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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 613 - (show annotations) (download)
Sat Jul 2 16:59:52 2011 UTC (3 years, 2 months ago) by ph10
File size: 204497 byte(s)
Fix problem with the interaction of (*ACCEPT) in an assertion with 
PCRE_NOTEMPTY.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12