1850 than about 50K.
1852 17. I have done a major re-factoring of the way pcre_compile() computes the
1853 amount of memory needed for a compiled pattern. Previously, there was code
1854 that made a preliminary scan of the pattern in order to do this. That was
1855 OK when PCRE was new, but as the facilities have expanded, it has become
1856 harder and harder to keep it in step with the real compile phase, and there
1857 have been a number of bugs (see for example, 4 above). I have now found a
1858 cunning way of running the real compile function in a "fake" mode that
1859 enables it to compute how much memory it would need, while actually only
1860 ever using a few hundred bytes of working memory and without too many
1861 tests of the mode. This should make future maintenance and development
1862 easier. A side effect of this work is that the limit of 200 on the nesting
1863 depth of parentheses has been removed (though this was never a serious
1864 limitation, I suspect). However, there is a downside: pcre_compile() now
1865 runs more slowly than before (30% or more, depending on the pattern). I
1866 hope this isn't a big issue. There is no effect on runtime performance.
1868 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1869 newline (only possible for the last line of a file) and it was a
1870 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1872 19. Added additional timing features to pcretest. (1) The -tm option now times
1873 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1874 separate command line item, by a number that specifies the number of
1875 repeats to use when timing. The default is 50000; this gives better
1876 precision, but takes uncomfortably long for very large patterns.
1878 20. Extended pcre_study() to be more clever in cases where a branch of a
1879 subpattern has no definite first character. For example, (a*|b*)[cd] would
1880 previously give no result from pcre_study(). Now it recognizes that the
1881 first character must be a, b, c, or d.
1883 21. There was an incorrect error "recursive call could loop indefinitely" if
1884 a subpattern (or the entire pattern) that was being tested for matching an
1885 empty string contained only one non-empty item after a nested subpattern.
1886 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1887 incorrectly, because the \d was being skipped in the check.
1889 22. The pcretest program now has a new pattern option /B and a command line
1890 option -b, which is equivalent to adding /B to every pattern. This causes
1891 it to show the compiled bytecode, without the additional information that
1892 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1893 is the same as /B/I).
1895 23. A new optimization is now able automatically to treat some sequences such
1896 as a*b as a*+b. More specifically, if something simple (such as a character
1897 or a simple class like \d) has an unlimited quantifier, and is followed by
1898 something that cannot possibly match the quantified thing, the quantifier
1899 is automatically "possessified".
1901 24. A recursive reference to a subpattern whose number was greater than 39
1902 went wrong under certain circumstances in UTF-8 mode. This bug could also
1903 have affected the operation of pcre_study().
1905 25. Realized that a little bit of performance could be had by replacing
1906 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1908 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1910 27. Possessive quantifiers such as a++ were previously implemented by turning
1911 them into atomic groups such as ($>a+). Now they have their own opcodes,
1912 which improves performance. This includes the automatically created ones
1913 from 23 above.
1915 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1916 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1917 the first matched character to be a colon. This applied both to named and
1918 numbered groups.
1920 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1922 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1923 building PCRE on Windows easier, so I have included it in the distribution.
1925 31. There is now a check in pcretest against a ridiculously large number being
1926 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1927 loop, the loop is abandoned.
1929 32. Forward references to subpatterns in conditions such as (?(2)...) where
1930 subpattern 2 is defined later cause pcre_compile() to search forwards in
1931 the pattern for the relevant set of parentheses. This search went wrong
1932 when there were unescaped parentheses in a character class, parentheses
1933 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1935 33. "Subroutine" calls and backreferences were previously restricted to
1936 referencing subpatterns earlier in the regex. This restriction has now
1937 been removed.
1939 34. Added a number of extra features that are going to be in Perl 5.10. On the
1940 whole, these are just syntactic alternatives for features that PCRE had
1941 previously implemented using the Python syntax or my own invention. The
1942 other formats are all retained for compatibility.
1944 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1945 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1946 also .NET compatible.
1948 (b) A recursion or subroutine call to a named group can now be defined as
1949 (?&name) as well as (?P>name).
1951 (c) A backreference to a named group can now be defined as \k<name> or
1952 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1953 5.10, are also .NET compatible.
1955 (d) A conditional reference to a named group can now use the syntax
1956 (?(<name>) or (?('name') as well as (?(name).
1958 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1959 groups (named and numbered) that are never evaluated inline, but can be
1960 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1961 is always false. There may be only one alternative in such a group.
1963 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1964 as the simple (?(R). The condition is true only if the most recent
1965 recursion is that of the given number or name. It does not search out
1966 through the entire recursion stack.
1968 (g) The escape \gN or \g{N} has been added, where N is a positive or
1969 negative number, specifying an absolute or relative reference.
1971 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1972 some "unreachable code" warnings.
1974 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1975 things, this adds five new scripts.
1977 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1978 There were also incompatibilities regarding the handling of \Q..\E inside
1979 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1980 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1982 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1983 matches an empty string, and forcibly breaks the loop. There were bugs in
1984 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1985 against aaaa the result was just "a" rather than "aaaa", for example. Two
1986 separate and independent bugs (that affected different cases) have been
1987 fixed.
1989 39. Refactored the code to abolish the use of different opcodes for small
1990 capturing bracket numbers. This is a tidy that I avoided doing when I
1991 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1992 The new approach is not only tidier, it makes it possible to reduce the
1993 memory needed to fix the previous bug (38).
1995 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1996 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1997 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1998 mode.
2000 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2001 report.
2003 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2004 copy construction and assignment in the C++ wrapper.
2006 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2007 couple of bugs that could have given wrong results in the "--newline=crlf"
2008 case.
2010 44. Added a number of casts and did some reorganization of signed/unsigned int
2011 variables following suggestions from Dair Grant. Also renamed the variable
2012 "this" as "item" because it is a C++ keyword.
2014 45. Arranged for dftables to add
2016 #include "pcre_internal.h"
2018 to pcre_chartables.c because without it, gcc 4.x may remove the array
2019 definition from the final binary if PCRE is built into a static library and
2020 dead code stripping is activated.
2022 46. For an unanchored pattern, if a match attempt fails at the start of a
2023 newline sequence, and the newline setting is CRLF or ANY, and the next two
2024 characters are CRLF, advance by two characters instead of one.
2027 Version 6.7 04-Jul-06
2028 ---------------------
2030 1. In order to handle tests when input lines are enormously long, pcretest has
2031 been re-factored so that it automatically extends its buffers when
2032 necessary. The code is crude, but this _is_ just a test program. The
2033 default size has been increased from 32K to 50K.
2035 2. The code in pcre_study() was using the value of the re argument before
2036 testing it for NULL. (Of course, in any sensible call of the function, it
2037 won't be NULL.)
2039 3. The memmove() emulation function in pcre_internal.h, which is used on
2040 systems that lack both memmove() and bcopy() - that is, hardly ever -
2041 was missing a "static" storage class specifier.
2043 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2044 containing an extended class (one that cannot be represented by a bitmap
2045 because it contains high-valued characters or Unicode property items, e.g.
2046 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2047 pattern, but PCRE should not loop if you do not (it no longer does).
2048 [Detail: two cases were found: (a) a repeated subpattern containing an
2049 extended class; (b) a recursive reference to a subpattern that followed a
2050 previous extended class. It wasn't skipping over the extended class
2051 correctly when UTF-8 mode was not set.]
2053 5. A negated single-character class was not being recognized as fixed-length
2054 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2055 compile error "lookbehind assertion is not fixed length".
2057 6. The RunPerlTest auxiliary script was showing an unexpected difference
2058 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2059 write a Perl script that can interpret lines of an input file either as
2060 byte characters or as UTF-8, which is what "perltest" was being required to
2061 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2062 can't do is switch easily at run time between having the "use utf8;" pragma
2063 or not. In the end, I fudged it by using the RunPerlTest script to insert
2064 "use utf8;" explicitly for the UTF-8 tests.
2066 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2067 the end of the subject string, contrary to the documentation and to what
2068 Perl does. This was true of both matching functions. Now it matches only at
2069 the start of the subject and immediately after *internal* newlines.
2071 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2072 a pointer to an int instead of a pointer to an unsigned long int. This
2073 caused problems on 64-bit systems.
2075 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2076 instance of the 'standard' template library not being so standard".
2078 10. There was no check on the number of named subpatterns nor the maximum
2079 length of a subpattern name. The product of these values is used to compute
2080 the size of the memory block for a compiled pattern. By supplying a very
2081 long subpattern name and a large number of named subpatterns, the size
2082 computation could be caused to overflow. This is now prevented by limiting
2083 the length of names to 32 characters, and the number of named subpatterns
2084 to 10,000.
2086 11. Subpatterns that are repeated with specific counts have to be replicated in
2087 the compiled pattern. The size of memory for this was computed from the
2088 length of the subpattern and the repeat count. The latter is limited to
2089 65535, but there was no limit on the former, meaning that integer overflow
2090 could in principle occur. The compiled length of a repeated subpattern is
2091 now limited to 30,000 bytes in order to prevent this.
2093 12. Added the optional facility to have named substrings with the same name.
2095 13. Added the ability to use a named substring as a condition, using the
2096 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2097 are numbers (not recommended). Forward references are permitted.
2099 14. Added forward references in named backreferences (if you see what I mean).
2101 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2102 pattern could run off the end of the subject. For example, the pattern
2103 "(?s)(.{1,5})"8 did this with the subject "ab".
2105 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2106 PCRE_CASELESS was set when matching characters that were quantified with ?
2107 or *.
2109 17. A character class other than a single negated character that had a minimum
2110 but no maximum quantifier - for example [ab]{6,} - was not handled
2111 correctly by pce_dfa_exec(). It would match only one character.
2113 18. A valid (though odd) pattern that looked like a POSIX character
2114 class but used an invalid character after [ (for example [[,abc,]]) caused
2115 pcre_compile() to give the error "Failed: internal error: code overflow" or
2116 in some cases to crash with a glibc free() error. This could even happen if
2117 the pattern terminated after [[ but there just happened to be a sequence of
2118 letters, a binary zero, and a closing ] in the memory that followed.
2120 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2121 over the years. Originally (before any Unicode support), just the bottom 8
2122 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2123 output from "man perlunicode" includes this:
2125 The regular expression compiler produces polymorphic opcodes. That
2126 is, the pattern adapts to the data and automatically switches to
2127 the Unicode character scheme when presented with Unicode data--or
2128 instead uses a traditional byte scheme when presented with byte
2129 data.
2131 Sadly, a wide octal escape does not cause a switch, and in a string with
2132 no other multibyte characters, these octal escapes are treated as before.
2133 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2134 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2135 Unicode string.
2137 I have not perpetrated such confusion in PCRE. Up till now, it took just
2138 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2139 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2140 translate to the appropriate multibyte character.
2142 29. Applied some refactoring to reduce the number of warnings from Microsoft
2143 and Borland compilers. This has included removing the fudge introduced
2144 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2145 a warning about an unused variable.
2147 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2148 characters since release 4.0, because Perl (from release 5.004) does not.
2149 [Or at least, is documented not to: some releases seem to be in conflict
2150 with the documentation.] However, when a pattern was studied with
2151 pcre_study() and all its branches started with \s, PCRE still included VT
2152 as a possible starting character. Of course, this did no harm; it just
2153 caused an unnecessary match attempt.
2155 22. Removed a now-redundant internal flag bit that recorded the fact that case
2156 dependency changed within the pattern. This was once needed for "required
2157 byte" processing, but is no longer used. This recovers a now-scarce options
2158 bit. Also moved the least significant internal flag bit to the most-
2159 significant bit of the word, which was not previously used (hangover from
2160 the days when it was an int rather than a uint) to free up another bit for
2161 the future.
2163 23. Added support for CRLF line endings as well as CR and LF. As well as the
2164 default being selectable at build time, it can now be changed at runtime
2165 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2166 specify that it is scanning data with non-default line endings.
2168 24. Changed the definition of CXXLINK to make it agree with the definition of
2169 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2171 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2172 recursions. This makes a big different to stack usage for some patterns.
2174 26. If a subpattern containing a named recursion or subroutine reference such
2175 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2176 the space required for the compiled pattern went wrong and gave too small a
2177 value. Depending on the environment, this could lead to "Failed: internal
2178 error: code overflow at offset 49" or "glibc detected double free or
2179 corruption" errors.
2181 27. Applied patches from Google (a) to support the new newline modes and (b) to
2182 advance over multibyte UTF-8 characters in GlobalReplace.
2184 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2185 difference for some implementation of PCRE in some Windows version.
2187 29. Added some extra testing facilities to pcretest:
2189 \q<number> in a data line sets the "match limit" value
2190 \Q<number> in a data line sets the "match recursion limt" value
2191 -S <number> sets the stack size, where <number> is in megabytes
2193 The -S option isn't available for Windows.
2196 Version 6.6 06-Feb-06
2197 ---------------------
2199 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2200 in pcreposix.h. I have copied the definition from pcre.h.
2202 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2203 because pcre.h is no longer a built file.
2205 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2206 not normally included in the compiled code.
2209 Version 6.5 01-Feb-06
2210 ---------------------
2212 1. When using the partial match feature with pcre_dfa_exec(), it was not
2213 anchoring the second and subsequent partial matches at the new starting
2214 point. This could lead to incorrect results. For example, with the pattern
2215 /1234/, partially matching against "123" and then "a4" gave a match.
2217 2. Changes to pcregrep:
2219 (a) All non-match returns from pcre_exec() were being treated as failures
2220 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2221 error message is output. Some extra information is given for the
2223 probably the only errors that are likely to be caused by users (by
2224 specifying a regex that has nested indefinite repeats, for instance).
2225 If there are more than 20 of these errors, pcregrep is abandoned.
2227 (b) A binary zero was treated as data while matching, but terminated the
2228 output line if it was written out. This has been fixed: binary zeroes
2229 are now no different to any other data bytes.
2231 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2232 used to set a locale for matching. The --locale=xxxx long option has
2233 been added (no short equivalent) to specify a locale explicitly on the
2234 pcregrep command, overriding the environment variables.
2236 (d) When -B was used with -n, some line numbers in the output were one less
2237 than they should have been.
2239 (e) Added the -o (--only-matching) option.
2241 (f) If -A or -C was used with -c (count only), some lines of context were
2242 accidentally printed for the final match.
2244 (g) Added the -H (--with-filename) option.
2246 (h) The combination of options -rh failed to suppress file names for files
2247 that were found from directory arguments.
2249 (i) Added the -D (--devices) and -d (--directories) options.
2251 (j) Added the -F (--fixed-strings) option.
2253 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2255 (l) Added the --colo(u)r option.
2257 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2258 is not present by default.
2260 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2261 items such as (?R) or (?1), when the recursion could match a number of
2262 alternatives. If it matched one of the alternatives, but subsequently,
2263 outside the recursion, there was a failure, the code tried to back up into
2264 the recursion. However, because of the way PCRE is implemented, this is not
2265 possible, and the result was an incorrect result from the match.
2267 In order to prevent this happening, the specification of recursion has
2268 been changed so that all such subpatterns are automatically treated as
2269 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2271 4. I had overlooked the fact that, in some locales, there are characters for
2272 which isalpha() is true but neither isupper() nor islower() are true. In
2273 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2274 and ordfeminine) are like this. This affected the treatment of \w and \W
2275 when they appeared in character classes, but not when they appeared outside
2276 a character class. The bit map for "word" characters is now created
2277 separately from the results of isalnum() instead of just taking it from the
2278 upper, lower, and digit maps. (Plus the underscore character, of course.)
2280 5. The above bug also affected the handling of POSIX character classes such as
2281 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2282 permanent tables. Instead, the bit maps for such a class were previously
2283 created as the appropriate unions of the upper, lower, and digit bitmaps.
2284 Now they are created by subtraction from the [[:word:]] class, which has
2285 its own bitmap.
2287 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2288 It is created by subtracting the vertical space characters (\x09, \x0a,
2289 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2290 subtraction was done in the overall bitmap for a character class, meaning
2291 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2292 be recognized. This bug has been fixed.
2294 7. Patches from the folks at Google:
2296 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2297 real life, but is still worth protecting against".
2299 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2300 regular expressions".
2302 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2303 have it.
2305 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2306 "configure" and the latter not, in order to fix a problem somebody had
2307 with compiling the Arg class on HP-UX.
2309 (e) Improve the error-handling of the C++ wrapper a little bit.
2311 (f) New tests for checking recursion limiting.
2313 8. The pcre_memmove() function, which is used only if the environment does not
2314 have a standard memmove() function (and is therefore rarely compiled),
2315 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2316 returning a result (though PCRE never actually uses the result).
2318 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2319 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2320 returned instead of calling malloc() with an overflowing number that would
2321 most likely cause subsequent chaos.
2323 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2325 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2326 with this option is matched, the nmatch and pmatch options of regexec() are
2327 ignored.
2329 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2330 provided in case anyone wants to the the POSIX interface with UTF-8
2331 strings.
2333 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2334 C++ linking (needed for some HP-UX environments).
2336 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2337 (unused parameter) and in the pcre_printint() function (omitted "default"
2338 switch label when the default is to do nothing).
2340 15. Added some code to make it possible, when PCRE is compiled as a C++
2341 library, to replace subject pointers for pcre_exec() with a smart pointer
2342 class, thus making it possible to process discontinuous strings.
2344 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2345 much the same function. They were added by different people who were trying
2346 to make PCRE easy to compile on non-Unix systems. It has been suggested
2347 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2348 for compiling on Windows systems. I have therefore replaced it with
2349 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2350 defaults to "extern" for C or "extern C" for C++, which works fine on
2351 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2352 SCOPE with something explicit in config.h. In addition:
2354 (a) pcreposix.h still had just "extern" instead of either of these macros;
2355 I have replaced it with PCRE_DATA_SCOPE.
2357 (b) Functions such as _pcre_xclass(), which are internal to the library,
2358 but external in the C sense, all had PCRE_EXPORT in their definitions.
2359 This is apparently wrong for the Windows case, so I have removed it.
2360 (It makes no difference on Unix-like systems.)
2362 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2363 of recursive calls to match(). This is different to MATCH_LIMIT because
2364 that limits the total number of calls to match(), not all of which increase
2365 the depth of recursion. Limiting the recursion depth limits the amount of
2366 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2367 when PCRE is compiled, and changed at run time. A patch from Google adds
2368 this functionality to the C++ interface.
2370 18. Changes to the handling of Unicode character properties:
2372 (a) Updated the table to Unicode 4.1.0.
2374 (b) Recognize characters that are not in the table as "Cn" (undefined).
2376 (c) I revised the way the table is implemented to a much improved format
2377 which includes recognition of ranges. It now supports the ranges that
2378 are defined in UnicodeData.txt, and it also amalgamates other
2379 characters into ranges. This has reduced the number of entries in the
2380 table from around 16,000 to around 3,000, thus reducing its size
2381 considerably. I realized I did not need to use a tree structure after
2382 all - a binary chop search is just as efficient. Having reduced the
2383 number of entries, I extended their size from 6 bytes to 8 bytes to
2384 allow for more data.
2386 (d) Added support for Unicode script names via properties such as \p{Han}.
2388 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2389 matching that character.
2391 20. When matching a repeated Unicode property with a minimum greater than zero,
2392 (for example \pL{2,}), PCRE could look past the end of the subject if it
2393 reached it while seeking the minimum number of characters. This could
2394 happen only if some of the characters were more than one byte long, because
2395 there is a check for at least the minimum number of bytes.
2397 21. Refactored the implementation of \p and \P so as to be more general, to
2398 allow for more different types of property in future. This has changed the
2399 compiled form incompatibly. Anybody with saved compiled patterns that use
2400 \p or \P will have to recompile them.
2402 22. Added "Any" and "L&" to the supported property types.
2404 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2405 but give a compile time error if the value is greater than 0xff.
2407 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2408 accidentally not being installed or uninstalled.
2410 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2411 made were to insert the current release number. This seemed silly, because
2412 it made things harder for people building PCRE on systems that don't run
2413 "configure". I have turned pcre.h into a distributed file, no longer built
2414 by "configure", with the version identification directly included. There is
2415 no longer a pcre.h.in file.
2417 However, this change necessitated a change to the pcre-config script as
2418 well. It is built from pcre-config.in, and one of the substitutions was the
2419 release number. I have updated configure.ac so that ./configure now finds
2420 the release number by grepping pcre.h.
2422 26. Added the ability to run the tests under valgrind.
2425 Version 6.4 05-Sep-05
2426 ---------------------
2428 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2429 "--" to be printed when multiple files were scanned, even when none of the
2430 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2431 consider it to be a bug, and have restored the previous behaviour.
2433 2. A couple of code tidies to get rid of compiler warnings.
2435 3. The pcretest program used to cheat by referring to symbols in the library
2436 whose names begin with _pcre_. These are internal symbols that are not
2437 really supposed to be visible externally, and in some environments it is
2438 possible to suppress them. The cheating is now confined to including
2439 certain files from the library's source, which is a bit cleaner.
2441 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2442 file's purpose clearer.
2444 5. Reorganized pcre_ucp_findchar().
2447 Version 6.3 15-Aug-05
2448 ---------------------
2450 1. The file libpcre.pc.in did not have general read permission in the tarball.
2452 2. There were some problems when building without C++ support:
2454 (a) If C++ support was not built, "make install" and "make test" still
2455 tried to test it.
2457 (b) There were problems when the value of CXX was explicitly set. Some
2458 changes have been made to try to fix these, and ...
2460 (c) --disable-cpp can now be used to explicitly disable C++ support.
2462 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2463 backslash in a target when C++ was disabled. This confuses some
2464 versions of "make", apparently. Using an intermediate variable solves
2465 this. (Same for CPP_LOBJ.)
2467 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2468 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2469 necessary on certain architectures.
2471 4. Added a setting of -export-symbols-regex to the link command to remove
2472 those symbols that are exported in the C sense, but actually are local
2473 within the library, and not documented. Their names all begin with
2474 "_pcre_". This is not a perfect job, because (a) we have to except some
2475 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2476 available (and never for static libraries). I have made a note to try to
2477 find a way round (a) in the future.
2480 Version 6.2 01-Aug-05
2481 ---------------------
2483 1. There was no test for integer overflow of quantifier values. A construction
2484 such as {1111111111111111} would give undefined results. What is worse, if
2485 a minimum quantifier for a parenthesized subpattern overflowed and became
2486 negative, the calculation of the memory size went wrong. This could have
2487 led to memory overwriting.
2489 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2491 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2492 operating environments where this matters.
2494 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2495 PCRE options from within the C++ wrapper.
2497 5. Named capturing subpatterns were not being correctly counted when a pattern
2498 was compiled. This caused two problems: (a) If there were more than 100
2499 such subpatterns, the calculation of the memory needed for the whole
2500 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2501 back references of the form \12, where the number was greater than 9, were
2502 not recognized as back references, even though there were sufficient
2503 previous subpatterns.
2505 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2506 versions of gcc, e.g. 2.95.4.
2509 Version 6.1 21-Jun-05
2510 ---------------------
2512 1. There was one reference to the variable "posix" in pcretest.c that was not
2513 surrounded by "#if !defined NOPOSIX".
2515 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2516 the cross-check on the old pcre_info() function, for the benefit of the
2517 cut-down version of PCRE that is currently imported into Exim.
2519 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2520 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2521 patterns that start (?i) but I don't think that matters. The use of (?i) is
2522 just an example; this all applies to the other options as well.
2524 4. Since libtool seems to echo the compile commands it is issuing, the output
2525 from "make" can be reduced a bit by putting "@" in front of each libtool
2526 compile command.
2528 5. Patch from the folks at Google for configure.in to be a bit more thorough
2529 in checking for a suitable C++ installation before trying to compile the
2530 C++ stuff. This should fix a reported problem when a compiler was present,
2531 but no suitable headers.
2533 6. The man pages all had just "PCRE" as their title. I have changed them to
2534 be the relevant file name. I have also arranged that these names are
2535 retained in the file doc/pcre.txt, which is a concatenation in text format
2536 of all the man pages except the little individual ones for each function.
2538 7. The NON-UNIX-USE file had not been updated for the different set of source
2539 files that come with release 6. I also added a few comments about the C++
2540 wrapper.
2543 Version 6.0 07-Jun-05
2544 ---------------------
2546 1. Some minor internal re-organization to help with my DFA experiments.
2548 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2549 didn't matter for the library itself when fully configured, but did matter
2550 when compiling without UCP support, or within Exim, where the ucp files are
2551 not imported.
2553 3. Refactoring of the library code to split up the various functions into
2554 different source modules. The addition of the new DFA matching code (see
2555 below) to a single monolithic source would have made it really too
2556 unwieldy, quite apart from causing all the code to be include in a
2557 statically linked application, when only some functions are used. This is
2558 relevant even without the DFA addition now that patterns can be compiled in
2559 one application and matched in another.
2561 The downside of splitting up is that there have to be some external
2562 functions and data tables that are used internally in different modules of
2563 the library but which are not part of the API. These have all had their
2564 names changed to start with "_pcre_" so that they are unlikely to clash
2565 with other external names.
2567 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2568 a different (DFA) algorithm. Although it is slower than the original
2569 function, it does have some advantages for certain types of matching
2570 problem.
2572 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2573 including restarting after a partial match.
2575 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2576 defined when compiling for Windows was sent to me. I have put it into the
2577 code, though I have no means of testing or verifying it.
2579 7. Added the pcre_refcount() auxiliary function.
2581 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2582 match before or at the first newline in the subject string. In pcretest,
2583 the /f option on a pattern can be used to set this.
2585 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2586 would behave wrongly. This has been present in PCRE since release 4.0.
2588 10. A number of changes to the pcregrep command:
2590 (a) Refactored how -x works; insert ^(...)$ instead of setting
2591 PCRE_ANCHORED and checking the length, in preparation for adding
2592 something similar for -w.
2594 (b) Added the -w (match as a word) option.
2596 (c) Refactored the way lines are read and buffered so as to have more
2597 than one at a time available.
2599 (d) Implemented a pcregrep test script.
2601 (e) Added the -M (multiline match) option. This allows patterns to match
2602 over several lines of the subject. The buffering ensures that at least
2603 8K, or the rest of the document (whichever is the shorter) is available
2604 for matching (and similarly the previous 8K for lookbehind assertions).
2606 (f) Changed the --help output so that it now says
2608 -w, --word-regex(p)
2610 instead of two lines, one with "regex" and the other with "regexp"
2611 because that confused at least one person since the short forms are the
2612 same. (This required a bit of code, as the output is generated
2613 automatically from a table. It wasn't just a text change.)
2615 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2616 option but starts with a hyphen. Could be a pattern or a path name
2617 starting with a hyphen, for instance.
2619 (h) "-" can be given as a file name to represent stdin.
2621 (i) When file names are being printed, "(standard input)" is used for
2622 the standard input, for compatibility with GNU grep. Previously
2623 "<stdin>" was used.
2625 (j) The option --label=xxx can be used to supply a name to be used for
2626 stdin when file names are being printed. There is no short form.
2628 (k) Re-factored the options decoding logic because we are going to add
2629 two more options that take data. Such options can now be given in four
2630 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2632 (l) Added the -A, -B, and -C options for requesting that lines of context
2633 around matches be printed.
2635 (m) Added the -L option to print the names of files that do not contain
2636 any matching lines, that is, the complement of -l.
2638 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2639 continue to scan other files.
2641 (o) The -s option was incorrectly implemented. For compatibility with other
2642 greps, it now suppresses the error message for a non-existent or non-
2643 accessible file (but not the return code). There is a new option called
2644 -q that suppresses the output of matching lines, which was what -s was
2645 previously doing.
2647 (p) Added --include and --exclude options to specify files for inclusion
2648 and exclusion when recursing.
2650 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2651 Hopefully, it now does.
2653 12. Missing cast in pcre_study().
2655 13. Added an "uninstall" target to the makefile.
2657 14. Replaced "extern" in the function prototypes in Makefile.in with
2658 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2659 world, but is set differently for Windows.
2661 15. Added a second compiling function called pcre_compile2(). The only
2662 difference is that it has an extra argument, which is a pointer to an
2663 integer error code. When there is a compile-time failure, this is set
2664 non-zero, in addition to the error test pointer being set to point to an
2665 error message. The new argument may be NULL if no error number is required
2666 (but then you may as well call pcre_compile(), which is now just a
2667 wrapper). This facility is provided because some applications need a
2668 numeric error indication, but it has also enabled me to tidy up the way
2669 compile-time errors are handled in the POSIX wrapper.
2671 16. Added VPATH=.libs to the makefile; this should help when building with one
2672 prefix path and installing with another. (Or so I'm told by someone who
2673 knows more about this stuff than I do.)
2675 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2676 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2677 match everything, including newlines. This is not POSIX-compatible, but
2678 somebody wanted the feature. From pcretest it can be activated by using
2679 both the P and the s flags.
2681 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2683 19. libpcre.pc was being incorrectly installed as executable.
2685 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2686 it now also looks for '\r' so that it will work unmodified on Windows.
2688 21. Added Google's contributed C++ wrapper to the distribution.
2690 22. Added some untidy missing memory free() calls in pcretest, to keep
2691 Electric Fence happy when testing.
2695 Version 5.0 13-Sep-04
2696 ---------------------
2698 1. Internal change: literal characters are no longer packed up into items
2699 containing multiple characters in a single byte-string. Each character
2700 is now matched using a separate opcode. However, there may be more than one
2701 byte in the character in UTF-8 mode.
2703 2. The pcre_callout_block structure has two new fields: pattern_position and
2704 next_item_length. These contain the offset in the pattern to the next match
2705 item, and its length, respectively.
2707 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2708 insertion of callouts before each pattern item. Added the /C option to
2709 pcretest to make use of this.
2711 4. On the advice of a Windows user, the lines
2713 #if defined(_WIN32) || defined(WIN32)
2714 _setmode( _fileno( stdout ), 0x8000 );
2715 #endif /* defined(_WIN32) || defined(WIN32) */
2717 have been added to the source of pcretest. This apparently does useful
2718 magic in relation to line terminators.
2720 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2721 for the benefit of those environments where the "b" makes a difference.
2723 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2724 to know about it. I have put a hack into configure.in that adds in code
2725 to set GCC=yes if CC=icc. This seems to end up at a point in the
2726 generated configure script that is early enough to affect the setting of
2727 compiler options, which is what is needed, but I have no means of testing
2728 whether it really works. (The user who reported this had patched the
2729 generated configure script, which of course I cannot do.)
2731 LATER: After change 22 below (new libtool files), the configure script
2732 seems to know about icc (and also ecc). Therefore, I have commented out
2733 this hack in configure.in.
2735 7. Added support for pkg-config (2 patches were sent in).
2737 8. Negated POSIX character classes that used a combination of internal tables
2738 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2739 [[:^ascii]]. Typically, they would match almost any characters. The other
2740 POSIX classes were not broken in this way.
2742 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2743 to find the match, as PCRE was deluded into thinking that the match had to
2744 start at the start point or following a newline. The same bug applied to
2745 patterns with negative forward assertions or any backward assertions
2746 preceding ".*" at the start, unless the pattern required a fixed first
2747 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2749 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2750 starting at the last subject character, bytes beyond the end of the subject
2751 string were read.
2753 11. Renamed the variable "class" as "classbits" to make life easier for C++
2754 users. (Previously there was a macro definition, but it apparently wasn't
2755 enough.)
2757 12. Added the new field "tables" to the extra data so that tables can be passed
2758 in at exec time, or the internal tables can be re-selected. This allows
2759 a compiled regex to be saved and re-used at a later time by a different
2760 program that might have everything at different addresses.
2762 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2763 -R library as well as a -L library.
2765 14. The debugging options of pcretest (-d on the command line or D on a
2766 pattern) showed incorrect output for anything following an extended class
2767 that contained multibyte characters and which was followed by a quantifier.
2769 15. Added optional support for general category Unicode character properties
2770 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2771 support. It adds about 90K to the size of the library. The meanings of the
2772 inbuilt class escapes such as \d and \s have NOT been changed.
2774 16. Updated pcredemo.c to include calls to free() to release the memory for the
2775 compiled pattern.
2777 17. The generated file chartables.c was being created in the source directory
2778 instead of in the building directory. This caused the build to fail if the
2779 source directory was different from the building directory, and was
2780 read-only.
2782 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2783 file. No doubt somebody will tell me if they don't make sense... Also added
2784 Dan Mooney's comments about building on OpenVMS.
2786 19. Added support for partial matching via the PCRE_PARTIAL option for
2787 pcre_exec() and the \P data escape in pcretest.
2789 20. Extended pcretest with 3 new pattern features:
2791 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2792 write the compiled pattern to the file whose name is "rest-of-line".
2793 This is a straight binary dump of the data, with the saved pointer to
2794 the character tables forced to be NULL. The study data, if any, is
2795 written too. After writing, pcretest reads a new pattern.
2797 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2798 compiled pattern from the given file. There must not be any
2799 occurrences of "<" in the file name (pretty unlikely); if there are,
2800 pcretest will instead treat the initial "<" as a pattern delimiter.
2801 After reading in the pattern, pcretest goes on to read data lines as
2802 usual.
2804 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2805 and 16-bit fields in a compiled pattern, to simulate a pattern that
2806 was compiled on a host of opposite endianness.
2808 21. The pcre-exec() function can now cope with patterns that were compiled on
2809 hosts of opposite endianness, with this restriction:
2811 As for any compiled expression that is saved and used later, the tables
2812 pointer field cannot be preserved; the extra_data field in the arguments
2813 to pcre_exec() should be used to pass in a tables address if a value
2814 other than the default internal tables were used at compile time.
2816 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2817 now diagnosed as an error. Previously, most of the time, a negative number
2818 would have been treated as zero, but if in addition "ovector" was passed as
2819 NULL, a crash could occur.
2821 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2822 new versions from the libtool 1.5 distribution (the last one is a copy of
2823 a file called libtool.m4). This seems to have fixed the need to patch
2824 "configure" to support Darwin 1.3 (which I used to do). However, I still
2825 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2826 workstation).
2828 24. Changed the PCRE licence to be the more standard "BSD" licence.
2831 Version 4.5 01-Dec-03
2832 ---------------------
2834 1. There has been some re-arrangement of the code for the match() function so
2835 that it can be compiled in a version that does not call itself recursively.
2836 Instead, it keeps those local variables that need separate instances for
2837 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2838 needs to "recurse". Keeping track of where control must go is done by means
2839 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2840 hide most of the details from the main code, and operates only if
2841 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2842 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2843 operating.
2845 To make it easier for callers to provide specially tailored get/free
2846 functions for this usage, two new functions, pcre_stack_malloc, and
2847 pcre_stack_free, are used. They are always called in strict stacking order,
2848 and the size of block requested is always the same.
2850 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2851 PCRE has been compiled to use the stack or the heap for recursion. The
2852 -C option of pcretest uses this to show which version is compiled.
2854 A new data escape \S, is added to pcretest; it causes the amounts of store
2855 obtained and freed by both kinds of malloc/free at match time to be added
2856 to the output.
2858 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2859 what's available on my current Linux desktop machine.
2861 3. When matching a UTF-8 string, the test for a valid string at the start has
2862 been extended. If start_offset is not zero, PCRE now checks that it points
2863 to a byte that is the start of a UTF-8 character. If not, it returns
2864 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2865 this is necessary because there may be backward assertions in the pattern.
2866 When matching the same subject several times, it may save resources to use
2867 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2869 4. The code for checking the validity of UTF-8 strings has been tightened so
2870 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2871 containing "overlong sequences".
2873 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2874 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2875 should have been "&", but it just so happened that all the cases this let
2876 through by mistake were picked up later in the function.
2878 6. I had used a variable called "isblank" - this is a C99 function, causing
2879 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2881 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2882 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2883 so the version is shown; (c) stop "make test" echoing "./RunTest".
2885 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2887 9. The prototype for memmove() for systems that don't have it was using
2888 size_t, but the inclusion of the header that defines size_t was later. I've
2889 moved the #includes for the C headers earlier to avoid this.
2891 10. Added some adjustments to the code to make it easier to compiler on certain
2892 special systems:
2894 (a) Some "const" qualifiers were missing.
2895 (b) Added the macro EXPORT before all exported functions; by default this
2896 is defined to be empty.
2897 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2898 that it reads its output file name as an argument instead of writing
2899 to the standard output and assuming this can be redirected.
2901 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2902 class containing characters with values greater than 255, PCRE compilation
2903 went into a loop.
2905 12. A recursive reference to a subpattern that was within another subpattern
2906 that had a minimum quantifier of zero caused PCRE to crash. For example,
2907 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2908 recursion. If the recursively-called subpattern itself had a zero repeat,
2909 that was OK.
2911 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2912 buffer into which it was copied (for escape processing) was still set at
2913 1024, so long lines caused crashes.
2915 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2916 "internal error: code overflow...". This applied to any character class
2917 that was followed by a possessive quantifier.
2919 15. Modified the Makefile to add libpcre.la as a prerequisite for
2920 libpcreposix.la because I was told this is needed for a parallel build to
2921 work.
2923 16. If a pattern that contained .* following optional items at the start was
2924 studied, the wrong optimizing data was generated, leading to matching
2925 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2926 matching string must start with a or b or c. The correct conclusion for
2927 this pattern is that a match can start with any character.
2930 Version 4.4 13-Aug-03
2931 ---------------------
2933 1. In UTF-8 mode, a character class containing characters with values between
2934 127 and 255 was not handled correctly if the compiled pattern was studied.
2935 In fixing this, I have also improved the studying algorithm for such
2936 classes (slightly).
2938 2. Three internal functions had redundant arguments passed to them. Removal
2939 might give a very teeny performance improvement.
2941 3. Documentation bug: the value of the capture_top field in a callout is *one
2942 more than* the number of the hightest numbered captured substring.
2944 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2945 in incorrectly linking with a previously installed version. They now link
2946 explicitly with libpcre.la.
2948 5. configure.in no longer needs to recognize Cygwin specially.
2950 6. A problem in pcre.in for Windows platforms is fixed.
2952 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2953 pcretest, it used to include the size of the study block as part of its
2954 output. Unfortunately, the structure contains a field that has a different
2955 size on different hardware architectures. This meant that the tests that
2956 showed this size failed. As the block is currently always of a fixed size,
2957 this information isn't actually particularly useful in pcretest output, so
2958 I have just removed it.
2960 8. Three pre-processor statements accidentally did not start in column 1.
2961 Sadly, there are *still* compilers around that complain, even though
2962 standard C has not required this for well over a decade. Sigh.
2964 9. In pcretest, the code for checking callouts passed small integers in the
2965 callout_data field, which is a void * field. However, some picky compilers
2966 complained about the casts involved for this on 64-bit systems. Now
2967 pcretest passes the address of the small integer instead, which should get
2968 rid of the warnings.
2970 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2971 both compile and run time, and gives an error if an invalid UTF-8 sequence
2972 is found. There is a option for disabling this check in cases where the
2973 string is known to be correct and/or the maximum performance is wanted.
2975 11. In response to a bug report, I changed one line in Makefile.in from
2977 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2978 to
2979 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2981 to look similar to other lines, but I have no way of telling whether this
2982 is the right thing to do, as I do not use Windows. No doubt I'll get told
2983 if it's wrong...
2986 Version 4.3 21-May-03
2987 ---------------------
2989 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2990 Makefile.
2992 2. Some refactoring to improve the quality of the code:
2994 (i) The utf8_table... variables are now declared "const".
2996 (ii) The code for \cx, which used the "case flipping" table to upper case
2997 lower case letters, now just substracts 32. This is ASCII-specific,
2998 but the whole concept of \cx is ASCII-specific, so it seems
2999 reasonable.
3001 (iii) PCRE was using its character types table to recognize decimal and
3002 hexadecimal digits in the pattern. This is silly, because it handles
3003 only 0-9, a-f, and A-F, but the character types table is locale-
3004 specific, which means strange things might happen. A private
3005 table is now used for this - though it costs 256 bytes, a table is
3006 much faster than multiple explicit tests. Of course, the standard
3007 character types table is still used for matching digits in subject
3008 strings against \d.
3010 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3011 ending in _t are). So I've renamed it as ESC_tee.
3013 3. The first argument for regexec() in the POSIX wrapper should have been
3014 defined as "const".
3016 4. Changed pcretest to use malloc() for its buffers so that they can be
3017 Electric Fenced for debugging.
3019 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3020 to read one or more bytes before the start of the subject string. Often this
3021 had no effect on PCRE's behaviour, but in some circumstances it could
3022 provoke a segmentation fault.
3024 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3025 to try to read one or more bytes before the start of the subject string.
3027 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3028 UTF-8 support could misbehave in various ways if the subject string
3029 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3030 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3031 back over UTF-8 characters.)
3034 Version 4.2 14-Apr-03
3035 ---------------------
3037 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3039 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3040 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3041 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3042 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3043 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3045 Note: automatic setting of the BUILD variables is not yet working
3046 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3047 compile-time but not at link-time
3048 [LINK]: use for linking executables only
3049 make different versions for Windows and non-Windows
3050 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3051 libraries
3052 [LINK_FOR_BUILD]: new variable
3053 [OBJEXT]: use throughout
3054 [EXEEXT]: use throughout
3055 <winshared>: new target
3056 <wininstall>: new target
3057 <dftables.o>: use native compiler
3058 <dftables>: use native linker
3059 <install>: handle Windows platform correctly
3060 <clean>: ditto
3061 <check>: ditto
3062 copy DLL to top builddir before testing
3064 As part of these changes, -no-undefined was removed again. This was reported
3065 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3066 in any case.
3068 3. Some tidies to get rid of compiler warnings:
3070 . In the match_data structure, match_limit was an unsigned long int, whereas
3071 match_call_count was an int. I've made them both unsigned long ints.
3073 . In pcretest the fact that a const uschar * doesn't automatically cast to
3074 a void * provoked a warning.
3076 . Turning on some more compiler warnings threw up some "shadow" variables
3077 and a few more missing casts.
3079 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3080 option, a class that contained a single character with a value between 128
3081 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3083 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3084 option, a class that contained several characters, but with at least one
3085 whose value was between 128 and 255 caused PCRE to crash.
3088 Version 4.1 12-Mar-03
3089 ---------------------
3091 1. Compiling with gcc -pedantic found a couple of places where casts were
3092 needed, and a string in dftables.c that was longer than standard compilers are
3093 required to support.
3095 2. Compiling with Sun's compiler found a few more places where the code could
3096 be tidied up in order to avoid warnings.
3098 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3099 first of these names is deprecated in the latest Autoconf in favour of the name
3100 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3101 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3102 analogy I have changed it to CFLAGS_FOR_BUILD.
3104 4. Added -no-undefined to the linking command in the Makefile, because this is
3105 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3106 linking step for the pcreposix library.
3108 5. PCRE was failing to diagnose the case of two named groups with the same
3109 name.
3111 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3112 literal character that is needed in the subject for a match, and scans along to
3113 ensure that it is present before embarking on the full matching process. This
3114 saves time in cases of nested unlimited repeats that are never going to match.
3115 Problem: the scan can take a lot of time if the subject is very long (e.g.
3116 megabytes), thus penalizing straightforward matches. It is now done only if the
3117 amount of subject to be scanned is less than 1000 bytes.
3119 7. A lesser problem with the same optimization is that it was recording the
3120 first character of an anchored pattern as "needed", thus provoking a search
3121 right along the subject, even when the first match of the pattern was going to
3122 fail. The "needed" character is now not set for anchored patterns, unless it
3123 follows something in the pattern that is of non-fixed length. Thus, it still
3124 fulfils its original purpose of finding quick non-matches in cases of nested
3125 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3128 Version 4.0 17-Feb-03
3129 ---------------------
3131 1. If a comment in an extended regex that started immediately after a meta-item
3132 extended to the end of string, PCRE compiled incorrect data. This could lead to
3133 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3135 2. Moved to autoconf 2.53 and libtool 1.4.2.
3137 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3138 the special perltest8 script is no longer needed - all the tests can be run
3139 from a single perltest script.
3141 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3142 by \s. It has now been removed in PCRE. This means it isn't recognized as
3143 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3144 class [:space:] *does* include VT, thereby creating a mess.
3146 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3147 space and tab.
3149 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3150 its new features into the main test script, reducing the number of scripts.
3152 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3153 were backward compatible, and made the (?i) apply to the whole pattern, as if
3154 /i were given. Now it behaves more logically, and applies the option setting
3155 only to what follows. PCRE has been changed to follow suit. However, if it
3156 finds options settings right at the start of the pattern, it extracts them into
3157 the global options, as before. Thus, they show up in the info data.
3159 8. Added support for the \Q...\E escape sequence. Characters in between are
3160 treated as literals. This is slightly different from Perl in that $ and @ are
3161 also handled as literals inside the quotes. In Perl, they will cause variable
3162 interpolation. Note the following examples:
3164 Pattern PCRE matches Perl matches
3166 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3167 \Qabc\$xyz\E abc\$xyz abc\$xyz
3168 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3170 For compatibility with Perl, \Q...\E sequences are recognized inside character
3171 classes as well as outside them.
3173 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3174 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3175 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3176 signed/unsigned warnings.
3178 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3179 option for pcretest, so I've replaced it by a simple function that does just
3180 that job.
3182 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3183 "pcregrep -".
3185 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3186 Java package. This provides some syntactic sugar for simple cases of what my
3187 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3188 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3189 item, you can use this simplified notation. Note that only makes sense with
3190 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3191 greediness, whatever the setting of the PCRE_UNGREEDY option.
3193 13. A change of greediness default within a pattern was not taking effect at
3194 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3195 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3196 was abstracted outside.
3198 14. PCRE now supports the \G assertion. It is true when the current matching
3199 position is at the start point of the match. This differs from \A when the
3200 starting offset is non-zero. Used with the /g option of pcretest (or similar
3201 code), it works in the same way as it does for Perl's /g option. If all
3202 alternatives of a regex begin with \G, the expression is anchored to the start
3203 match position, and the "anchored" flag is set in the compiled expression.
3205 15. Some bugs concerning the handling of certain option changes within patterns
3206 have been fixed. These applied to options other than (?ims). For example,
3207 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3208 been the other way round. Some of this was related to change 7 above.
3210 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3211 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3212 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3213 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3215 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3216 mode. Unlike ".", it always matches newline, whatever the setting of
3217 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3218 assertions. Perl allows it, but it doesn't (in general) work because it can't
3219 calculate the length of the lookbehind. At least, that's the case for Perl
3220 5.8.0 - I've been told they are going to document that it doesn't work in
3221 future.
3223 18. Added an error diagnosis for escapes that PCRE does not support: these are
3224 \L, \l, \N, \P, \p, \U, \u, and \X.
3226 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3227 reading past the end of the pattern in cases such as /[abcd/.
3229 20. PCRE was getting more memory than necessary for patterns with classes that
3230 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3232 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3233 compiling PCRE for use with Virtual Pascal.
3235 22. Small fix to the Makefile to make it work properly if the build is done
3236 outside the source tree.
3238 23. Added a new extension: a condition to go with recursion. If a conditional
3239 subpattern starts with (?(R) the "true" branch is used if recursion has
3240 happened, whereas the "false" branch is used only at the top level.
3242 24. When there was a very long string of literal characters (over 255 bytes
3243 without UTF support, over 250 bytes with UTF support), the computation of how
3244 much memory was required could be incorrect, leading to segfaults or other
3245 strange effects.
3247 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3248 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3249 there was a subsequent back reference to those brackets. This meant that, for
3250 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3251 possible to check for precisely this case. All we can do is abandon the
3252 optimization if .* occurs inside capturing brackets when there are any back
3253 references whatsoever. (See below for a better fix that came later.)
3255 26. The handling of the optimization for finding the first character of a
3256 non-anchored pattern, and for finding a character that is required later in the
3257 match were failing in some cases. This didn't break the matching; it just
3258 failed to optimize when it could. The way this is done has been re-implemented.
3260 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3262 28. Added a new feature that provides some of the functionality that Perl
3263 provides with (?{...}). The facility is termed a "callout". The way it is done
3264 in PCRE is for the caller to provide an optional function, by setting
3265 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3266 global variable. By default it is unset, which disables all calling out. To get
3267 the function called, the regex must include (?C) at appropriate points. This
3268 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3269 This provides a means of identifying different callout points. When PCRE
3270 reaches such a point in the regex, if pcre_callout has been set, the external
3271 function is called. It is provided with data in a structure called
3272 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3273 matching continues; if it returns a non-zero value, the match at the current
3274 point fails. However, backtracking will occur if possible. [This was changed
3275 later and other features added - see item 49 below.]
3277 29. pcretest is upgraded to test the callout functionality. It provides a
3278 callout function that displays information. By default, it shows the start of
3279 the match and the current position in the text. There are some new data escapes
3280 to vary what happens:
3282 \C+ in addition, show current contents of captured substrings
3283 \C- do not supply a callout function
3284 \C!n return 1 when callout number n is reached
3285 \C!n!m return 1 when callout number n is reached for the mth time
3287 30. If pcregrep was called with the -l option and just a single file name, it
3288 output "<stdin>" if a match was found, instead of the file name.
3290 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3291 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3292 pcre_exec(). This saves a malloc/free per call. The default value of
3293 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3294 when configuring.
3296 32. The default maximum size of a compiled pattern is 64K. There have been a
3297 few cases of people hitting this limit. The code now uses macros to handle the
3298 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3299 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3300 configuring. Tests 2 and 5 work only with 2-byte links because they output
3301 debugging information about compiled patterns.
3303 33. Internal code re-arrangements:
3305 (a) Moved the debugging function for printing out a compiled regex into
3306 its own source file (printint.c) and used #include to pull it into
3307 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3308 separate copies.
3310 (b) Defined the list of op-code names for debugging as a macro in
3311 internal.h so that it is next to the definition of the opcodes.
3313 (c) Defined a table of op-code lengths for simpler skipping along compiled
3314 code. This is again a macro in internal.h so that it is next to the
3315 definition of the opcodes.
3317 34. Added support for recursive calls to individual subpatterns, along the
3318 lines of Robin Houston's patch (but implemented somewhat differently).
3320 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3321 allow it to read and process whole directories in Win32. This code was
3322 contributed by Lionel Fourquaux; it has not been tested by me.
3324 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3325 used to name a group. Names consist of alphanumerics and underscores, and must
3326 be unique. Back references use the syntax (?P=name) and recursive calls use
3327 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3328 numbers. The function pcre_fullinfo() can be used after compilation to extract
3329 a name/number map. There are three relevant calls:
3331 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3332 PCRE_INFO_NAMECOUNT yields the number of entries
3333 PCRE_INFO_NAMETABLE yields a pointer to the map.
3335 The map is a vector of fixed-size entries. The size of each entry depends on
3336 the length of the longest name used. The first two bytes of each entry are the
3337 group number, most significant byte first. There follows the corresponding
3338 name, zero terminated. The names are in alphabetical order.
3340 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3341 case instead of 255. Making it the same both with and without UTF-8 support
3342 means that the same test output works with both.
3344 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3345 calling malloc() with a zero argument.
3347 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3348 optimization. I've improved things by keeping a bitmap of backreferences with
3349 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3350 fact referenced, the optimization can be applied. It is unlikely that a
3351 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3352 the match to follow \n) will appear inside brackets with a number greater than
3353 31, but if it does, any back reference > 31 suppresses the optimization.
3355 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3356 of disabling numbered capturing parentheses. Any opening parenthesis that is
3357 not followed by ? behaves as if it were followed by ?: but named parentheses
3358 can still be used for capturing (and they will acquire numbers in the usual
3359 way).
3361 41. Redesigned the return codes from the match() function into yes/no/error so
3362 that errors can be passed back from deep inside the nested calls. A malloc
3363 failure while inside a recursive subpattern call now causes the
3364 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3366 42. It is now possible to set a limit on the number of times the match()
3367 function is called in a call to pcre_exec(). This facility makes it possible to
3368 limit the amount of recursion and backtracking, though not in a directly
3369 obvious way, because the match() function is used in a number of different
3370 circumstances. The count starts from zero for each position in the subject
3371 string (for non-anchored patterns). The default limit is, for compatibility, a
3372 large number, namely 10 000 000. You can change this in two ways:
3374 (a) When configuring PCRE before making, you can use --with-match-limit=n
3375 to set a default value for the compiled library.
3377 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3378 a different value is set. See 45 below.
3380 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3382 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3383 of things that can be changed at compile time. The first argument specifies
3384 what is wanted and the second points to where the information is to be placed.
3385 The current list of available information is:
3389 The output is an integer that is set to one if UTF-8 support is available;
3390 otherwise it is set to zero.
3394 The output is an integer that it set to the value of the code that is used for
3395 newline. It is either LF (10) or CR (13).
3399 The output is an integer that contains the number of bytes used for internal
3400 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3404 The output is an integer that contains the threshold above which the POSIX
3405 interface uses malloc() for output vectors. See item 31 above.
3409 The output is an unsigned integer that contains the default limit of the number
3410 of match() calls in a pcre_exec() execution. See 42 above.
3412 44. pcretest has been upgraded by the addition of the -C option. This causes it
3413 to extract all the available output from the new pcre_config() function, and to
3414 output it. The program then exits immediately.
3416 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3417 order to support additional features. One way would have been to define
3418 pcre_exec2() (for example) with extra arguments, but this would not have been
3419 extensible, and would also have required all calls to the original function to
3420 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3421 is used for passing in "extra" data from pcre_study().
3423 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3424 contains the following fields:
3426 flags a bitmap indicating which of the following fields are set
3427 study_data opaque data from pcre_study()
3428 match_limit a way of specifying a limit on match() calls for a specific
3429 call to pcre_exec()
3430 callout_data data for callouts (see 49 below)
3432 The flag bits are also defined in pcre.h, and are
3438 The pcre_study() function now returns one of these new pcre_extra blocks, with
3439 the actual study data pointed to by the study_data field, and the
3440 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3441 before. That is, this change is entirely upwards-compatible and requires no
3442 change to existing code.
3444 If you want to pass in additional data to pcre_exec(), you can either place it
3445 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3446 block.
3448 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3449 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3450 times with different match limits, until it finds the minimum value needed for
3451 pcre_exec() to complete. The value is then output. This can be instructive; for
3452 most simple matches the number is quite small, but for pathological cases it
3453 gets very large very quickly.
3455 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3456 returns the size of the data block pointed to by the study_data field in a
3457 pcre_extra block, that is, the value that was passed as the argument to
3458 pcre_malloc() when PCRE was getting memory in which to place the information
3459 created by pcre_study(). The fourth argument should point to a size_t variable.
3460 pcretest has been extended so that this information is shown after a successful
3461 pcre_study() call when information about the compiled regex is being displayed.
3463 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3464 because what follows is always an absolute path. (Later: it turns out that this
3465 is more than cosmetic for MinGW, because it doesn't like empty path
3466 components.)
3468 49. Some changes have been made to the callout feature (see 28 above):
3470 (i) A callout function now has three choices for what it returns:
3472 0 => success, carry on matching
3473 > 0 => failure at this point, but backtrack if possible
3474 < 0 => serious error, return this value from pcre_exec()
3476 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3477 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3478 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3479 use by callout functions. It will never be used by PCRE itself.
3481 (ii) The pcre_extra structure (see 45 above) has a void * field called
3482 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3483 pcre_callout_block structure has a field of the same name. The contents of
3484 the field passed in the pcre_extra structure are passed to the callout
3485 function in the corresponding field in the callout block. This makes it
3486 easier to use the same callout-containing regex from multiple threads. For
3487 testing, the pcretest program has a new data escape
3489 \C*n pass the number n (may be negative) as callout_data
3491 If the callout function in pcretest receives a non-zero value as
3492 callout_data, it returns that value.
3494 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3495 there were some redundant $(CFLAGS) in commands that are now specified as
3496 $(LINK), which already includes $(CFLAGS).
3498 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3499 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3500 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3501 one-byte characters throughout. Note that case-insensitive matching applies
3502 only to characters whose values are less than 256. PCRE doesn't support the
3503 notion of cases for higher-valued characters.
3505 (i) A character class whose characters are all within 0-255 is handled as
3506 a bit map, and the map is inverted for negative classes. Previously, a
3507 character > 255 always failed to match such a class; however it should
3508 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3510 (ii) A negated character class with a single character < 255 is coded as
3511 "not this character" (OP_NOT). This wasn't working properly when the test
3512 character was multibyte, either singly or repeated.
3514 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3515 mode, for example: \x{100}{2,3}.
3517 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3518 singly or repeated) now correctly test multibyte characters. However,
3519 PCRE doesn't recognize any characters with values greater than 255 as
3520 digits, spaces, or word characters. Such characters always match \D, \S,
3521 and \W, and never match \d, \s, or \w.
3523 (v) Classes may now contain characters and character ranges with values
3524 greater than 255. For example: [ab\x{100}-\x{400}].
3526 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3527 PCRE in UTF-8 mode.
3529 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3530 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3531 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3532 value.)
3534 53. The single man page has become too large. I have therefore split it up into
3535 a number of separate man pages. These also give rise to individual HTML pages;
3536 these are now put in a separate directory, and there is an index.html page that
3537 lists them all. Some hyperlinking between the pages has been installed.
3539 54. Added convenience functions for handling named capturing parentheses.
3541 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3542 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3543 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3544 are faulted.
3546 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3547 calling configure. These values are used when compiling the dftables.c program
3548 which is run to generate the source of the default character tables. They
3549 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3550 you will need to set these values.
3552 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3555 Version 3.9 02-Jan-02
3556 ---------------------
3558 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3560 2. If --disable-static was given, the building process failed when trying to
3561 build pcretest and pcregrep. (For some reason it was using libtool to compile
3562 them, which is not right, as they aren't part of the library.)
3565 Version 3.8 18-Dec-01
3566 ---------------------
3568 1. The experimental UTF-8 code was completely screwed up. It was packing the
3569 bytes in the wrong order. How dumb can you get?
3572 Version 3.7 29-Oct-01
3573 ---------------------
3575 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3576 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3577 this didn't happen under Solaris 8, where I normally test things.
3579 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3580 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3581 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3582 this fix an hour or so after the initial 3.7 release.)
3585 Version 3.6 23-Oct-01
3586 ---------------------
3588 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3589 offsets passed as NULL with zero offset count.
3591 2. The config.guess and config.sub files had not been updated when I moved to
3592 the latest autoconf.
3595 Version 3.5 15-Aug-01
3596 ---------------------
3598 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3599 had been forgotten.
3601 2. By using declared but undefined structures, we can avoid using "void"
3602 definitions in pcre.h while keeping the internal definitions of the structures
3603 private.
3605 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3606 user point of view, this means that both static and shared libraries are built
3607 by default, but this can be individually controlled. More of the work of
3608 handling this static/shared cases is now inside libtool instead of PCRE's make
3609 file.
3611 4. The pcretest utility is now installed along with pcregrep because it is
3612 useful for users (to test regexs) and by doing this, it automatically gets
3613 relinked by libtool. The documentation has been turned into a man page, so
3614 there are now .1, .txt, and .html versions in /doc.
3616 5. Upgrades to pcregrep:
3617 (i) Added long-form option names like gnu grep.
3618 (ii) Added --help to list all options with an explanatory phrase.
3619 (iii) Added -r, --recursive to recurse into sub-directories.
3620 (iv) Added -f, --file to read patterns from a file.
3622 6. pcre_exec() was referring to its "code" argument before testing that
3623 argument for NULL (and giving an error if it was NULL).
3625 7. Upgraded Makefile.in to allow for compiling in a different directory from
3626 the source directory.
3628 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3629 options bits, the pointer it was passed was to an int instead of to an unsigned
3630 long int. This mattered only on 64-bit systems.
3632 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3633 generated) instead of pcre.in, which it its source. Also made the same change
3634 in several of the .c files.
3636 10. A new release of gcc defines printf() as a macro, which broke pcretest
3637 because it had an ifdef in the middle of a string argument for printf(). Fixed
3638 by using separate calls to printf().
3640 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3641 script, to force use of CR or LF instead of \n in the source. On non-Unix
3642 systems, the value can be set in config.h.
3644 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3645 absolute limit. Changed the text of the error message to make this clear, and
3646 likewise updated the man page.
3648 13. The limit of 99 on the number of capturing subpatterns has been removed.
3649 The new limit is 65535, which I hope will not be a "real" limit.
3652 Version 3.4 22-Aug-00
3653 ---------------------
3655 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3657 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3660 Version 3.3 01-Aug-00
3661 ---------------------
3663 1. If an octal character was given, but the value was greater than \377, it
3664 was not getting masked to the least significant bits, as documented. This could
3665 lead to crashes in some systems.
3667 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3668 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3670 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3671 These just pass their arguments on to (pcre_free)(), but they are provided
3672 because some uses of PCRE bind it to non-C systems that can call its functions,
3673 but cannot call free() or pcre_free() directly.
3675 4. Add "make test" as a synonym for "make check". Corrected some comments in
3676 the Makefile.
3678 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3679 Makefile.
3681 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3682 command called pgrep for grepping around the active processes.
3684 7. Added the beginnings of support for UTF-8 character strings.
3686 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3687 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3688 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3689 out for the ar command.)
3692 Version 3.2 12-May-00
3693 ---------------------
3695 This is purely a bug fixing release.
3697 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3698 of ZA. This was just one example of several cases that could provoke this bug,
3699 which was introduced by change 9 of version 2.00. The code for breaking
3700 infinite loops after an iteration that matches an empty string was't working
3701 correctly.
3703 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3704 when matched against abbab (for example). After matching an empty string, it
3705 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3706 caused it to match further down the string than it should.
3708 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3709 was there because it doesn't seem to be needed, and it causes trouble on some
3710 systems, as it is not a Standard C header. It has been removed.
3712 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3713 were reported on the Macintosh. The changes were from
3715 while ((c = *(++ptr)) != 0 && c != '\n');
3716 to
3717 while ((c = *(++ptr)) != 0 && c != '\n') ;
3719 Totally extraordinary, but if that's what it takes...
3721 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3722 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3723 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3724 assumes the way PCRE uses memmove() (always moving upwards).
3726 6. PCRE is being used in one environment where strchr() is not available. There
3727 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3728 faster code anyway.
3731 Version 3.1 09-Feb-00
3732 ---------------------
3734 The only change in this release is the fixing of some bugs in Makefile.in for
3735 the "install" target:
3737 (1) It was failing to install pcreposix.h.
3739 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3742 Version 3.0 01-Feb-00
3743 ---------------------
3745 1. Add support for the /+ modifier to perltest (to output $` like it does in
3746 pcretest).
3748 2. Add support for the /g modifier to perltest.
3750 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3751 matches null strings.
3753 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3754 pattern. Perl treats empty patterns specially - it reuses the most recent
3755 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3756 effect.
3758 5. The POSIX interface was broken in that it was just handing over the POSIX
3759 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3760 required a bigger vector, with some working space on the end. This means that
3761 the POSIX wrapper now has to get and free some memory, and copy the results.
3763 6. Added some simple autoconf support, placing the test data and the
3764 documentation in separate directories, re-organizing some of the
3765 information files, and making it build pcre-config (a GNU standard). Also added
3766 libtool support for building PCRE as a shared library, which is now the
3767 default.
3769 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3770 09 are not valid octal constants. Single digits will be used for minor values
3771 less than 10.
3773 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3774 existing programs that set these in the POSIX interface can use PCRE without
3775 modification.
3777 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3778 return all that pcre_info() returns, plus additional data. The pcre_info()
3779 function is retained for compatibility, but is considered to be obsolete.
3781 10. Added experimental recursion feature (?R) to handle one common case that
3782 Perl 5.6 will be able to do with (?p{...}).
3784 11. Added support for POSIX character classes like [:alpha:], which Perl is
3785 adopting.
3788 Version 2.08 31-Aug-99
3789 ----------------------
3791 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3792 trying to match at the startoffset position, but instead was moving forward to
3793 the next newline as if a previous match had failed.
3795 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3796 and could get into a loop if a null string was matched other than at the start
3797 of the subject.
3799 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3800 be distinguished at compile time, and for completeness also added PCRE_DATE.
3802 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3803 in GnuWin32 environments.
3806 Version 2.07 29-Jul-99
3807 ----------------------
3809 1. The documentation is now supplied in plain text form and HTML as well as in
3810 the form of man page sources.
3812 2. C++ compilers don't like assigning (void *) values to other pointer types.
3813 In particular this affects malloc(). Although there is no problem in Standard
3814 C, I've put in casts to keep C++ compilers happy.
3816 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3817 should be (const char *).
3819 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3820 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3821 However, I haven't made this a standard facility. The documentation doesn't
3822 mention it, and the Makefile doesn't support it.
3824 5. The Makefile now contains an "install" target, with editable destinations at
3825 the top of the file. The pcretest program is not installed.
3827 6. pgrep -V now gives the PCRE version number and date.
3829 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3830 causing the entire string to be ignored, instead of just the last character.
3832 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3833 non-matching string, it can take a very, very long time, even for strings of
3834 quite modest length, because of the nested recursion. PCRE now does better in
3835 some of these cases. It does this by remembering the last required literal
3836 character in the pattern, and pre-searching the subject to ensure it is present
3837 before running the real match. In other words, it applies a heuristic to detect
3838 some types of certain failure quickly, and in the above example, if presented
3839 with a string that has no trailing " it gives "no match" very quickly.
3841 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3842 other alternatives are tried instead.
3845 Version 2.06 09-Jun-99
3846 ----------------------
3848 1. Change pcretest's output for amount of store used to show just the code
3849 space, because the remainder (the data block) varies in size between 32-bit and
3850 64-bit systems.
3852 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3853 start matching at. This allows lookbehinds to work when searching for multiple
3854 occurrences in a string.
3856 3. Added additional options to pcretest for testing multiple occurrences:
3858 /+ outputs the rest of the string that follows a match
3859 /g loops for multiple occurrences, using the new startoffset argument
3860 /G loops for multiple occurrences by passing an incremented pointer
3862 4. PCRE wasn't doing the "first character" optimization for patterns starting
3863 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3864 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3865 the letter 'x'. On long subject strings, this gives a significant speed-up.
3868 Version 2.05 21-Apr-99
3869 ----------------------
3871 1. Changed the type of magic_number from int to long int so that it works
3872 properly on 16-bit systems.
3874 2. Fixed a bug which caused patterns starting with .* not to work correctly
3875 when the subject string contained newline characters. PCRE was assuming
3876 anchoring for such patterns in all cases, which is not correct because .* will
3877 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3878 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3879 must be retried after every newline in the subject.
3882 Version 2.04 18-Feb-99
3883 ----------------------
3885 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3886 computation of the store needed to hold the pattern was incorrect (too large).
3887 If such patterns were nested a few deep, this could multiply and become a real
3888 problem.
3890 2. Added /M option to pcretest to show the memory requirement of a specific
3891 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3893 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3894 compiled in such a way that the backtracking after subsequent failure was
3895 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3896 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3899 Version 2.03 02-Feb-99
3900 ----------------------
3902 1. Fixed typo and small mistake in man page.
3904 2. Added 4th condition (GPL supersedes if conflict) and created separate
3905 LICENCE file containing the conditions.
3907 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3908 Perl, that is the internal \ allows the delimiter to be included in the
3909 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3910 the final delimiter, add \ to the end of the pattern (to test the error).
3912 4. Added the convenience functions for extracting substrings after a successful
3913 match. Updated pcretest to make it able to test these functions.
3916 Version 2.02 14-Jan-99
3917 ----------------------
3919 1. Initialized the working variables associated with each extraction so that
3920 their saving and restoring doesn't refer to uninitialized store.
3922 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3923 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3924 fix the problem.
3926 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3927 calls, and wasn't printing the correct value for compiling calls. Increased the
3928 default value of LOOPREPEAT, and the number of significant figures in the
3929 times.
3931 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3933 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3934 a building problem on Windows NT with a FAT file system.
3937 Version 2.01 21-Oct-98
3938 ----------------------
3940 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3941 to character tables built by pcre_maketables() in the current locale. If NULL
3942 is passed, the default tables are used.
3945 Version 2.00 24-Sep-98
3946 ----------------------
3948 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3949 it any more.
3951 2. Allow quantification of (?>) groups, and make it work correctly.
3953 3. The first character computation wasn't working for (?>) groups.
3955 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3956 end of the subject) and add 5.005's \z, which really does match only at the
3957 very end of the subject.
3959 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3961 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3962 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3963 localized options. All options to pcre_study() were also removed.
3965 7. Add other new features from 5.005:
3967 $(?<= positive lookbehind
3968 $(?<! negative lookbehind
3969 (?imsx-imsx) added the unsetting capability
3970 such a setting is global if at outer level; local otherwise
3971 (?imsx-imsx:) non-capturing groups with option setting
3972 (?(cond)re|re) conditional pattern matching
3974 A backreference to itself in a repeated group matches the previous
3975 captured string.
3977 8. General tidying up of studying (both automatic and via "study")
3978 consequential on the addition of new assertions.
3980 9. As in 5.005, unlimited repeated groups that could match an empty substring
3981 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3982 runtime if any iteration does actually match an empty substring.
3984 10. Include the RunTest script in the distribution.
3986 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3987 discrepancies, some of which were old and were also with respect to 5.004. They
3988 have now been fixed.
3991 Version 1.09 28-Apr-98
3992 ----------------------
3994 1. A negated single character class followed by a quantifier with a minimum
3995 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3996 program crashes, or just wrong answers. This did not apply to negated classes
3997 containing more than one character, or to minima other than one.
4000 Version 1.08 27-Mar-98
4001 ----------------------
4003 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
4005 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
4006 latter must appear before anything that relies on it in the pattern.
4009 Version 1.07 16-Feb-98
4010 ----------------------
4012 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
4013 repeat of a potentially empty string).
4016 Version 1.06 23-Jan-98
4017 ----------------------
4019 1. Added Markus Oberhumer's little patches for C++.
4021 2. Literal strings longer than 255 characters were broken.
4024 Version 1.05 23-Dec-97
4025 ----------------------
4027 1. Negated character classes containing more than one character were failing if
4028 PCRE_CASELESS was set at run time.
4031 Version 1.04 19-Dec-97
4032 ----------------------
4034 1. Corrected the man page, where some "const" qualifiers had been omitted.
4036 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4037 input syntax.
4039 3. Fixed memory leak which occurred when a regex with back references was
4040 matched with an offsets vector that wasn't big enough. The temporary memory
4041 that is used in this case wasn't being freed if the match failed.
4043 4. Tidied pcretest to ensure it frees memory that it gets.
4045 5. Temporary memory was being obtained in the case where the passed offsets
4046 vector was exactly big enough.
4048 6. Corrected definition of offsetof() from change 5 below.
4050 7. I had screwed up change 6 below and broken the rules for the use of
4051 setjmp(). Now fixed.
4054 Version 1.03 18-Dec-97
4055 ----------------------
4057 1. A erroneous regex with a missing opening parenthesis was correctly
4058 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4059 on some systems.
4061 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4062 it was reported that one broken compiler failed on the former because "code" is
4063 also an independent variable.
4065 3. The erroneous regex a[]b caused an array overrun reference.
4067 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4068 fail on data ending with that character. (It was going on too far, and checking
4069 the next character, typically a binary zero.) This was specific to the
4070 optimized code for single-character negative classes.
4072 5. Added a contributed patch from the TIN world which does the following:
4074 + Add an undef for memmove, in case the the system defines a macro for it.
4076 + Add a definition of offsetof(), in case there isn't one. (I don't know
4077 the reason behind this - offsetof() is part of the ANSI standard - but
4078 it does no harm).
4080 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4081 most of the places where whitespace preceded '#'. I have given up and
4082 allowed the remaining 2 cases to be at the margin.
4084 + Rename some variables in pcre to eliminate shadowing. This seems very
4085 pedantic, but does no harm, of course.
4087 6. Moved the call to setjmp() into its own function, to get rid of warnings
4088 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4090 7. Constructs such as \d{8,} were compiling into the equivalent of
4091 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4092 outcome, but in this particular case used more store than had been allocated,
4093 which caused the bug to be discovered because it threw up an internal error.
4095 8. The debugging code in both pcre and pcretest for outputting the compiled
4096 form of a regex was going wrong in the case of back references followed by
4097 curly-bracketed repeats.
4100 Version 1.02 12-Dec-97
4101 ----------------------
4103 1. Typos in pcre.3 and comments in the source fixed.
4105 2. Applied a contributed patch to get rid of places where it used to remove
4106 'const' from variables, and fixed some signed/unsigned and uninitialized
4107 variable warnings.
4109 3. Added the "runtest" target to Makefile.
4111 4. Set default compiler flag to -O2 rather than just -O.
4114 Version 1.01 19-Nov-97
4115 ----------------------
4117 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4118 like /([ab]*)*/, that is, for classes with more than one character in them.
4120 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4121 as /((?>a*))*/ (a PCRE_EXTRA facility).
4124 Version 1.00 18-Nov-97
4125 ----------------------
4127 1. Added compile-time macros to support systems such as SunOS4 which don't have
4128 memmove() or strerror() but have other things that can be used instead.
4130 2. Arranged that "make clean" removes the executables.
4133 Version 0.99 27-Oct-97
4134 ----------------------
4136 1. Fixed bug in code for optimizing classes with only one character. It was
4137 initializing a 32-byte map regardless, which could cause it to run off the end
4138 of the memory it had got.
4140 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4143 Version 0.98 22-Oct-97
4144 ----------------------
4146 1. Fixed bug in code for handling temporary memory usage when there are more
4147 back references than supplied space in the ovector. This could cause segfaults.
4150 Version 0.97 21-Oct-97
4151 ----------------------
4153 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4155 2. Optimized negated single characters not to use a bit map.
4157 3. Brought error texts together as macro definitions; clarified some of them;
4158 fixed one that was wrong - it said "range out of order" when it meant "invalid
4159 escape sequence".
4161 4. Changed some char * arguments to const char *.
4163 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4165 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4166 pcretest.
4169 Version 0.96 16-Oct-97
4170 ----------------------
4172 1. Added a simple "pgrep" utility to the distribution.
4174 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4175 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4176 where "ddd" means "one or more decimal digits".
4178 3. Fixed serious bug. If a pattern had a back reference, but the call to
4179 pcre_exec() didn't supply a large enough ovector to record the related
4180 identifying subpattern, the match always failed. PCRE now remembers the number
4181 of the largest back reference, and gets some temporary memory in which to save
4182 the offsets during matching if necessary, in order to ensure that
4183 backreferences always work.
4185 4. Increased the compatibility with Perl in a number of ways:
4187 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4188 to request this handling. The option can be set at compile or exec time.
4190 (b) $ matches before a terminating newline by default; an option
4191 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4192 mode). The option can be set at compile or exec time.
4194 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4195 the same as Perl's. If the decimal number it represents is less than 10
4196 or there aren't that many previous left capturing parentheses, an octal
4197 escape is read. Inside a character class, it's always an octal escape,
4198 even if it is a single digit.
4200 (d) An escaped but undefined alphabetic character is taken as a literal,
4201 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4202 escapes.
4204 (e) {0} is now permitted. (The previous item is removed from the compiled
4205 pattern).
4207 5. Changed all the names of code files so that the basic parts are no longer
4208 than 10 characters, and abolished the teeny "globals.c" file.
4210 6. Changed the handling of character classes; they are now done with a 32-byte
4211 bit map always.
4213 7. Added the -d and /D options to pcretest to make it possible to look at the
4214 internals of compilation without having to recompile pcre.
4217 Version 0.95 23-Sep-97
4218 ----------------------
4220 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4221 \x20 at the start of a run of normal characters. These were being treated as
4222 real characters, instead of the source characters being re-checked.
4225 Version 0.94 18-Sep-97
4226 ----------------------
4228 1. The functions are now thread-safe, with the caveat that the global variables
4229 containing pointers to malloc() and free() or alternative functions are the
4230 same for all threads.
4232 2. Get pcre_study() to generate a bitmap of initial characters for non-
4233 anchored patterns when this is possible, and use it if passed to pcre_exec().
4236 Version 0.93 15-Sep-97
4237 ----------------------
4239 1. /(b)|(:+)/ was computing an incorrect first character.
4241 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4242 but not actually doing anything yet.
4244 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4245 as Perl does (e.g. [-az] or [az-]).
4247 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4248 all possible positions.
4250 5. Split up into different modules to avoid including unneeded functions in a
4251 compiled binary. However, compile and exec are still in one module. The "study"
4252 function is split off.
4254 6. The character tables are now in a separate module whose source is generated
4255 by an auxiliary program - but can then be edited by hand if required. There are
4256 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4257 toupper() in the code.
4259 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4260 make them global. Abolish the function for setting them, as the caller can now
4261 set them directly.
4264 Version 0.92 11-Sep-97
4265 ----------------------
4267 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4268 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4270 2. Caseless matching was not working in character classes if the characters in
4271 the pattern were in upper case.
4273 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4275 4. Make PCRE_ANCHORED public and accept as a compile option.
4277 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4278 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4279 pass them.
4281 6. Give an error if bad option bits passed at compile or run time.
4283 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4284 pcretest to cause it to pass that flag.
4286 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4287 options, and the first character, if set.
4289 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4292 Version 0.91 10-Sep-97
4293 ----------------------
4295 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4296 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4298 2. PCRE was looping on encountering an indefinitely repeated back reference to
4299 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4300 Perl does - treats the match as successful.
4302 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5