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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 576 - (show annotations)
Sun Nov 21 18:45:10 2010 UTC (4 years, 6 months ago) by ph10
File size: 196203 byte(s)
Error occurred while calculating annotation data.
Added support for (*NO_START_OPT)
1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.11 10-Oct-2010
5 ------------------------
6
7 1. (*THEN) was not working properly if there were untried alternatives prior
8 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
9 backtracked to try for "b" instead of moving to the next alternative branch
10 at the same level (in this case, to look for "c"). The Perl documentation
11 is clear that when (*THEN) is backtracked onto, it goes to the "next
12 alternative in the innermost enclosing group".
13
14 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
15 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
16 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
17 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
18 (*THEN).
19
20 3. If \s appeared in a character class, it removed the VT character from
21 the class, even if it had been included by some previous item, for example
22 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
23 of \s, but is part of the POSIX "space" class.)
24
25 4. A partial match never returns an empty string (because you can always
26 match an empty string at the end of the subject); however the checking for
27 an empty string was starting at the "start of match" point. This has been
28 changed to the "earliest inspected character" point, because the returned
29 data for a partial match starts at this character. This means that, for
30 example, /(?<=abc)def/ gives a partial match for the subject "abc"
31 (previously it gave "no match").
32
33 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
34 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
35 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
36 has an implication that the given string is incomplete (because a partial
37 match is preferred over a full match). For this reason, these items now
38 give a partial match in this situation. [Aside: previously, the one case
39 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
40 match rather than a full match, which was wrong by the old rules, but is
41 now correct.]
42
43 6. There was a bug in the handling of #-introduced comments, recognized when
44 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
45 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
46 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
47 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
48 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
49 places in pcre_compile().
50
51 7. Related to (6) above, when pcre_compile() was skipping #-introduced
52 comments when looking ahead for named forward references to subpatterns,
53 the only newline sequence it recognized was NL. It now handles newlines
54 according to the set newline convention.
55
56 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
57 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
58 cater for a lack of strerror(). These oversights have been fixed.
59
60 9. Added --match-limit and --recursion-limit to pcregrep.
61
62 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
63
64 11. When the -o option was used, pcregrep was setting a return code of 1, even
65 when matches were found, and --line-buffered was not being honoured.
66
67 12. Added an optional parentheses number to the -o and --only-matching options
68 of pcregrep.
69
70 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
71 can match an empty string. The code to do it in pcretest and pcredemo
72 needed fixing:
73
74 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
75 only one byte after an empty string match just before CRLF (this case
76 just got forgotten; "any" and "anycrlf" were OK).
77
78 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
79 mode when an empty string match preceded an ASCII character followed by
80 a non-ASCII character. (The code for advancing by one character rather
81 than one byte was nonsense.)
82
83 (c) The pcredemo.c sample program did not have any code at all to handle
84 the cases when CRLF is a valid newline sequence.
85
86 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
87 as a starting offset was within the subject string. There is now a new
88 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
89 negative or greater than the length of the string. In order to test this,
90 pcretest is extended to allow the setting of negative starting offsets.
91
92 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
93 starting offset points to the beginning of a UTF-8 character was
94 unnecessarily clumsy. I tidied it up.
95
96 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
97 bad UTF-8 sequence and one that is incomplete.
98
99 17. Nobody had reported that the --include_dir option, which was added in
100 release 7.7 should have been called --include-dir (hyphen, not underscore)
101 for compatibility with GNU grep. I have changed it to --include-dir, but
102 left --include_dir as an undocumented synonym, and the same for
103 --exclude-dir, though that is not available in GNU grep, at least as of
104 release 2.5.4.
105
106 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
107 characters from a string of bytes) have been redefined so as not to use
108 loops, in order to improve performance in some environments. At the same
109 time, I abstracted some of the common code into auxiliary macros to save
110 repetition (this should not affect the compiled code).
111
112 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
113 compile-time error is now given if \c is not followed by an ASCII
114 character, that is, a byte less than 128. (In EBCDIC mode, the code is
115 different, and any byte value is allowed.)
116
117 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
118 START_OPTIMIZE option, which is now allowed at compile time - but just
119 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
120 to pcregrep and other applications that have no direct access to PCRE
121 options. The new /Y option in pcretest sets this option when calling
122 pcre_compile().
123
124
125 Version 8.10 25-Jun-2010
126 ------------------------
127
128 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
129 THEN.
130
131 2. (*ACCEPT) was not working when inside an atomic group.
132
133 3. Inside a character class, \B is treated as a literal by default, but
134 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
135 causes the error). The code is unchanged, but I tidied the documentation.
136
137 4. Inside a character class, PCRE always treated \R and \X as literals,
138 whereas Perl faults them if its -w option is set. I have changed PCRE so
139 that it faults them when PCRE_EXTRA is set.
140
141 5. Added support for \N, which always matches any character other than
142 newline. (It is the same as "." when PCRE_DOTALL is not set.)
143
144 6. When compiling pcregrep with newer versions of gcc which may have
145 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
146 declared with attribute warn_unused_result" were given. Just casting the
147 result to (void) does not stop the warnings; a more elaborate fudge is
148 needed. I've used a macro to implement this.
149
150 7. Minor change to pcretest.c to avoid a compiler warning.
151
152 8. Added four artifical Unicode properties to help with an option to make
153 \s etc use properties (see next item). The new properties are: Xan
154 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
155
156 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
157 use Unicode properties. (*UCP) at the start of a pattern can be used to set
158 this option. Modified pcretest to add /W to test this facility. Added
159 REG_UCP to make it available via the POSIX interface.
160
161 10. Added --line-buffered to pcregrep.
162
163 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
164 studied, and the match started with a letter with a code point greater than
165 127 whose first byte was different to the first byte of the other case of
166 the letter, the other case of this starting letter was not recognized
167 (#976).
168
169 12. If a pattern that was studied started with a repeated Unicode property
170 test, for example, \p{Nd}+, there was the theoretical possibility of
171 setting up an incorrect bitmap of starting bytes, but fortunately it could
172 not have actually happened in practice until change 8 above was made (it
173 added property types that matched character-matching opcodes).
174
175 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
176 possible starting bytes for non-anchored patterns.
177
178 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
179 \R, and also a number of cases that involve Unicode properties, both
180 explicit and implicit when PCRE_UCP is set.
181
182 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
183 input, it could crash or give wrong results if characters with values
184 greater than 0xc0 were present in the subject string. (Detail: it assumed
185 UTF-8 input when processing these items.)
186
187 16. Added a lot of (int) casts to avoid compiler warnings in systems where
188 size_t is 64-bit (#991).
189
190 17. Added a check for running out of memory when PCRE is compiled with
191 --disable-stack-for-recursion (#990).
192
193 18. If the last data line in a file for pcretest does not have a newline on
194 the end, a newline was missing in the output.
195
196 19. The default pcre_chartables.c file recognizes only ASCII characters (values
197 less than 128) in its various bitmaps. However, there is a facility for
198 generating tables according to the current locale when PCRE is compiled. It
199 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
200 characters, are recognized by isspace() and therefore were getting set in
201 these tables, and indeed these tables seem to approximate to ISO 8859. This
202 caused a problem in UTF-8 mode when pcre_study() was used to create a list
203 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
204 which of course cannot start UTF-8 characters. I have changed the code so
205 that only real ASCII characters (less than 128) and the correct starting
206 bytes for UTF-8 encodings are set for characters greater than 127 when in
207 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
208 altogether.)
209
210 20. Added the /T option to pcretest so as to be able to run tests with non-
211 standard character tables, thus making it possible to include the tests
212 used for 19 above in the standard set of tests.
213
214 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
215 reference to a subpattern the other side of a comment that contains an
216 opening parenthesis caused either an internal compiling error, or a
217 reference to the wrong subpattern.
218
219
220 Version 8.02 19-Mar-2010
221 ------------------------
222
223 1. The Unicode data tables have been updated to Unicode 5.2.0.
224
225 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
226 configured.
227
228 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
229 original author of that file, following a query about its status.
230
231 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
232 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
233
234 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
235 quantifier applied to a forward-referencing subroutine call, could compile
236 incorrect code or give the error "internal error: previously-checked
237 referenced subpattern not found".
238
239 6. Both MS Visual Studio and Symbian OS have problems with initializing
240 variables to point to external functions. For these systems, therefore,
241 pcre_malloc etc. are now initialized to local functions that call the
242 relevant global functions.
243
244 7. There were two entries missing in the vectors called coptable and poptable
245 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
246 I've fixed the data, and added a kludgy way of testing at compile time that
247 the lengths are correct (equal to the number of opcodes).
248
249 8. Following on from 7, I added a similar kludge to check the length of the
250 eint vector in pcreposix.c.
251
252 9. Error texts for pcre_compile() are held as one long string to avoid too
253 much relocation at load time. To find a text, the string is searched,
254 counting zeros. There was no check for running off the end of the string,
255 which could happen if a new error number was added without updating the
256 string.
257
258 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
259
260 11. \K was not working if it appeared in an atomic group or in a group that
261 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
262 \K is "not well defined" if used in an assertion. PCRE now accepts it if
263 the assertion is positive, but not if it is negative.
264
265 12. Change 11 fortuitously reduced the size of the stack frame used in the
266 "match()" function of pcre_exec.c by one pointer. Forthcoming
267 implementation of support for (*MARK) will need an extra pointer on the
268 stack; I have reserved it now, so that the stack frame size does not
269 decrease.
270
271 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
272 item in branch that calls a recursion is a subroutine call - as in the
273 second branch in the above example - was incorrectly given the compile-
274 time error "recursive call could loop indefinitely" because pcre_compile()
275 was not correctly checking the subroutine for matching a non-empty string.
276
277 14. The checks for overrunning compiling workspace could trigger after an
278 overrun had occurred. This is a "should never occur" error, but it can be
279 triggered by pathological patterns such as hundreds of nested parentheses.
280 The checks now trigger 100 bytes before the end of the workspace.
281
282 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
283
284
285 Version 8.01 19-Jan-2010
286 ------------------------
287
288 1. If a pattern contained a conditional subpattern with only one branch (in
289 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
290 computed the wrong minimum data length (which is of course zero for such
291 subpatterns). This could cause incorrect "no match" results.
292
293 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
294 the pattern is reset in the first branch, pcre_compile() failed with
295 "internal error: code overflow at offset...". This happened only when
296 the reset was to the original external option setting. (An optimization
297 abstracts leading options settings into an external setting, which was the
298 cause of this.)
299
300 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
301 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
302 assertion pattern did not match (meaning that the assertion was true), it
303 was incorrectly treated as false if the SKIP had been reached during the
304 matching. This also applied to assertions used as conditions.
305
306 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
307 assertion subpattern, including such a pattern used as a condition,
308 unpredictable results occurred, instead of the error return
309 PCRE_ERROR_DFA_UITEM.
310
311 5. The C++ GlobalReplace function was not working like Perl for the special
312 situation when an empty string is matched. It now does the fancy magic
313 stuff that is necessary.
314
315 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
316 removed. (These were left over from very, very early versions of PCRE.)
317
318 7. Some cosmetic changes to the code to make life easier when compiling it
319 as part of something else:
320
321 (a) Change DEBUG to PCRE_DEBUG.
322
323 (b) In pcre_compile(), rename the member of the "branch_chain" structure
324 called "current" as "current_branch", to prevent a collision with the
325 Linux macro when compiled as a kernel module.
326
327 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
328 prevent a collision with the Linux macro when compiled as a kernel
329 module.
330
331 8. In pcre_compile() there are some checks for integer overflows that used to
332 cast potentially large values to (double). This has been changed to that
333 when building, a check for int64_t is made, and if it is found, it is used
334 instead, thus avoiding the use of floating point arithmetic. (There is no
335 other use of FP in PCRE.) If int64_t is not found, the fallback is to
336 double.
337
338 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
339 2005 (difference between two addresses compared to an unsigned value).
340
341 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
342 custom one, because of the following reported problem in Windows:
343
344 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
345 under Win32.
346 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
347 therefore missing the function definition.
348 - The compiler thus generates a "C" signature for the test function.
349 - The linker fails to find the "C" function.
350 - PCRE fails to configure if asked to do so against libbz2.
351
352 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
353 messages were output:
354
355 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
356 rerunning libtoolize, to keep the correct libtool macros in-tree.
357 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
358
359 I have done both of these things.
360
361 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
362 most of the time, it *can* run out if it is given a pattern that contains a
363 runaway infinite recursion. I updated the discussion in the pcrestack man
364 page.
365
366 13. Now that we have gone to the x.xx style of version numbers, the minor
367 version may start with zero. Using 08 or 09 is a bad idea because users
368 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
369 interpreted as invalid octal numbers. I've updated the previous comment in
370 configure.ac, and also added a check that gives an error if 08 or 09 are
371 used.
372
373 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
374 causing partial matching to fail when the end of the subject matched \W
375 in a UTF-8 pattern where \W was quantified with a minimum of 3.
376
377 15. There were some discrepancies between the declarations in pcre_internal.h
378 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
379 their definitions. The declarations used "const uschar *" and the
380 definitions used USPTR. Even though USPTR is normally defined as "const
381 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
382 reported that: "This difference in casting confuses some C++ compilers, for
383 example, SunCC recognizes above declarations as different functions and
384 generates broken code for hbpcre." I have changed the declarations to use
385 USPTR.
386
387 16. GNU libtool is named differently on some systems. The autogen.sh script now
388 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
389 (FreeBSD).
390
391 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
392 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
393 comment: "Figure out how to create a longlong from a string: strtoll and
394 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
395 instance, but it only takes 2 args instead of 3!"
396
397 18. A subtle bug concerned with back references has been fixed by a change of
398 specification, with a corresponding code fix. A pattern such as
399 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
400 refers, was giving matches when it shouldn't. For example, xa=xaaa would
401 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
402 same bug. Such groups have to be quantified to be useful, or contained
403 inside another quantified group. (If there's no repetition, the reference
404 can never match.) The problem arises because, having left the group and
405 moved on to the rest of the pattern, a later failure that backtracks into
406 the group uses the captured value from the final iteration of the group
407 rather than the correct earlier one. I have fixed this in PCRE by forcing
408 any group that contains a reference to itself to be an atomic group; that
409 is, there cannot be any backtracking into it once it has completed. This is
410 similar to recursive and subroutine calls.
411
412
413 Version 8.00 19-Oct-09
414 ----------------------
415
416 1. The table for translating pcre_compile() error codes into POSIX error codes
417 was out-of-date, and there was no check on the pcre_compile() error code
418 being within the table. This could lead to an OK return being given in
419 error.
420
421 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
422 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
423 in a Windows environment.
424
425 3. The pcregrep --count option prints the count for each file even when it is
426 zero, as does GNU grep. However, pcregrep was also printing all files when
427 --files-with-matches was added. Now, when both options are given, it prints
428 counts only for those files that have at least one match. (GNU grep just
429 prints the file name in this circumstance, but including the count seems
430 more useful - otherwise, why use --count?) Also ensured that the
431 combination -clh just lists non-zero counts, with no names.
432
433 4. The long form of the pcregrep -F option was incorrectly implemented as
434 --fixed_strings instead of --fixed-strings. This is an incompatible change,
435 but it seems right to fix it, and I didn't think it was worth preserving
436 the old behaviour.
437
438 5. The command line items --regex=pattern and --regexp=pattern were not
439 recognized by pcregrep, which required --regex pattern or --regexp pattern
440 (with a space rather than an '='). The man page documented the '=' forms,
441 which are compatible with GNU grep; these now work.
442
443 6. No libpcreposix.pc file was created for pkg-config; there was just
444 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
445
446 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
447 when UCP support is not needed, by modifying the Python script that
448 generates it from Unicode data files. This should not matter if the module
449 is correctly used as a library, but I received one complaint about 50K of
450 unwanted data. My guess is that the person linked everything into his
451 program rather than using a library. Anyway, it does no harm.
452
453 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
454 was a minimum greater than 1 for a wide character in a possessive
455 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
456 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
457 character. Chaos in the form of incorrect output or a compiling loop could
458 result.
459
460 9. The restrictions on what a pattern can contain when partial matching is
461 requested for pcre_exec() have been removed. All patterns can now be
462 partially matched by this function. In addition, if there are at least two
463 slots in the offset vector, the offset of the earliest inspected character
464 for the match and the offset of the end of the subject are set in them when
465 PCRE_ERROR_PARTIAL is returned.
466
467 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
468 synonymous with PCRE_PARTIAL, for backwards compatibility, and
469 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
470 and may be more useful for multi-segment matching.
471
472 11. Partial matching with pcre_exec() is now more intuitive. A partial match
473 used to be given if ever the end of the subject was reached; now it is
474 given only if matching could not proceed because another character was
475 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
476 string "Z", which now yields "no match" instead of "partial match". In the
477 case of pcre_dfa_exec(), "no match" is given if every matching path for the
478 final character ended with (*FAIL).
479
480 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
481 if the pattern had a "must contain" character that was already found in the
482 earlier partial match, unless partial matching was again requested. For
483 example, with the pattern /dog.(body)?/, the "must contain" character is
484 "g". If the first part-match was for the string "dog", restarting with
485 "sbody" failed. This bug has been fixed.
486
487 13. The string returned by pcre_dfa_exec() after a partial match has been
488 changed so that it starts at the first inspected character rather than the
489 first character of the match. This makes a difference only if the pattern
490 starts with a lookbehind assertion or \b or \B (\K is not supported by
491 pcre_dfa_exec()). It's an incompatible change, but it makes the two
492 matching functions compatible, and I think it's the right thing to do.
493
494 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
495 so that the demonstration program is easily available in environments where
496 PCRE has not been installed from source.
497
498 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
499 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
500 library.
501
502 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
503 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
504 is not the first non-POSIX option to be added. Clearly some people find
505 these options useful.
506
507 17. If a caller to the POSIX matching function regexec() passes a non-zero
508 value for nmatch with a NULL value for pmatch, the value of
509 nmatch is forced to zero.
510
511 18. RunGrepTest did not have a test for the availability of the -u option of
512 the diff command, as RunTest does. It now checks in the same way as
513 RunTest, and also checks for the -b option.
514
515 19. If an odd number of negated classes containing just a single character
516 interposed, within parentheses, between a forward reference to a named
517 subpattern and the definition of the subpattern, compilation crashed with
518 an internal error, complaining that it could not find the referenced
519 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
520 [The bug was that it was starting one character too far in when skipping
521 over the character class, thus treating the ] as data rather than
522 terminating the class. This meant it could skip too much.]
523
524 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
525 /g option in pcretest when the pattern contains \K, which makes it possible
526 to have an empty string match not at the start, even when the pattern is
527 anchored. Updated pcretest and pcredemo to use this option.
528
529 21. If the maximum number of capturing subpatterns in a recursion was greater
530 than the maximum at the outer level, the higher number was returned, but
531 with unset values at the outer level. The correct (outer level) value is
532 now given.
533
534 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
535 PCRE did not set those parentheses (unlike Perl). I have now found a way to
536 make it do so. The string so far is captured, making this feature
537 compatible with Perl.
538
539 23. The tests have been re-organized, adding tests 11 and 12, to make it
540 possible to check the Perl 5.10 features against Perl 5.10.
541
542 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
543 pattern matches a fixed length string. PCRE did not allow this; now it
544 does. Neither allows recursion.
545
546 25. I finally figured out how to implement a request to provide the minimum
547 length of subject string that was needed in order to match a given pattern.
548 (It was back references and recursion that I had previously got hung up
549 on.) This code has now been added to pcre_study(); it finds a lower bound
550 to the length of subject needed. It is not necessarily the greatest lower
551 bound, but using it to avoid searching strings that are too short does give
552 some useful speed-ups. The value is available to calling programs via
553 pcre_fullinfo().
554
555 26. While implementing 25, I discovered to my embarrassment that pcretest had
556 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
557 study optimizations had never been tested with that matching function.
558 Oops. What is worse, even when it was passed study data, there was a bug in
559 pcre_dfa_exec() that meant it never actually used it. Double oops. There
560 were also very few tests of studied patterns with pcre_dfa_exec().
561
562 27. If (?| is used to create subpatterns with duplicate numbers, they are now
563 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
564 on the other side of the coin, they are no longer allowed to have different
565 names, because these cannot be distinguished in PCRE, and this has caused
566 confusion. (This is a difference from Perl.)
567
568 28. When duplicate subpattern names are present (necessarily with different
569 numbers, as required by 27 above), and a test is made by name in a
570 conditional pattern, either for a subpattern having been matched, or for
571 recursion in such a pattern, all the associated numbered subpatterns are
572 tested, and the overall condition is true if the condition is true for any
573 one of them. This is the way Perl works, and is also more like the way
574 testing by number works.
575
576
577 Version 7.9 11-Apr-09
578 ---------------------
579
580 1. When building with support for bzlib/zlib (pcregrep) and/or readline
581 (pcretest), all targets were linked against these libraries. This included
582 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
583 libraries. This caused unwanted dependencies to be created. This problem
584 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
585 pcretest is linked with readline.
586
587 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
588 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
589 moved outside it again, because FALSE and TRUE are already defined in AIX,
590 but BOOL is not.
591
592 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
593 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
594
595 4. The pcregrep documentation said spaces were inserted as well as colons (or
596 hyphens) following file names and line numbers when outputting matching
597 lines. This is not true; no spaces are inserted. I have also clarified the
598 wording for the --colour (or --color) option.
599
600 5. In pcregrep, when --colour was used with -o, the list of matching strings
601 was not coloured; this is different to GNU grep, so I have changed it to be
602 the same.
603
604 6. When --colo(u)r was used in pcregrep, only the first matching substring in
605 each matching line was coloured. Now it goes on to look for further matches
606 of any of the test patterns, which is the same behaviour as GNU grep.
607
608 7. A pattern that could match an empty string could cause pcregrep to loop; it
609 doesn't make sense to accept an empty string match in pcregrep, so I have
610 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
611 seems to be how GNU grep behaves.
612
613 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
614 start or after a newline", because the conditional assertion was not being
615 correctly handled. The rule now is that both the assertion and what follows
616 in the first alternative must satisfy the test.
617
618 9. If auto-callout was enabled in a pattern with a conditional group whose
619 condition was an assertion, PCRE could crash during matching, both with
620 pcre_exec() and pcre_dfa_exec().
621
622 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
623 used for matching.
624
625 11. Unicode property support in character classes was not working for
626 characters (bytes) greater than 127 when not in UTF-8 mode.
627
628 12. Added the -M command line option to pcretest.
629
630 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
631
632 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
633
634 16. Added comments and documentation about mis-use of no_arg in the C++
635 wrapper.
636
637 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
638 from Martin Jerabek that uses macro names for all relevant character and
639 string constants.
640
641 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
642 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
643 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
644 these, but not everybody uses configure.
645
646 19. A conditional group that had only one branch was not being correctly
647 recognized as an item that could match an empty string. This meant that an
648 enclosing group might also not be so recognized, causing infinite looping
649 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
650 with the subject "ab", where knowledge that the repeated group can match
651 nothing is needed in order to break the loop.
652
653 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
654 exec(), but without supplying a callout function, matching went wrong.
655
656 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
657 leak if the size of the offset vector was greater than 30. When the vector
658 is smaller, the saved offsets during recursion go onto a local stack
659 vector, but for larger vectors malloc() is used. It was failing to free
660 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
661 error, in fact).
662
663 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
664 heapframe that is used only when UTF-8 support is enabled. This caused no
665 problem, but was untidy.
666
667 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
668 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
669 included within another project.
670
671 24. Steven Van Ingelgem's patches to add more options to the CMake support,
672 slightly modified by me:
673
674 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
675 not building pcregrep.
676
677 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
678 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
679
680 25. Forward references, both numeric and by name, in patterns that made use of
681 duplicate group numbers, could behave incorrectly or give incorrect errors,
682 because when scanning forward to find the reference group, PCRE was not
683 taking into account the duplicate group numbers. A pattern such as
684 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
685
686 26. Changed a few more instances of "const unsigned char *" to USPTR, making
687 the feature of a custom pointer more persuasive (as requested by a user).
688
689 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
690 pcretest.c, inside #ifndefs, because it seems they are sometimes already
691 pre-defined.
692
693 28. Added support for (*UTF8) at the start of a pattern.
694
695 29. Arrange for flags added by the "release type" setting in CMake to be shown
696 in the configuration summary.
697
698
699 Version 7.8 05-Sep-08
700 ---------------------
701
702 1. Replaced UCP searching code with optimized version as implemented for Ad
703 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
704 stage table and inline lookup instead of a function, giving speed ups of 2
705 to 5 times on some simple patterns that I tested. Permission was given to
706 distribute the MultiStage2.py script that generates the tables (it's not in
707 the tarball, but is in the Subversion repository).
708
709 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
710 scripts.
711
712 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
713 a group with a zero qualifier. The result of the study could be incorrect,
714 or the function might crash, depending on the pattern.
715
716 4. Caseless matching was not working for non-ASCII characters in back
717 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
718 It now works when Unicode Property Support is available.
719
720 5. In pcretest, an escape such as \x{de} in the data was always generating
721 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
722 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
723 truncation.
724
725 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
726
727 7. Added two (int) casts to pcregrep when printing the difference of two
728 pointers, in case they are 64-bit values.
729
730 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
731 test 2 if it fails.
732
733 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
734 and a #define of that name to empty if it is not externally set. This is to
735 allow users of MSVC to set it if necessary.
736
737 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
738 the convenience functions in the pcre_get.c source file.
739
740 11. An option change at the start of a pattern that had top-level alternatives
741 could cause overwriting and/or a crash. This command provoked a crash in
742 some environments:
743
744 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
745
746 This potential security problem was recorded as CVE-2008-2371.
747
748 12. For a pattern where the match had to start at the beginning or immediately
749 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
750 pcre_dfa_exec() could read past the end of the passed subject if there was
751 no match. To help with detecting such bugs (e.g. with valgrind), I modified
752 pcretest so that it places the subject at the end of its malloc-ed buffer.
753
754 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
755 exec() might read past the end of the data buffer in UTF-8 mode.
756
757 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
758 the data contained the byte 0x85 as part of a UTF-8 character within its
759 first line. This applied both to normal and DFA matching.
760
761 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
762 /^[^d]*?$/8 failed to match "abc".
763
764 16. Added a missing copyright notice to pcrecpp_internal.h.
765
766 17. Make it more clear in the documentation that values returned from
767 pcre_exec() in ovector are byte offsets, not character counts.
768
769 18. Tidied a few places to stop certain compilers from issuing warnings.
770
771 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
772 supplied by Stefan Weber. I made a further small update for 7.8 because
773 there is a change of source arrangements: the pcre_searchfuncs.c module is
774 replaced by pcre_ucd.c.
775
776
777 Version 7.7 07-May-08
778 ---------------------
779
780 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
781 a string to a long long, pretend we don't even have a long long." This is
782 done by checking for the strtoq, strtoll, and _strtoi64 functions.
783
784 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
785 pre-7.6 versions, which defined a global no_arg variable instead of putting
786 it in the RE class. (See also #8 below.)
787
788 3. Remove a line of dead code, identified by coverity and reported by Nuno
789 Lopes.
790
791 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
792
793 (1) The include/exclude patterns were being applied to the whole pathnames
794 of files, instead of just to the final components.
795
796 (2) If there was more than one level of directory, the subdirectories were
797 skipped unless they satisfied the include/exclude conditions. This is
798 inconsistent with GNU grep (and could even be seen as contrary to the
799 pcregrep specification - which I improved to make it absolutely clear).
800 The action now is always to scan all levels of directory, and just
801 apply the include/exclude patterns to regular files.
802
803 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
804 --exclude_dir in the tests to avoid scanning .svn directories.
805
806 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
807 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
808 doesn't support NULs in patterns.
809
810 7. Added some missing "const"s to declarations of static tables in
811 pcre_compile.c and pcre_dfa_exec.c.
812
813 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
814 caused by fix #2 above. (Subsequently also a second patch to fix the
815 first patch. And a third patch - this was a messy problem.)
816
817 9. Applied Craig's patch to remove the use of push_back().
818
819 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
820 matching function regexec().
821
822 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
823 which, however, unlike Perl's \g{...}, are subroutine calls, not back
824 references. PCRE supports relative numbers with this syntax (I don't think
825 Oniguruma does).
826
827 12. Previously, a group with a zero repeat such as (...){0} was completely
828 omitted from the compiled regex. However, this means that if the group
829 was called as a subroutine from elsewhere in the pattern, things went wrong
830 (an internal error was given). Such groups are now left in the compiled
831 pattern, with a new opcode that causes them to be skipped at execution
832 time.
833
834 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
835 to the way PCRE behaves:
836
837 (a) A lone ] character is dis-allowed (Perl treats it as data).
838
839 (b) A back reference to an unmatched subpattern matches an empty string
840 (Perl fails the current match path).
841
842 (c) A data ] in a character class must be notated as \] because if the
843 first data character in a class is ], it defines an empty class. (In
844 Perl it is not possible to have an empty class.) The empty class []
845 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
846 The negative empty class [^] matches any one character, independently
847 of the DOTALL setting.
848
849 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
850 non-existent subpattern following a character class starting with ']' and
851 containing () gave an internal compiling error instead of "reference to
852 non-existent subpattern". Fortunately, when the pattern did exist, the
853 compiled code was correct. (When scanning forwards to check for the
854 existencd of the subpattern, it was treating the data ']' as terminating
855 the class, so got the count wrong. When actually compiling, the reference
856 was subsequently set up correctly.)
857
858 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
859 it was being rejected as not supported by pcre_dfa_exec(), even though
860 other assertions are supported. I have made pcre_dfa_exec() support
861 (*FAIL).
862
863 16. The implementation of 13c above involved the invention of a new opcode,
864 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
865 cannot be changed at match time, I realized I could make a small
866 improvement to matching performance by compiling OP_ALLANY instead of
867 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
868 on the OP_ANY path.
869
870 17. Compiling pcretest on Windows with readline support failed without the
871 following two fixes: (1) Make the unistd.h include conditional on
872 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
873
874 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
875 ncurses library to be included for pcretest when ReadLine support is
876 requested, but also to allow for it to be overridden. This patch came from
877 Daniel Bergström.
878
879 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
880 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
881 any errors with the current Unicode tables. Thanks to Peter Kankowski for
882 spotting this.
883
884
885 Version 7.6 28-Jan-08
886 ---------------------
887
888 1. A character class containing a very large number of characters with
889 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
890 overflow.
891
892 2. Patch to cut out the "long long" test in pcrecpp_unittest when
893 HAVE_LONG_LONG is not defined.
894
895 3. Applied Christian Ehrlicher's patch to update the CMake build files to
896 bring them up to date and include new features. This patch includes:
897
898 - Fixed PH's badly added libz and libbz2 support.
899 - Fixed a problem with static linking.
900 - Added pcredemo. [But later removed - see 7 below.]
901 - Fixed dftables problem and added an option.
902 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
903 HAVE_LONG_LONG.
904 - Added readline support for pcretest.
905 - Added an listing of the option settings after cmake has run.
906
907 4. A user submitted a patch to Makefile that makes it easy to create
908 "pcre.dll" under mingw when using Configure/Make. I added stuff to
909 Makefile.am that cause it to include this special target, without
910 affecting anything else. Note that the same mingw target plus all
911 the other distribution libraries and programs are now supported
912 when configuring with CMake (see 6 below) instead of with
913 Configure/Make.
914
915 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
916 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
917 exported in the Windows port". It has not yet been confirmed that the patch
918 solves the problem, but it does no harm.
919
920 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
921 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
922 with CMake, and also correct the comment about stack recursion.
923
924 7. Remove the automatic building of pcredemo from the ./configure system and
925 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
926 of a program that users should build themselves after PCRE is installed, so
927 building it automatically is not really right. What is more, it gave
928 trouble in some build environments.
929
930 8. Further tidies to CMakeLists.txt from Sheri and Christian.
931
932
933 Version 7.5 10-Jan-08
934 ---------------------
935
936 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
937 values in parens when parsing an RE using the C++ wrapper."
938
939 2. Negative specials like \S did not work in character classes in UTF-8 mode.
940 Characters greater than 255 were excluded from the class instead of being
941 included.
942
943 3. The same bug as (2) above applied to negated POSIX classes such as
944 [:^space:].
945
946 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
947 defined or documented. It seems to have been a typo for PCRE_STATIC, so
948 I have changed it.
949
950 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
951 first named subpattern) and a construct such as (?&a) would reference the
952 first named subpattern whose name started with "a" (in other words, the
953 length check was missing). Both these problems are fixed. "Subpattern name
954 expected" is now given for (?&) (a zero-length name), and this patch also
955 makes it give the same error for \k'' (previously it complained that that
956 was a reference to a non-existent subpattern).
957
958 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
959 this is right because (?- can be followed by option settings as well as by
960 digits. I have, however, made the messages clearer.
961
962 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
963 than the number used in the conditional) now cause a compile-time error.
964 This is actually not compatible with Perl, which accepts such patterns, but
965 treats the conditional as always being FALSE (as PCRE used to), but it
966 seems to me that giving a diagnostic is better.
967
968 8. Change "alphameric" to the more common word "alphanumeric" in comments
969 and messages.
970
971 9. Fix two occurrences of "backslash" in comments that should have been
972 "backspace".
973
974 10. Remove two redundant lines of code that can never be obeyed (their function
975 was moved elsewhere).
976
977 11. The program that makes PCRE's Unicode character property table had a bug
978 which caused it to generate incorrect table entries for sequences of
979 characters that have the same character type, but are in different scripts.
980 It amalgamated them into a single range, with the script of the first of
981 them. In other words, some characters were in the wrong script. There were
982 thirteen such cases, affecting characters in the following ranges:
983
984 U+002b0 - U+002c1
985 U+0060c - U+0060d
986 U+0061e - U+00612
987 U+0064b - U+0065e
988 U+0074d - U+0076d
989 U+01800 - U+01805
990 U+01d00 - U+01d77
991 U+01d9b - U+01dbf
992 U+0200b - U+0200f
993 U+030fc - U+030fe
994 U+03260 - U+0327f
995 U+0fb46 - U+0fbb1
996 U+10450 - U+1049d
997
998 12. The -o option (show only the matching part of a line) for pcregrep was not
999 compatible with GNU grep in that, if there was more than one match in a
1000 line, it showed only the first of them. It now behaves in the same way as
1001 GNU grep.
1002
1003 13. If the -o and -v options were combined for pcregrep, it printed a blank
1004 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1005 does the same. The return code can be used to tell if there were any
1006 non-matching lines.
1007
1008 14. Added --file-offsets and --line-offsets to pcregrep.
1009
1010 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1011 infinitely looping recursion. The bug was that positive lookaheads were not
1012 being skipped when checking for a possible empty match (negative lookaheads
1013 and both kinds of lookbehind were skipped).
1014
1015 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1016 inclusion of <windows.h> to before rather than after the definition of
1017 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1018
1019 17. Specifying a possessive quantifier with a specific limit for a Unicode
1020 character property caused pcre_compile() to compile bad code, which led at
1021 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1022 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1023 caused the error; without that there was no problem.
1024
1025 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1026
1027 19. Added --enable-pcretest-libreadline.
1028
1029 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1030 RE::GlobalReplace(). As a result, the number of replacements returned was
1031 double what it should be. I removed one of the increments, but Craig sent a
1032 later patch that removed the other one (the right fix) and added unit tests
1033 that check the return values (which was not done before).
1034
1035 21. Several CMake things:
1036
1037 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1038 the names libpcre and libpcreposix, not just pcre and pcreposix.
1039
1040 (2) The above change means that pcretest and pcregrep are now correctly
1041 linked with the newly-built libraries, not previously installed ones.
1042
1043 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1044
1045 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1046 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1047 UTF-8 newline character). The key issue is that the pattern starts .*;
1048 this means that the match must be either at the beginning, or after a
1049 newline. The bug was in the code for advancing after a failed match and
1050 checking that the new position followed a newline. It was not taking
1051 account of UTF-8 characters correctly.
1052
1053 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1054 character classes. PCRE was not treating the sequence [:...:] as a
1055 character class unless the ... were all letters. Perl, however, seems to
1056 allow any characters between [: and :], though of course it rejects as
1057 unknown any "names" that contain non-letters, because all the known class
1058 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1059 for example, whereas PCRE did not - it did not recognize a POSIX character
1060 class. This seemed a bit dangerous, so the code has been changed to be
1061 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1062 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1063 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1064 Perl does, and where it didn't before.
1065
1066 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1067 Windows environments %n is disabled by default.
1068
1069
1070 Version 7.4 21-Sep-07
1071 ---------------------
1072
1073 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1074 means that a class such as [\s] counted as "explicit reference to CR or
1075 LF". That isn't really right - the whole point of the change was to try to
1076 help when there was an actual mention of one of the two characters. So now
1077 the change happens only if \r or \n (or a literal CR or LF) character is
1078 encountered.
1079
1080 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1081 of both had grown to the point where there were only 3 bits left.
1082 Fortunately, there was spare space in the data structure, and so I have
1083 moved the internal flags into a new 16-bit field to free up more option
1084 bits.
1085
1086 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1087 but did not set the internal JCHANGED flag - either of these is enough to
1088 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1089 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1090 start sets both bits.
1091
1092 4. Added options (at build time, compile time, exec time) to change \R from
1093 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1094
1095 5. doc/pcresyntax.html was missing from the distribution.
1096
1097 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1098 compatibility, even though it is no longer used.
1099
1100 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1101 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1102 windows.h file is present (where different names are used). [This was
1103 reversed later after testing - see 16 below.]
1104
1105 8. Changed all #include <config.h> to #include "config.h". There were also
1106 some further <pcre.h> cases that I changed to "pcre.h".
1107
1108 9. When pcregrep was used with the --colour option, it missed the line ending
1109 sequence off the lines that it output.
1110
1111 10. It was pointed out to me that arrays of string pointers cause lots of
1112 relocations when a shared library is dynamically loaded. A technique of
1113 using a single long string with a table of offsets can drastically reduce
1114 these. I have refactored PCRE in four places to do this. The result is
1115 dramatic:
1116
1117 Originally: 290
1118 After changing UCP table: 187
1119 After changing error message table: 43
1120 After changing table of "verbs" 36
1121 After changing table of Posix names 22
1122
1123 Thanks to the folks working on Gregex for glib for this insight.
1124
1125 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1126 unicode-properties was also set.
1127
1128 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1129
1130 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1131 checked only for CRLF.
1132
1133 14. Added casts to pcretest.c to avoid compiler warnings.
1134
1135 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1136
1137 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1138 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1139 entirely. This removes changes made in 7 above.
1140
1141 17. The CMake files have been updated, and there is now more information about
1142 building with CMake in the NON-UNIX-USE document.
1143
1144
1145 Version 7.3 28-Aug-07
1146 ---------------------
1147
1148 1. In the rejigging of the build system that eventually resulted in 7.1, the
1149 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1150 brackets there is not right, since it causes compilers to look for an
1151 installed pcre.h, not the version that is in the source that is being
1152 compiled (which of course may be different). I have changed it back to:
1153
1154 #include "pcre.h"
1155
1156 I have a vague recollection that the change was concerned with compiling in
1157 different directories, but in the new build system, that is taken care of
1158 by the VPATH setting the Makefile.
1159
1160 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1161 when the subject happened to end in the byte 0x85 (e.g. if the last
1162 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1163 characters but of course it shouldn't be taken as a newline when it is part
1164 of another character. The bug was that, for an unlimited repeat of . in
1165 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1166 characters when looking for a newline.
1167
1168 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1169
1170 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1171 in debug output.
1172
1173 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1174 long printing in the pcrecpp unittest when running under MinGW.
1175
1176 6. ESC_K was left out of the EBCDIC table.
1177
1178 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1179 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1180 limit also applies to "virtual nesting" when a pattern is recursive, and in
1181 this case 1000 isn't so big. I have been able to remove this limit at the
1182 expense of backing off one optimization in certain circumstances. Normally,
1183 when pcre_exec() would call its internal match() function recursively and
1184 immediately return the result unconditionally, it uses a "tail recursion"
1185 feature to save stack. However, when a subpattern that can match an empty
1186 string has an unlimited repetition quantifier, it no longer makes this
1187 optimization. That gives it a stack frame in which to save the data for
1188 checking that an empty string has been matched. Previously this was taken
1189 from the 1000-entry workspace that had been reserved. So now there is no
1190 explicit limit, but more stack is used.
1191
1192 8. Applied Daniel's patches to solve problems with the import/export magic
1193 syntax that is required for Windows, and which was going wrong for the
1194 pcreposix and pcrecpp parts of the library. These were overlooked when this
1195 problem was solved for the main library.
1196
1197 9. There were some crude static tests to avoid integer overflow when computing
1198 the size of patterns that contain repeated groups with explicit upper
1199 limits. As the maximum quantifier is 65535, the maximum group length was
1200 set at 30,000 so that the product of these two numbers did not overflow a
1201 32-bit integer. However, it turns out that people want to use groups that
1202 are longer than 30,000 bytes (though not repeat them that many times).
1203 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1204 made it possible to implement the integer overflow checks in a much more
1205 dynamic way, which I have now done. The artificial limitation on group
1206 length has been removed - we now have only the limit on the total length of
1207 the compiled pattern, which depends on the LINK_SIZE setting.
1208
1209 10. Fixed a bug in the documentation for get/copy named substring when
1210 duplicate names are permitted. If none of the named substrings are set, the
1211 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1212 empty string.
1213
1214 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1215 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1216 because the ] is interpreted as the first data character and the
1217 terminating ] is not found. PCRE has been made compatible with Perl in this
1218 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1219 cause memory overwriting.
1220
1221 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1222 string has been matched (to stop an infinite loop). It was not recognizing
1223 a conditional subpattern that could match an empty string if that
1224 subpattern was within another subpattern. For example, it looped when
1225 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1226 condition was not nested. This bug has been fixed.
1227
1228 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1229 past the start of the subject in the presence of bytes with the top bit
1230 set, for example "\x8aBCD".
1231
1232 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1233 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1234
1235 14. Optimized (?!) to (*FAIL).
1236
1237 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1238 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1239 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1240 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1241 does: it's just the validity check that is more restrictive.
1242
1243 16. Inserted checks for integer overflows during escape sequence (backslash)
1244 processing, and also fixed erroneous offset values for syntax errors during
1245 backslash processing.
1246
1247 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1248 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1249
1250 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1251 caused an overrun.
1252
1253 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1254 something other than just ASCII characters) inside a group that had an
1255 unlimited repeat caused a loop at compile time (while checking to see
1256 whether the group could match an empty string).
1257
1258 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1259 [\P{Any}] did so. (Error in the code for printing property names.)
1260
1261 21. An orphan \E inside a character class could cause a crash.
1262
1263 22. A repeated capturing bracket such as (A)? could cause a wild memory
1264 reference during compilation.
1265
1266 23. There are several functions in pcre_compile() that scan along a compiled
1267 expression for various reasons (e.g. to see if it's fixed length for look
1268 behind). There were bugs in these functions when a repeated \p or \P was
1269 present in the pattern. These operators have additional parameters compared
1270 with \d, etc, and these were not being taken into account when moving along
1271 the compiled data. Specifically:
1272
1273 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1274 length.
1275
1276 (b) An item such as \pL+ within a repeated group could cause crashes or
1277 loops.
1278
1279 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1280 "reference to non-existent subpattern" error.
1281
1282 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1283
1284 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1285 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1286
1287 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1288
1289 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1290 character were causing crashes (broken optimization).
1291
1292 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1293 \p or \P) caused a compile-time loop.
1294
1295 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1296 break. For example, the unstudied pattern [\r\n]A does not match the string
1297 "\r\nA" because change 7.0/46 below moves the current point on by two
1298 characters after failing to match at the start. However, the pattern \nA
1299 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1300 the same is true. There doesn't seem any very clean way out of this, but
1301 what I have chosen to do makes the common cases work: PCRE now takes note
1302 of whether there can be an explicit match for \r or \n anywhere in the
1303 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1304 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1305 pattern has explicit CR or LF references.
1306
1307 29. Added (*CR) etc for changing newline setting at start of pattern.
1308
1309
1310 Version 7.2 19-Jun-07
1311 ---------------------
1312
1313 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1314 which is apparently normally available under Windows.
1315
1316 2. Re-jig the pcregrep tests with different newline settings in an attempt
1317 to make them independent of the local environment's newline setting.
1318
1319 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1320
1321 4. Some of the "internals" tests were previously cut out when the link size
1322 was not 2, because the output contained actual offsets. The recent new
1323 "Z" feature of pcretest means that these can be cut out, making the tests
1324 usable with all link sizes.
1325
1326 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1327 stack recursion. This gives a massive performance boost under BSD, but just
1328 a small improvement under Linux. However, it saves one field in the frame
1329 in all cases.
1330
1331 6. Added more features from the forthcoming Perl 5.10:
1332
1333 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1334 recursion call. It refers to the nth most recently opened parentheses.
1335
1336 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1337 to be opened parentheses.
1338
1339 (c) Conditions that refer to capturing parentheses can be specified
1340 relatively, for example, (?(-2)... or (?(+3)...
1341
1342 (d) \K resets the start of the current match so that everything before
1343 is not part of it.
1344
1345 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1346
1347 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1348 reference syntax.
1349
1350 (g) (?| introduces a group in which the numbering of parentheses in each
1351 alternative starts with the same number.
1352
1353 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1354
1355 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1356 PCRE_INFO_JCHANGED.
1357
1358 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1359 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1360 for detecting groups that can match an empty string.
1361
1362 9. A pattern with a very large number of alternatives (more than several
1363 hundred) was running out of internal workspace during the pre-compile
1364 phase, where pcre_compile() figures out how much memory will be needed. A
1365 bit of new cunning has reduced the workspace needed for groups with
1366 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1367 workspace instead of running out of the 4096 that are available.
1368
1369 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1370
1371 11. Applied patch from Google to remove an optimization that didn't quite work.
1372 The report of the bug said:
1373
1374 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1375 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1376 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1377
1378 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1379 it matched the wrong number of bytes.
1380
1381
1382 Version 7.1 24-Apr-07
1383 ---------------------
1384
1385 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1386 that is more "standard", making use of automake and other Autotools. There
1387 is some re-arrangement of the files and adjustment of comments consequent
1388 on this.
1389
1390 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1391 for recursive directory scanning broke on some systems because the files
1392 are not scanned in any specific order and on different systems the order
1393 was different. A call to "sort" has been inserted into RunGrepTest for the
1394 approprate test as a short-term fix. In the longer term there may be an
1395 alternative.
1396
1397 3. I had an email from Eric Raymond about problems translating some of PCRE's
1398 man pages to HTML (despite the fact that I distribute HTML pages, some
1399 people do their own conversions for various reasons). The problems
1400 concerned the use of low-level troff macros .br and .in. I have therefore
1401 removed all such uses from the man pages (some were redundant, some could
1402 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1403 HTML has been updated to handle .nf/.fi and to complain if it encounters
1404 .br or .in.
1405
1406 4. Updated comments in configure.ac that get placed in config.h.in and also
1407 arranged for config.h to be included in the distribution, with the name
1408 config.h.generic, for the benefit of those who have to compile without
1409 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1410
1411 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1412 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1413 makevp.bat for the current PCRE, using the additional files
1414 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1415
1416 6. A Windows user reported a minor discrepancy with test 2, which turned out
1417 to be caused by a trailing space on an input line that had got lost in his
1418 copy. The trailing space was an accident, so I've just removed it.
1419
1420 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1421 that is needed.
1422
1423 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1424 as "const" (a) because they are and (b) because it helps the PHP
1425 maintainers who have recently made a script to detect big data structures
1426 in the php code that should be moved to the .rodata section. I remembered
1427 to update Builducptable as well, so it won't revert if ucptable.h is ever
1428 re-created.
1429
1430 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1431 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1432 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1433 support is not required. This saves 1.5-2K of code, which is important in
1434 some applications.
1435
1436 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1437 so as not to refer to the tables, even though these functions will never be
1438 called when UTF-8 support is disabled. Otherwise there are problems with a
1439 shared library.
1440
1441 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1442
1443 (a) It was defining its arguments as char * instead of void *.
1444
1445 (b) It was assuming that all moves were upwards in memory; this was true
1446 a long time ago when I wrote it, but is no longer the case.
1447
1448 The emulated memove() is provided for those environments that have neither
1449 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1450 is clearly not the case, as these two bugs were recently reported.
1451
1452 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1453 and Detrail to create the HTML documentation, the .txt form of the man
1454 pages, and it removes trailing spaces from listed files. It also creates
1455 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1456 case, it wraps all the #defines with #ifndefs. This script should be run
1457 before "make dist".
1458
1459 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1460 with Unicode property support.
1461
1462 (a) For a maximizing quantifier, if the two different cases of the
1463 character were of different lengths in their UTF-8 codings (there are
1464 some cases like this - I found 11), and the matching function had to
1465 back up over a mixture of the two cases, it incorrectly assumed they
1466 were both the same length.
1467
1468 (b) When PCRE was configured to use the heap rather than the stack for
1469 recursion during matching, it was not correctly preserving the data for
1470 the other case of a UTF-8 character when checking ahead for a match
1471 while processing a minimizing repeat. If the check also involved
1472 matching a wide character, but failed, corruption could cause an
1473 erroneous result when trying to check for a repeat of the original
1474 character.
1475
1476 13. Some tidying changes to the testing mechanism:
1477
1478 (a) The RunTest script now detects the internal link size and whether there
1479 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1480 values substituted by "configure". (The RunGrepTest script already did
1481 this for UTF-8.) The configure.ac script no longer substitutes the
1482 relevant variables.
1483
1484 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1485 with length and offset values. This means that the output is different
1486 for different internal link sizes. Test 2 is skipped for link sizes
1487 other than 2 because of this, bypassing the problem. Unfortunately,
1488 there was also a test in test 3 (the locale tests) that used /B and
1489 failed for link sizes other than 2. Rather than cut the whole test out,
1490 I have added a new /Z option to pcretest that replaces the length and
1491 offset values with spaces. This is now used to make test 3 independent
1492 of link size. (Test 2 will be tidied up later.)
1493
1494 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1495 segmentation fault instead of returning the appropriate error message.
1496
1497 15. In multiline mode when the newline sequence was set to "any", the pattern
1498 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1499 This doesn't seem right; it now treats the CRLF combination as the line
1500 ending, and so does not match in that case. It's only a pattern such as ^$
1501 that would hit this one: something like ^ABC$ would have failed after \r
1502 and then tried again after \r\n.
1503
1504 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1505 in an attempt to make files that differ only in their line terminators
1506 compare equal. This works on Linux.
1507
1508 17. Under certain error circumstances pcregrep might try to free random memory
1509 as it exited. This is now fixed, thanks to valgrind.
1510
1511 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1512 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1513 was because its rules for how to advance for /g after matching an empty
1514 string at the end of a line did not allow for this case. They now check for
1515 it specially.
1516
1517 20. pcretest is supposed to handle patterns and data of any length, by
1518 extending its buffers when necessary. It was getting this wrong when the
1519 buffer for a data line had to be extended.
1520
1521 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1522 CRLF as a newline sequence.
1523
1524 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1525 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1526 I have nevertheless tidied it up.
1527
1528 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1529
1530 24. Added a man page for pcre-config.
1531
1532
1533 Version 7.0 19-Dec-06
1534 ---------------------
1535
1536 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1537 moving to gcc 4.1.1.
1538
1539 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1540 sys/time.h, which is documented as needed for this function. It doesn't
1541 seem to matter on Linux, but it showed up on some releases of OS X.
1542
1543 3. It seems that there are systems where bytes whose values are greater than
1544 127 match isprint() in the "C" locale. The "C" locale should be the
1545 default when a C program starts up. In most systems, only ASCII printing
1546 characters match isprint(). This difference caused the output from pcretest
1547 to vary, making some of the tests fail. I have changed pcretest so that:
1548
1549 (a) When it is outputting text in the compiled version of a pattern, bytes
1550 other than 32-126 are always shown as hex escapes.
1551
1552 (b) When it is outputting text that is a matched part of a subject string,
1553 it does the same, unless a different locale has been set for the match
1554 (using the /L modifier). In this case, it uses isprint() to decide.
1555
1556 4. Fixed a major bug that caused incorrect computation of the amount of memory
1557 required for a compiled pattern when options that changed within the
1558 pattern affected the logic of the preliminary scan that determines the
1559 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1560 that the computed length was too small. The symptoms of this bug were
1561 either the PCRE error "internal error: code overflow" from pcre_compile(),
1562 or a glibc crash with a message such as "pcretest: free(): invalid next
1563 size (fast)". Examples of patterns that provoked this bug (shown in
1564 pcretest format) are:
1565
1566 /(?-x: )/x
1567 /(?x)(?-x: \s*#\s*)/
1568 /((?i)[\x{c0}])/8
1569 /(?i:[\x{c0}])/8
1570
1571 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1572 is now done differently.
1573
1574 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1575 wrapper classes; (b) implement a new function in the C++ scanner that is
1576 more efficient than the old way of doing things because it avoids levels of
1577 recursion in the regex matching; (c) add a paragraph to the documentation
1578 for the FullMatch() function.
1579
1580 6. The escape sequence \n was being treated as whatever was defined as
1581 "newline". Not only was this contrary to the documentation, which states
1582 that \n is character 10 (hex 0A), but it also went horribly wrong when
1583 "newline" was defined as CRLF. This has been fixed.
1584
1585 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1586 was being set to -1 for the "end of line" case (supposedly a value that no
1587 character can have). Though this value is never used (the check for end of
1588 line is "zero bytes in current character"), it caused compiler complaints.
1589 I've changed it to 0xffffffff.
1590
1591 8. In pcre_version.c, the version string was being built by a sequence of
1592 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1593 string (as it is for production releases) called a macro with an empty
1594 argument. The C standard says the result of this is undefined. The gcc
1595 compiler treats it as an empty string (which was what was wanted) but it is
1596 reported that Visual C gives an error. The source has been hacked around to
1597 avoid this problem.
1598
1599 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1600 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1601 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1602 of them did).
1603
1604 10. Originally, pcretest opened its input and output without "b"; then I was
1605 told that "b" was needed in some environments, so it was added for release
1606 5.0 to both the input and output. (It makes no difference on Unix-like
1607 systems.) Later I was told that it is wrong for the input on Windows. I've
1608 now abstracted the modes into two macros, to make it easier to fiddle with
1609 them, and removed "b" from the input mode under Windows.
1610
1611 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1612
1613 12. Added -help and --help to pcretest as an official way of being reminded
1614 of the options.
1615
1616 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1617 and pcrecpp.cc because they annoy compilers at high warning levels.
1618
1619 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1620
1621 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1622 scripts are not C programs :-) and which was not noticed because it works
1623 on Linux.
1624
1625 16. pcretest is supposed to handle any length of pattern and data line (as one
1626 line or as a continued sequence of lines) by extending its input buffer if
1627 necessary. This feature was broken for very long pattern lines, leading to
1628 a string of junk being passed to pcre_compile() if the pattern was longer
1629 than about 50K.
1630
1631 17. I have done a major re-factoring of the way pcre_compile() computes the
1632 amount of memory needed for a compiled pattern. Previously, there was code
1633 that made a preliminary scan of the pattern in order to do this. That was
1634 OK when PCRE was new, but as the facilities have expanded, it has become
1635 harder and harder to keep it in step with the real compile phase, and there
1636 have been a number of bugs (see for example, 4 above). I have now found a
1637 cunning way of running the real compile function in a "fake" mode that
1638 enables it to compute how much memory it would need, while actually only
1639 ever using a few hundred bytes of working memory and without too many
1640 tests of the mode. This should make future maintenance and development
1641 easier. A side effect of this work is that the limit of 200 on the nesting
1642 depth of parentheses has been removed (though this was never a serious
1643 limitation, I suspect). However, there is a downside: pcre_compile() now
1644 runs more slowly than before (30% or more, depending on the pattern). I
1645 hope this isn't a big issue. There is no effect on runtime performance.
1646
1647 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1648 newline (only possible for the last line of a file) and it was a
1649 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1650
1651 19. Added additional timing features to pcretest. (1) The -tm option now times
1652 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1653 separate command line item, by a number that specifies the number of
1654 repeats to use when timing. The default is 50000; this gives better
1655 precision, but takes uncomfortably long for very large patterns.
1656
1657 20. Extended pcre_study() to be more clever in cases where a branch of a
1658 subpattern has no definite first character. For example, (a*|b*)[cd] would
1659 previously give no result from pcre_study(). Now it recognizes that the
1660 first character must be a, b, c, or d.
1661
1662 21. There was an incorrect error "recursive call could loop indefinitely" if
1663 a subpattern (or the entire pattern) that was being tested for matching an
1664 empty string contained only one non-empty item after a nested subpattern.
1665 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1666 incorrectly, because the \d was being skipped in the check.
1667
1668 22. The pcretest program now has a new pattern option /B and a command line
1669 option -b, which is equivalent to adding /B to every pattern. This causes
1670 it to show the compiled bytecode, without the additional information that
1671 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1672 is the same as /B/I).
1673
1674 23. A new optimization is now able automatically to treat some sequences such
1675 as a*b as a*+b. More specifically, if something simple (such as a character
1676 or a simple class like \d) has an unlimited quantifier, and is followed by
1677 something that cannot possibly match the quantified thing, the quantifier
1678 is automatically "possessified".
1679
1680 24. A recursive reference to a subpattern whose number was greater than 39
1681 went wrong under certain circumstances in UTF-8 mode. This bug could also
1682 have affected the operation of pcre_study().
1683
1684 25. Realized that a little bit of performance could be had by replacing
1685 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1686
1687 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1688
1689 27. Possessive quantifiers such as a++ were previously implemented by turning
1690 them into atomic groups such as ($>a+). Now they have their own opcodes,
1691 which improves performance. This includes the automatically created ones
1692 from 23 above.
1693
1694 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1695 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1696 the first matched character to be a colon. This applied both to named and
1697 numbered groups.
1698
1699 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1700
1701 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1702 building PCRE on Windows easier, so I have included it in the distribution.
1703
1704 31. There is now a check in pcretest against a ridiculously large number being
1705 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1706 loop, the loop is abandoned.
1707
1708 32. Forward references to subpatterns in conditions such as (?(2)...) where
1709 subpattern 2 is defined later cause pcre_compile() to search forwards in
1710 the pattern for the relevant set of parentheses. This search went wrong
1711 when there were unescaped parentheses in a character class, parentheses
1712 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1713
1714 33. "Subroutine" calls and backreferences were previously restricted to
1715 referencing subpatterns earlier in the regex. This restriction has now
1716 been removed.
1717
1718 34. Added a number of extra features that are going to be in Perl 5.10. On the
1719 whole, these are just syntactic alternatives for features that PCRE had
1720 previously implemented using the Python syntax or my own invention. The
1721 other formats are all retained for compatibility.
1722
1723 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1724 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1725 also .NET compatible.
1726
1727 (b) A recursion or subroutine call to a named group can now be defined as
1728 (?&name) as well as (?P>name).
1729
1730 (c) A backreference to a named group can now be defined as \k<name> or
1731 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1732 5.10, are also .NET compatible.
1733
1734 (d) A conditional reference to a named group can now use the syntax
1735 (?(<name>) or (?('name') as well as (?(name).
1736
1737 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1738 groups (named and numbered) that are never evaluated inline, but can be
1739 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1740 is always false. There may be only one alternative in such a group.
1741
1742 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1743 as the simple (?(R). The condition is true only if the most recent
1744 recursion is that of the given number or name. It does not search out
1745 through the entire recursion stack.
1746
1747 (g) The escape \gN or \g{N} has been added, where N is a positive or
1748 negative number, specifying an absolute or relative reference.
1749
1750 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1751 some "unreachable code" warnings.
1752
1753 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1754 things, this adds five new scripts.
1755
1756 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1757 There were also incompatibilities regarding the handling of \Q..\E inside
1758 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1759 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1760
1761 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1762 matches an empty string, and forcibly breaks the loop. There were bugs in
1763 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1764 against aaaa the result was just "a" rather than "aaaa", for example. Two
1765 separate and independent bugs (that affected different cases) have been
1766 fixed.
1767
1768 39. Refactored the code to abolish the use of different opcodes for small
1769 capturing bracket numbers. This is a tidy that I avoided doing when I
1770 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1771 The new approach is not only tidier, it makes it possible to reduce the
1772 memory needed to fix the previous bug (38).
1773
1774 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1775 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1776 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1777 mode.
1778
1779 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1780 report.
1781
1782 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1783 copy construction and assignment in the C++ wrapper.
1784
1785 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1786 couple of bugs that could have given wrong results in the "--newline=crlf"
1787 case.
1788
1789 44. Added a number of casts and did some reorganization of signed/unsigned int
1790 variables following suggestions from Dair Grant. Also renamed the variable
1791 "this" as "item" because it is a C++ keyword.
1792
1793 45. Arranged for dftables to add
1794
1795 #include "pcre_internal.h"
1796
1797 to pcre_chartables.c because without it, gcc 4.x may remove the array
1798 definition from the final binary if PCRE is built into a static library and
1799 dead code stripping is activated.
1800
1801 46. For an unanchored pattern, if a match attempt fails at the start of a
1802 newline sequence, and the newline setting is CRLF or ANY, and the next two
1803 characters are CRLF, advance by two characters instead of one.
1804
1805
1806 Version 6.7 04-Jul-06
1807 ---------------------
1808
1809 1. In order to handle tests when input lines are enormously long, pcretest has
1810 been re-factored so that it automatically extends its buffers when
1811 necessary. The code is crude, but this _is_ just a test program. The
1812 default size has been increased from 32K to 50K.
1813
1814 2. The code in pcre_study() was using the value of the re argument before
1815 testing it for NULL. (Of course, in any sensible call of the function, it
1816 won't be NULL.)
1817
1818 3. The memmove() emulation function in pcre_internal.h, which is used on
1819 systems that lack both memmove() and bcopy() - that is, hardly ever -
1820 was missing a "static" storage class specifier.
1821
1822 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1823 containing an extended class (one that cannot be represented by a bitmap
1824 because it contains high-valued characters or Unicode property items, e.g.
1825 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1826 pattern, but PCRE should not loop if you do not (it no longer does).
1827 [Detail: two cases were found: (a) a repeated subpattern containing an
1828 extended class; (b) a recursive reference to a subpattern that followed a
1829 previous extended class. It wasn't skipping over the extended class
1830 correctly when UTF-8 mode was not set.]
1831
1832 5. A negated single-character class was not being recognized as fixed-length
1833 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1834 compile error "lookbehind assertion is not fixed length".
1835
1836 6. The RunPerlTest auxiliary script was showing an unexpected difference
1837 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1838 write a Perl script that can interpret lines of an input file either as
1839 byte characters or as UTF-8, which is what "perltest" was being required to
1840 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1841 can't do is switch easily at run time between having the "use utf8;" pragma
1842 or not. In the end, I fudged it by using the RunPerlTest script to insert
1843 "use utf8;" explicitly for the UTF-8 tests.
1844
1845 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1846 the end of the subject string, contrary to the documentation and to what
1847 Perl does. This was true of both matching functions. Now it matches only at
1848 the start of the subject and immediately after *internal* newlines.
1849
1850 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1851 a pointer to an int instead of a pointer to an unsigned long int. This
1852 caused problems on 64-bit systems.
1853
1854 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1855 instance of the 'standard' template library not being so standard".
1856
1857 10. There was no check on the number of named subpatterns nor the maximum
1858 length of a subpattern name. The product of these values is used to compute
1859 the size of the memory block for a compiled pattern. By supplying a very
1860 long subpattern name and a large number of named subpatterns, the size
1861 computation could be caused to overflow. This is now prevented by limiting
1862 the length of names to 32 characters, and the number of named subpatterns
1863 to 10,000.
1864
1865 11. Subpatterns that are repeated with specific counts have to be replicated in
1866 the compiled pattern. The size of memory for this was computed from the
1867 length of the subpattern and the repeat count. The latter is limited to
1868 65535, but there was no limit on the former, meaning that integer overflow
1869 could in principle occur. The compiled length of a repeated subpattern is
1870 now limited to 30,000 bytes in order to prevent this.
1871
1872 12. Added the optional facility to have named substrings with the same name.
1873
1874 13. Added the ability to use a named substring as a condition, using the
1875 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1876 are numbers (not recommended). Forward references are permitted.
1877
1878 14. Added forward references in named backreferences (if you see what I mean).
1879
1880 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1881 pattern could run off the end of the subject. For example, the pattern
1882 "(?s)(.{1,5})"8 did this with the subject "ab".
1883
1884 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1885 PCRE_CASELESS was set when matching characters that were quantified with ?
1886 or *.
1887
1888 17. A character class other than a single negated character that had a minimum
1889 but no maximum quantifier - for example [ab]{6,} - was not handled
1890 correctly by pce_dfa_exec(). It would match only one character.
1891
1892 18. A valid (though odd) pattern that looked like a POSIX character
1893 class but used an invalid character after [ (for example [[,abc,]]) caused
1894 pcre_compile() to give the error "Failed: internal error: code overflow" or
1895 in some cases to crash with a glibc free() error. This could even happen if
1896 the pattern terminated after [[ but there just happened to be a sequence of
1897 letters, a binary zero, and a closing ] in the memory that followed.
1898
1899 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1900 over the years. Originally (before any Unicode support), just the bottom 8
1901 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1902 output from "man perlunicode" includes this:
1903
1904 The regular expression compiler produces polymorphic opcodes. That
1905 is, the pattern adapts to the data and automatically switches to
1906 the Unicode character scheme when presented with Unicode data--or
1907 instead uses a traditional byte scheme when presented with byte
1908 data.
1909
1910 Sadly, a wide octal escape does not cause a switch, and in a string with
1911 no other multibyte characters, these octal escapes are treated as before.
1912 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1913 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1914 Unicode string.
1915
1916 I have not perpetrated such confusion in PCRE. Up till now, it took just
1917 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1918 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1919 translate to the appropriate multibyte character.
1920
1921 29. Applied some refactoring to reduce the number of warnings from Microsoft
1922 and Borland compilers. This has included removing the fudge introduced
1923 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1924 a warning about an unused variable.
1925
1926 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1927 characters since release 4.0, because Perl (from release 5.004) does not.
1928 [Or at least, is documented not to: some releases seem to be in conflict
1929 with the documentation.] However, when a pattern was studied with
1930 pcre_study() and all its branches started with \s, PCRE still included VT
1931 as a possible starting character. Of course, this did no harm; it just
1932 caused an unnecessary match attempt.
1933
1934 22. Removed a now-redundant internal flag bit that recorded the fact that case
1935 dependency changed within the pattern. This was once needed for "required
1936 byte" processing, but is no longer used. This recovers a now-scarce options
1937 bit. Also moved the least significant internal flag bit to the most-
1938 significant bit of the word, which was not previously used (hangover from
1939 the days when it was an int rather than a uint) to free up another bit for
1940 the future.
1941
1942 23. Added support for CRLF line endings as well as CR and LF. As well as the
1943 default being selectable at build time, it can now be changed at runtime
1944 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1945 specify that it is scanning data with non-default line endings.
1946
1947 24. Changed the definition of CXXLINK to make it agree with the definition of
1948 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1949
1950 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1951 recursions. This makes a big different to stack usage for some patterns.
1952
1953 26. If a subpattern containing a named recursion or subroutine reference such
1954 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1955 the space required for the compiled pattern went wrong and gave too small a
1956 value. Depending on the environment, this could lead to "Failed: internal
1957 error: code overflow at offset 49" or "glibc detected double free or
1958 corruption" errors.
1959
1960 27. Applied patches from Google (a) to support the new newline modes and (b) to
1961 advance over multibyte UTF-8 characters in GlobalReplace.
1962
1963 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1964 difference for some implementation of PCRE in some Windows version.
1965
1966 29. Added some extra testing facilities to pcretest:
1967
1968 \q<number> in a data line sets the "match limit" value
1969 \Q<number> in a data line sets the "match recursion limt" value
1970 -S <number> sets the stack size, where <number> is in megabytes
1971
1972 The -S option isn't available for Windows.
1973
1974
1975 Version 6.6 06-Feb-06
1976 ---------------------
1977
1978 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1979 in pcreposix.h. I have copied the definition from pcre.h.
1980
1981 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1982 because pcre.h is no longer a built file.
1983
1984 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1985 not normally included in the compiled code.
1986
1987
1988 Version 6.5 01-Feb-06
1989 ---------------------
1990
1991 1. When using the partial match feature with pcre_dfa_exec(), it was not
1992 anchoring the second and subsequent partial matches at the new starting
1993 point. This could lead to incorrect results. For example, with the pattern
1994 /1234/, partially matching against "123" and then "a4" gave a match.
1995
1996 2. Changes to pcregrep:
1997
1998 (a) All non-match returns from pcre_exec() were being treated as failures
1999 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2000 error message is output. Some extra information is given for the
2001 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2002 probably the only errors that are likely to be caused by users (by
2003 specifying a regex that has nested indefinite repeats, for instance).
2004 If there are more than 20 of these errors, pcregrep is abandoned.
2005
2006 (b) A binary zero was treated as data while matching, but terminated the
2007 output line if it was written out. This has been fixed: binary zeroes
2008 are now no different to any other data bytes.
2009
2010 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2011 used to set a locale for matching. The --locale=xxxx long option has
2012 been added (no short equivalent) to specify a locale explicitly on the
2013 pcregrep command, overriding the environment variables.
2014
2015 (d) When -B was used with -n, some line numbers in the output were one less
2016 than they should have been.
2017
2018 (e) Added the -o (--only-matching) option.
2019
2020 (f) If -A or -C was used with -c (count only), some lines of context were
2021 accidentally printed for the final match.
2022
2023 (g) Added the -H (--with-filename) option.
2024
2025 (h) The combination of options -rh failed to suppress file names for files
2026 that were found from directory arguments.
2027
2028 (i) Added the -D (--devices) and -d (--directories) options.
2029
2030 (j) Added the -F (--fixed-strings) option.
2031
2032 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2033
2034 (l) Added the --colo(u)r option.
2035
2036 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2037 is not present by default.
2038
2039 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2040 items such as (?R) or (?1), when the recursion could match a number of
2041 alternatives. If it matched one of the alternatives, but subsequently,
2042 outside the recursion, there was a failure, the code tried to back up into
2043 the recursion. However, because of the way PCRE is implemented, this is not
2044 possible, and the result was an incorrect result from the match.
2045
2046 In order to prevent this happening, the specification of recursion has
2047 been changed so that all such subpatterns are automatically treated as
2048 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2049
2050 4. I had overlooked the fact that, in some locales, there are characters for
2051 which isalpha() is true but neither isupper() nor islower() are true. In
2052 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2053 and ordfeminine) are like this. This affected the treatment of \w and \W
2054 when they appeared in character classes, but not when they appeared outside
2055 a character class. The bit map for "word" characters is now created
2056 separately from the results of isalnum() instead of just taking it from the
2057 upper, lower, and digit maps. (Plus the underscore character, of course.)
2058
2059 5. The above bug also affected the handling of POSIX character classes such as
2060 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2061 permanent tables. Instead, the bit maps for such a class were previously
2062 created as the appropriate unions of the upper, lower, and digit bitmaps.
2063 Now they are created by subtraction from the [[:word:]] class, which has
2064 its own bitmap.
2065
2066 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2067 It is created by subtracting the vertical space characters (\x09, \x0a,
2068 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2069 subtraction was done in the overall bitmap for a character class, meaning
2070 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2071 be recognized. This bug has been fixed.
2072
2073 7. Patches from the folks at Google:
2074
2075 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2076 real life, but is still worth protecting against".
2077
2078 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2079 regular expressions".
2080
2081 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2082 have it.
2083
2084 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2085 "configure" and the latter not, in order to fix a problem somebody had
2086 with compiling the Arg class on HP-UX.
2087
2088 (e) Improve the error-handling of the C++ wrapper a little bit.
2089
2090 (f) New tests for checking recursion limiting.
2091
2092 8. The pcre_memmove() function, which is used only if the environment does not
2093 have a standard memmove() function (and is therefore rarely compiled),
2094 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2095 returning a result (though PCRE never actually uses the result).
2096
2097 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2098 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2099 returned instead of calling malloc() with an overflowing number that would
2100 most likely cause subsequent chaos.
2101
2102 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2103
2104 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2105 with this option is matched, the nmatch and pmatch options of regexec() are
2106 ignored.
2107
2108 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2109 provided in case anyone wants to the the POSIX interface with UTF-8
2110 strings.
2111
2112 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2113 C++ linking (needed for some HP-UX environments).
2114
2115 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2116 (unused parameter) and in the pcre_printint() function (omitted "default"
2117 switch label when the default is to do nothing).
2118
2119 15. Added some code to make it possible, when PCRE is compiled as a C++
2120 library, to replace subject pointers for pcre_exec() with a smart pointer
2121 class, thus making it possible to process discontinuous strings.
2122
2123 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2124 much the same function. They were added by different people who were trying
2125 to make PCRE easy to compile on non-Unix systems. It has been suggested
2126 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2127 for compiling on Windows systems. I have therefore replaced it with
2128 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2129 defaults to "extern" for C or "extern C" for C++, which works fine on
2130 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2131 SCOPE with something explicit in config.h. In addition:
2132
2133 (a) pcreposix.h still had just "extern" instead of either of these macros;
2134 I have replaced it with PCRE_DATA_SCOPE.
2135
2136 (b) Functions such as _pcre_xclass(), which are internal to the library,
2137 but external in the C sense, all had PCRE_EXPORT in their definitions.
2138 This is apparently wrong for the Windows case, so I have removed it.
2139 (It makes no difference on Unix-like systems.)
2140
2141 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2142 of recursive calls to match(). This is different to MATCH_LIMIT because
2143 that limits the total number of calls to match(), not all of which increase
2144 the depth of recursion. Limiting the recursion depth limits the amount of
2145 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2146 when PCRE is compiled, and changed at run time. A patch from Google adds
2147 this functionality to the C++ interface.
2148
2149 18. Changes to the handling of Unicode character properties:
2150
2151 (a) Updated the table to Unicode 4.1.0.
2152
2153 (b) Recognize characters that are not in the table as "Cn" (undefined).
2154
2155 (c) I revised the way the table is implemented to a much improved format
2156 which includes recognition of ranges. It now supports the ranges that
2157 are defined in UnicodeData.txt, and it also amalgamates other
2158 characters into ranges. This has reduced the number of entries in the
2159 table from around 16,000 to around 3,000, thus reducing its size
2160 considerably. I realized I did not need to use a tree structure after
2161 all - a binary chop search is just as efficient. Having reduced the
2162 number of entries, I extended their size from 6 bytes to 8 bytes to
2163 allow for more data.
2164
2165 (d) Added support for Unicode script names via properties such as \p{Han}.
2166
2167 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2168 matching that character.
2169
2170 20. When matching a repeated Unicode property with a minimum greater than zero,
2171 (for example \pL{2,}), PCRE could look past the end of the subject if it
2172 reached it while seeking the minimum number of characters. This could
2173 happen only if some of the characters were more than one byte long, because
2174 there is a check for at least the minimum number of bytes.
2175
2176 21. Refactored the implementation of \p and \P so as to be more general, to
2177 allow for more different types of property in future. This has changed the
2178 compiled form incompatibly. Anybody with saved compiled patterns that use
2179 \p or \P will have to recompile them.
2180
2181 22. Added "Any" and "L&" to the supported property types.
2182
2183 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2184 but give a compile time error if the value is greater than 0xff.
2185
2186 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2187 accidentally not being installed or uninstalled.
2188
2189 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2190 made were to insert the current release number. This seemed silly, because
2191 it made things harder for people building PCRE on systems that don't run
2192 "configure". I have turned pcre.h into a distributed file, no longer built
2193 by "configure", with the version identification directly included. There is
2194 no longer a pcre.h.in file.
2195
2196 However, this change necessitated a change to the pcre-config script as
2197 well. It is built from pcre-config.in, and one of the substitutions was the
2198 release number. I have updated configure.ac so that ./configure now finds
2199 the release number by grepping pcre.h.
2200
2201 26. Added the ability to run the tests under valgrind.
2202
2203
2204 Version 6.4 05-Sep-05
2205 ---------------------
2206
2207 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2208 "--" to be printed when multiple files were scanned, even when none of the
2209 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2210 consider it to be a bug, and have restored the previous behaviour.
2211
2212 2. A couple of code tidies to get rid of compiler warnings.
2213
2214 3. The pcretest program used to cheat by referring to symbols in the library
2215 whose names begin with _pcre_. These are internal symbols that are not
2216 really supposed to be visible externally, and in some environments it is
2217 possible to suppress them. The cheating is now confined to including
2218 certain files from the library's source, which is a bit cleaner.
2219
2220 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2221 file's purpose clearer.
2222
2223 5. Reorganized pcre_ucp_findchar().
2224
2225
2226 Version 6.3 15-Aug-05
2227 ---------------------
2228
2229 1. The file libpcre.pc.in did not have general read permission in the tarball.
2230
2231 2. There were some problems when building without C++ support:
2232
2233 (a) If C++ support was not built, "make install" and "make test" still
2234 tried to test it.
2235
2236 (b) There were problems when the value of CXX was explicitly set. Some
2237 changes have been made to try to fix these, and ...
2238
2239 (c) --disable-cpp can now be used to explicitly disable C++ support.
2240
2241 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2242 backslash in a target when C++ was disabled. This confuses some
2243 versions of "make", apparently. Using an intermediate variable solves
2244 this. (Same for CPP_LOBJ.)
2245
2246 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2247 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2248 necessary on certain architectures.
2249
2250 4. Added a setting of -export-symbols-regex to the link command to remove
2251 those symbols that are exported in the C sense, but actually are local
2252 within the library, and not documented. Their names all begin with
2253 "_pcre_". This is not a perfect job, because (a) we have to except some
2254 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2255 available (and never for static libraries). I have made a note to try to
2256 find a way round (a) in the future.
2257
2258
2259 Version 6.2 01-Aug-05
2260 ---------------------
2261
2262 1. There was no test for integer overflow of quantifier values. A construction
2263 such as {1111111111111111} would give undefined results. What is worse, if
2264 a minimum quantifier for a parenthesized subpattern overflowed and became
2265 negative, the calculation of the memory size went wrong. This could have
2266 led to memory overwriting.
2267
2268 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2269
2270 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2271 operating environments where this matters.
2272
2273 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2274 PCRE options from within the C++ wrapper.
2275
2276 5. Named capturing subpatterns were not being correctly counted when a pattern
2277 was compiled. This caused two problems: (a) If there were more than 100
2278 such subpatterns, the calculation of the memory needed for the whole
2279 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2280 back references of the form \12, where the number was greater than 9, were
2281 not recognized as back references, even though there were sufficient
2282 previous subpatterns.
2283
2284 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2285 versions of gcc, e.g. 2.95.4.
2286
2287
2288 Version 6.1 21-Jun-05
2289 ---------------------
2290
2291 1. There was one reference to the variable "posix" in pcretest.c that was not
2292 surrounded by "#if !defined NOPOSIX".
2293
2294 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2295 the cross-check on the old pcre_info() function, for the benefit of the
2296 cut-down version of PCRE that is currently imported into Exim.
2297
2298 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2299 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2300 patterns that start (?i) but I don't think that matters. The use of (?i) is
2301 just an example; this all applies to the other options as well.
2302
2303 4. Since libtool seems to echo the compile commands it is issuing, the output
2304 from "make" can be reduced a bit by putting "@" in front of each libtool
2305 compile command.
2306
2307 5. Patch from the folks at Google for configure.in to be a bit more thorough
2308 in checking for a suitable C++ installation before trying to compile the
2309 C++ stuff. This should fix a reported problem when a compiler was present,
2310 but no suitable headers.
2311
2312 6. The man pages all had just "PCRE" as their title. I have changed them to
2313 be the relevant file name. I have also arranged that these names are
2314 retained in the file doc/pcre.txt, which is a concatenation in text format
2315 of all the man pages except the little individual ones for each function.
2316
2317 7. The NON-UNIX-USE file had not been updated for the different set of source
2318 files that come with release 6. I also added a few comments about the C++
2319 wrapper.
2320
2321
2322 Version 6.0 07-Jun-05
2323 ---------------------
2324
2325 1. Some minor internal re-organization to help with my DFA experiments.
2326
2327 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2328 didn't matter for the library itself when fully configured, but did matter
2329 when compiling without UCP support, or within Exim, where the ucp files are
2330 not imported.
2331
2332 3. Refactoring of the library code to split up the various functions into
2333 different source modules. The addition of the new DFA matching code (see
2334 below) to a single monolithic source would have made it really too
2335 unwieldy, quite apart from causing all the code to be include in a
2336 statically linked application, when only some functions are used. This is
2337 relevant even without the DFA addition now that patterns can be compiled in
2338 one application and matched in another.
2339
2340 The downside of splitting up is that there have to be some external
2341 functions and data tables that are used internally in different modules of
2342 the library but which are not part of the API. These have all had their
2343 names changed to start with "_pcre_" so that they are unlikely to clash
2344 with other external names.
2345
2346 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2347 a different (DFA) algorithm. Although it is slower than the original
2348 function, it does have some advantages for certain types of matching
2349 problem.
2350
2351 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2352 including restarting after a partial match.
2353
2354 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2355 defined when compiling for Windows was sent to me. I have put it into the
2356 code, though I have no means of testing or verifying it.
2357
2358 7. Added the pcre_refcount() auxiliary function.
2359
2360 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2361 match before or at the first newline in the subject string. In pcretest,
2362 the /f option on a pattern can be used to set this.
2363
2364 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2365 would behave wrongly. This has been present in PCRE since release 4.0.
2366
2367 10. A number of changes to the pcregrep command:
2368
2369 (a) Refactored how -x works; insert ^(...)$ instead of setting
2370 PCRE_ANCHORED and checking the length, in preparation for adding
2371 something similar for -w.
2372
2373 (b) Added the -w (match as a word) option.
2374
2375 (c) Refactored the way lines are read and buffered so as to have more
2376 than one at a time available.
2377
2378 (d) Implemented a pcregrep test script.
2379
2380 (e) Added the -M (multiline match) option. This allows patterns to match
2381 over several lines of the subject. The buffering ensures that at least
2382 8K, or the rest of the document (whichever is the shorter) is available
2383 for matching (and similarly the previous 8K for lookbehind assertions).
2384
2385 (f) Changed the --help output so that it now says
2386
2387 -w, --word-regex(p)
2388
2389 instead of two lines, one with "regex" and the other with "regexp"
2390 because that confused at least one person since the short forms are the
2391 same. (This required a bit of code, as the output is generated
2392 automatically from a table. It wasn't just a text change.)
2393
2394 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2395 option but starts with a hyphen. Could be a pattern or a path name
2396 starting with a hyphen, for instance.
2397
2398 (h) "-" can be given as a file name to represent stdin.
2399
2400 (i) When file names are being printed, "(standard input)" is used for
2401 the standard input, for compatibility with GNU grep. Previously
2402 "<stdin>" was used.
2403
2404 (j) The option --label=xxx can be used to supply a name to be used for
2405 stdin when file names are being printed. There is no short form.
2406
2407 (k) Re-factored the options decoding logic because we are going to add
2408 two more options that take data. Such options can now be given in four
2409 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2410
2411 (l) Added the -A, -B, and -C options for requesting that lines of context
2412 around matches be printed.
2413
2414 (m) Added the -L option to print the names of files that do not contain
2415 any matching lines, that is, the complement of -l.
2416
2417 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2418 continue to scan other files.
2419
2420 (o) The -s option was incorrectly implemented. For compatibility with other
2421 greps, it now suppresses the error message for a non-existent or non-
2422 accessible file (but not the return code). There is a new option called
2423 -q that suppresses the output of matching lines, which was what -s was
2424 previously doing.
2425
2426 (p) Added --include and --exclude options to specify files for inclusion
2427 and exclusion when recursing.
2428
2429 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2430 Hopefully, it now does.
2431
2432 12. Missing cast in pcre_study().
2433
2434 13. Added an "uninstall" target to the makefile.
2435
2436 14. Replaced "extern" in the function prototypes in Makefile.in with
2437 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2438 world, but is set differently for Windows.
2439
2440 15. Added a second compiling function called pcre_compile2(). The only
2441 difference is that it has an extra argument, which is a pointer to an
2442 integer error code. When there is a compile-time failure, this is set
2443 non-zero, in addition to the error test pointer being set to point to an
2444 error message. The new argument may be NULL if no error number is required
2445 (but then you may as well call pcre_compile(), which is now just a
2446 wrapper). This facility is provided because some applications need a
2447 numeric error indication, but it has also enabled me to tidy up the way
2448 compile-time errors are handled in the POSIX wrapper.
2449
2450 16. Added VPATH=.libs to the makefile; this should help when building with one
2451 prefix path and installing with another. (Or so I'm told by someone who
2452 knows more about this stuff than I do.)
2453
2454 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2455 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2456 match everything, including newlines. This is not POSIX-compatible, but
2457 somebody wanted the feature. From pcretest it can be activated by using
2458 both the P and the s flags.
2459
2460 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2461
2462 19. libpcre.pc was being incorrectly installed as executable.
2463
2464 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2465 it now also looks for '\r' so that it will work unmodified on Windows.
2466
2467 21. Added Google's contributed C++ wrapper to the distribution.
2468
2469 22. Added some untidy missing memory free() calls in pcretest, to keep
2470 Electric Fence happy when testing.
2471
2472
2473
2474 Version 5.0 13-Sep-04
2475 ---------------------
2476
2477 1. Internal change: literal characters are no longer packed up into items
2478 containing multiple characters in a single byte-string. Each character
2479 is now matched using a separate opcode. However, there may be more than one
2480 byte in the character in UTF-8 mode.
2481
2482 2. The pcre_callout_block structure has two new fields: pattern_position and
2483 next_item_length. These contain the offset in the pattern to the next match
2484 item, and its length, respectively.
2485
2486 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2487 insertion of callouts before each pattern item. Added the /C option to
2488 pcretest to make use of this.
2489
2490 4. On the advice of a Windows user, the lines
2491
2492 #if defined(_WIN32) || defined(WIN32)
2493 _setmode( _fileno( stdout ), 0x8000 );
2494 #endif /* defined(_WIN32) || defined(WIN32) */
2495
2496 have been added to the source of pcretest. This apparently does useful
2497 magic in relation to line terminators.
2498
2499 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2500 for the benefit of those environments where the "b" makes a difference.
2501
2502 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2503 to know about it. I have put a hack into configure.in that adds in code
2504 to set GCC=yes if CC=icc. This seems to end up at a point in the
2505 generated configure script that is early enough to affect the setting of
2506 compiler options, which is what is needed, but I have no means of testing
2507 whether it really works. (The user who reported this had patched the
2508 generated configure script, which of course I cannot do.)
2509
2510 LATER: After change 22 below (new libtool files), the configure script
2511 seems to know about icc (and also ecc). Therefore, I have commented out
2512 this hack in configure.in.
2513
2514 7. Added support for pkg-config (2 patches were sent in).
2515
2516 8. Negated POSIX character classes that used a combination of internal tables
2517 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2518 [[:^ascii]]. Typically, they would match almost any characters. The other
2519 POSIX classes were not broken in this way.
2520
2521 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2522 to find the match, as PCRE was deluded into thinking that the match had to
2523 start at the start point or following a newline. The same bug applied to
2524 patterns with negative forward assertions or any backward assertions
2525 preceding ".*" at the start, unless the pattern required a fixed first
2526 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2527
2528 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2529 starting at the last subject character, bytes beyond the end of the subject
2530 string were read.
2531
2532 11. Renamed the variable "class" as "classbits" to make life easier for C++
2533 users. (Previously there was a macro definition, but it apparently wasn't
2534 enough.)
2535
2536 12. Added the new field "tables" to the extra data so that tables can be passed
2537 in at exec time, or the internal tables can be re-selected. This allows
2538 a compiled regex to be saved and re-used at a later time by a different
2539 program that might have everything at different addresses.
2540
2541 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2542 -R library as well as a -L library.
2543
2544 14. The debugging options of pcretest (-d on the command line or D on a
2545 pattern) showed incorrect output for anything following an extended class
2546 that contained multibyte characters and which was followed by a quantifier.
2547
2548 15. Added optional support for general category Unicode character properties
2549 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2550 support. It adds about 90K to the size of the library. The meanings of the
2551 inbuilt class escapes such as \d and \s have NOT been changed.
2552
2553 16. Updated pcredemo.c to include calls to free() to release the memory for the
2554 compiled pattern.
2555
2556 17. The generated file chartables.c was being created in the source directory
2557 instead of in the building directory. This caused the build to fail if the
2558 source directory was different from the building directory, and was
2559 read-only.
2560
2561 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2562 file. No doubt somebody will tell me if they don't make sense... Also added
2563 Dan Mooney's comments about building on OpenVMS.
2564
2565 19. Added support for partial matching via the PCRE_PARTIAL option for
2566 pcre_exec() and the \P data escape in pcretest.
2567
2568 20. Extended pcretest with 3 new pattern features:
2569
2570 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2571 write the compiled pattern to the file whose name is "rest-of-line".
2572 This is a straight binary dump of the data, with the saved pointer to
2573 the character tables forced to be NULL. The study data, if any, is
2574 written too. After writing, pcretest reads a new pattern.
2575
2576 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2577 compiled pattern from the given file. There must not be any
2578 occurrences of "<" in the file name (pretty unlikely); if there are,
2579 pcretest will instead treat the initial "<" as a pattern delimiter.
2580 After reading in the pattern, pcretest goes on to read data lines as
2581 usual.
2582
2583 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2584 and 16-bit fields in a compiled pattern, to simulate a pattern that
2585 was compiled on a host of opposite endianness.
2586
2587 21. The pcre-exec() function can now cope with patterns that were compiled on
2588 hosts of opposite endianness, with this restriction:
2589
2590 As for any compiled expression that is saved and used later, the tables
2591 pointer field cannot be preserved; the extra_data field in the arguments
2592 to pcre_exec() should be used to pass in a tables address if a value
2593 other than the default internal tables were used at compile time.
2594
2595 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2596 now diagnosed as an error. Previously, most of the time, a negative number
2597 would have been treated as zero, but if in addition "ovector" was passed as
2598 NULL, a crash could occur.
2599
2600 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2601 new versions from the libtool 1.5 distribution (the last one is a copy of
2602 a file called libtool.m4). This seems to have fixed the need to patch
2603 "configure" to support Darwin 1.3 (which I used to do). However, I still
2604 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2605 workstation).
2606
2607 24. Changed the PCRE licence to be the more standard "BSD" licence.
2608
2609
2610 Version 4.5 01-Dec-03
2611 ---------------------
2612
2613 1. There has been some re-arrangement of the code for the match() function so
2614 that it can be compiled in a version that does not call itself recursively.
2615 Instead, it keeps those local variables that need separate instances for
2616 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2617 needs to "recurse". Keeping track of where control must go is done by means
2618 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2619 hide most of the details from the main code, and operates only if
2620 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2621 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2622 operating.
2623
2624 To make it easier for callers to provide specially tailored get/free
2625 functions for this usage, two new functions, pcre_stack_malloc, and
2626 pcre_stack_free, are used. They are always called in strict stacking order,
2627 and the size of block requested is always the same.
2628
2629 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2630 PCRE has been compiled to use the stack or the heap for recursion. The
2631 -C option of pcretest uses this to show which version is compiled.
2632
2633 A new data escape \S, is added to pcretest; it causes the amounts of store
2634 obtained and freed by both kinds of malloc/free at match time to be added
2635 to the output.
2636
2637 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2638 what's available on my current Linux desktop machine.
2639
2640 3. When matching a UTF-8 string, the test for a valid string at the start has
2641 been extended. If start_offset is not zero, PCRE now checks that it points
2642 to a byte that is the start of a UTF-8 character. If not, it returns
2643 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2644 this is necessary because there may be backward assertions in the pattern.
2645 When matching the same subject several times, it may save resources to use
2646 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2647
2648 4. The code for checking the validity of UTF-8 strings has been tightened so
2649 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2650 containing "overlong sequences".
2651
2652 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2653 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2654 should have been "&", but it just so happened that all the cases this let
2655 through by mistake were picked up later in the function.
2656
2657 6. I had used a variable called "isblank" - this is a C99 function, causing
2658 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2659
2660 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2661 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2662 so the version is shown; (c) stop "make test" echoing "./RunTest".
2663
2664 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2665
2666 9. The prototype for memmove() for systems that don't have it was using
2667 size_t, but the inclusion of the header that defines size_t was later. I've
2668 moved the #includes for the C headers earlier to avoid this.
2669
2670 10. Added some adjustments to the code to make it easier to compiler on certain
2671 special systems:
2672
2673 (a) Some "const" qualifiers were missing.
2674 (b) Added the macro EXPORT before all exported functions; by default this
2675 is defined to be empty.
2676 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2677 that it reads its output file name as an argument instead of writing
2678 to the standard output and assuming this can be redirected.
2679
2680 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2681 class containing characters with values greater than 255, PCRE compilation
2682 went into a loop.
2683
2684 12. A recursive reference to a subpattern that was within another subpattern
2685 that had a minimum quantifier of zero caused PCRE to crash. For example,
2686 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2687 recursion. If the recursively-called subpattern itself had a zero repeat,
2688 that was OK.
2689
2690 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2691 buffer into which it was copied (for escape processing) was still set at
2692 1024, so long lines caused crashes.
2693
2694 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2695 "internal error: code overflow...". This applied to any character class
2696 that was followed by a possessive quantifier.
2697
2698 15. Modified the Makefile to add libpcre.la as a prerequisite for
2699 libpcreposix.la because I was told this is needed for a parallel build to
2700 work.
2701
2702 16. If a pattern that contained .* following optional items at the start was
2703 studied, the wrong optimizing data was generated, leading to matching
2704 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2705 matching string must start with a or b or c. The correct conclusion for
2706 this pattern is that a match can start with any character.
2707
2708
2709 Version 4.4 13-Aug-03
2710 ---------------------
2711
2712 1. In UTF-8 mode, a character class containing characters with values between
2713 127 and 255 was not handled correctly if the compiled pattern was studied.
2714 In fixing this, I have also improved the studying algorithm for such
2715 classes (slightly).
2716
2717 2. Three internal functions had redundant arguments passed to them. Removal
2718 might give a very teeny performance improvement.
2719
2720 3. Documentation bug: the value of the capture_top field in a callout is *one
2721 more than* the number of the hightest numbered captured substring.
2722
2723 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2724 in incorrectly linking with a previously installed version. They now link
2725 explicitly with libpcre.la.
2726
2727 5. configure.in no longer needs to recognize Cygwin specially.
2728
2729 6. A problem in pcre.in for Windows platforms is fixed.
2730
2731 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2732 pcretest, it used to include the size of the study block as part of its
2733 output. Unfortunately, the structure contains a field that has a different
2734 size on different hardware architectures. This meant that the tests that
2735 showed this size failed. As the block is currently always of a fixed size,
2736 this information isn't actually particularly useful in pcretest output, so
2737 I have just removed it.
2738
2739 8. Three pre-processor statements accidentally did not start in column 1.
2740 Sadly, there are *still* compilers around that complain, even though
2741 standard C has not required this for well over a decade. Sigh.
2742
2743 9. In pcretest, the code for checking callouts passed small integers in the
2744 callout_data field, which is a void * field. However, some picky compilers
2745 complained about the casts involved for this on 64-bit systems. Now
2746 pcretest passes the address of the small integer instead, which should get
2747 rid of the warnings.
2748
2749 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2750 both compile and run time, and gives an error if an invalid UTF-8 sequence
2751 is found. There is a option for disabling this check in cases where the
2752 string is known to be correct and/or the maximum performance is wanted.
2753
2754 11. In response to a bug report, I changed one line in Makefile.in from
2755
2756 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2757 to
2758 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2759
2760 to look similar to other lines, but I have no way of telling whether this
2761 is the right thing to do, as I do not use Windows. No doubt I'll get told
2762 if it's wrong...
2763
2764
2765 Version 4.3 21-May-03
2766 ---------------------
2767
2768 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2769 Makefile.
2770
2771 2. Some refactoring to improve the quality of the code:
2772
2773 (i) The utf8_table... variables are now declared "const".
2774
2775 (ii) The code for \cx, which used the "case flipping" table to upper case
2776 lower case letters, now just substracts 32. This is ASCII-specific,
2777 but the whole concept of \cx is ASCII-specific, so it seems
2778 reasonable.
2779
2780 (iii) PCRE was using its character types table to recognize decimal and
2781 hexadecimal digits in the pattern. This is silly, because it handles
2782 only 0-9, a-f, and A-F, but the character types table is locale-
2783 specific, which means strange things might happen. A private
2784 table is now used for this - though it costs 256 bytes, a table is
2785 much faster than multiple explicit tests. Of course, the standard
2786 character types table is still used for matching digits in subject
2787 strings against \d.
2788
2789 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2790 ending in _t are). So I've renamed it as ESC_tee.
2791
2792 3. The first argument for regexec() in the POSIX wrapper should have been
2793 defined as "const".
2794
2795 4. Changed pcretest to use malloc() for its buffers so that they can be
2796 Electric Fenced for debugging.
2797
2798 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2799 to read one or more bytes before the start of the subject string. Often this
2800 had no effect on PCRE's behaviour, but in some circumstances it could
2801 provoke a segmentation fault.
2802
2803 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2804 to try to read one or more bytes before the start of the subject string.
2805
2806 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2807 UTF-8 support could misbehave in various ways if the subject string
2808 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2809 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2810 back over UTF-8 characters.)
2811
2812
2813 Version 4.2 14-Apr-03
2814 ---------------------
2815
2816 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2817
2818 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2819 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2820 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2821 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2822 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2823 and BUILD_EXEEXT
2824 Note: automatic setting of the BUILD variables is not yet working
2825 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2826 compile-time but not at link-time
2827 [LINK]: use for linking executables only
2828 make different versions for Windows and non-Windows
2829 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2830 libraries
2831 [LINK_FOR_BUILD]: new variable
2832 [OBJEXT]: use throughout
2833 [EXEEXT]: use throughout
2834 <winshared>: new target
2835 <wininstall>: new target
2836 <dftables.o>: use native compiler
2837 <dftables>: use native linker
2838 <install>: handle Windows platform correctly
2839 <clean>: ditto
2840 <check>: ditto
2841 copy DLL to top builddir before testing
2842
2843 As part of these changes, -no-undefined was removed again. This was reported
2844 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2845 in any case.
2846
2847 3. Some tidies to get rid of compiler warnings:
2848
2849 . In the match_data structure, match_limit was an unsigned long int, whereas
2850 match_call_count was an int. I've made them both unsigned long ints.
2851
2852 . In pcretest the fact that a const uschar * doesn't automatically cast to
2853 a void * provoked a warning.
2854
2855 . Turning on some more compiler warnings threw up some "shadow" variables
2856 and a few more missing casts.
2857
2858 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2859 option, a class that contained a single character with a value between 128
2860 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2861
2862 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2863 option, a class that contained several characters, but with at least one
2864 whose value was between 128 and 255 caused PCRE to crash.
2865
2866
2867 Version 4.1 12-Mar-03
2868 ---------------------
2869
2870 1. Compiling with gcc -pedantic found a couple of places where casts were
2871 needed, and a string in dftables.c that was longer than standard compilers are
2872 required to support.
2873
2874 2. Compiling with Sun's compiler found a few more places where the code could
2875 be tidied up in order to avoid warnings.
2876
2877 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2878 first of these names is deprecated in the latest Autoconf in favour of the name
2879 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2880 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2881 analogy I have changed it to CFLAGS_FOR_BUILD.
2882
2883 4. Added -no-undefined to the linking command in the Makefile, because this is
2884 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2885 linking step for the pcreposix library.
2886
2887 5. PCRE was failing to diagnose the case of two named groups with the same
2888 name.
2889
2890 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2891 literal character that is needed in the subject for a match, and scans along to
2892 ensure that it is present before embarking on the full matching process. This
2893 saves time in cases of nested unlimited repeats that are never going to match.
2894 Problem: the scan can take a lot of time if the subject is very long (e.g.
2895 megabytes), thus penalizing straightforward matches. It is now done only if the
2896 amount of subject to be scanned is less than 1000 bytes.
2897
2898 7. A lesser problem with the same optimization is that it was recording the
2899 first character of an anchored pattern as "needed", thus provoking a search
2900 right along the subject, even when the first match of the pattern was going to
2901 fail. The "needed" character is now not set for anchored patterns, unless it
2902 follows something in the pattern that is of non-fixed length. Thus, it still
2903 fulfils its original purpose of finding quick non-matches in cases of nested
2904 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2905
2906
2907 Version 4.0 17-Feb-03
2908 ---------------------
2909
2910 1. If a comment in an extended regex that started immediately after a meta-item
2911 extended to the end of string, PCRE compiled incorrect data. This could lead to
2912 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2913
2914 2. Moved to autoconf 2.53 and libtool 1.4.2.
2915
2916 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2917 the special perltest8 script is no longer needed - all the tests can be run
2918 from a single perltest script.
2919
2920 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2921 by \s. It has now been removed in PCRE. This means it isn't recognized as
2922 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2923 class [:space:] *does* include VT, thereby creating a mess.
2924
2925 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2926 space and tab.
2927
2928 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2929 its new features into the main test script, reducing the number of scripts.
2930
2931 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2932 were backward compatible, and made the (?i) apply to the whole pattern, as if
2933 /i were given. Now it behaves more logically, and applies the option setting
2934 only to what follows. PCRE has been changed to follow suit. However, if it
2935 finds options settings right at the start of the pattern, it extracts them into
2936 the global options, as before. Thus, they show up in the info data.
2937
2938 8. Added support for the \Q...\E escape sequence. Characters in between are
2939 treated as literals. This is slightly different from Perl in that $ and @ are
2940 also handled as literals inside the quotes. In Perl, they will cause variable
2941 interpolation. Note the following examples:
2942
2943 Pattern PCRE matches Perl matches
2944
2945 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2946 \Qabc\$xyz\E abc\$xyz abc\$xyz
2947 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2948
2949 For compatibility with Perl, \Q...\E sequences are recognized inside character
2950 classes as well as outside them.
2951
2952 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2953 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2954 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2955 signed/unsigned warnings.
2956
2957 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2958 option for pcretest, so I've replaced it by a simple function that does just
2959 that job.
2960
2961 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2962 "pcregrep -".
2963
2964 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2965 Java package. This provides some syntactic sugar for simple cases of what my
2966 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2967 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2968 item, you can use this simplified notation. Note that only makes sense with
2969 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2970 greediness, whatever the setting of the PCRE_UNGREEDY option.
2971
2972 13. A change of greediness default within a pattern was not taking effect at
2973 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2974 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2975 was abstracted outside.
2976
2977 14. PCRE now supports the \G assertion. It is true when the current matching
2978 position is at the start point of the match. This differs from \A when the
2979 starting offset is non-zero. Used with the /g option of pcretest (or similar
2980 code), it works in the same way as it does for Perl's /g option. If all
2981 alternatives of a regex begin with \G, the expression is anchored to the start
2982 match position, and the "anchored" flag is set in the compiled expression.
2983
2984 15. Some bugs concerning the handling of certain option changes within patterns
2985 have been fixed. These applied to options other than (?ims). For example,
2986 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2987 been the other way round. Some of this was related to change 7 above.
2988
2989 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2990 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2991 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2992 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2993
2994 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2995 mode. Unlike ".", it always matches newline, whatever the setting of
2996 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2997 assertions. Perl allows it, but it doesn't (in general) work because it can't
2998 calculate the length of the lookbehind. At least, that's the case for Perl
2999 5.8.0 - I've been told they are going to document that it doesn't work in
3000 future.
3001
3002 18. Added an error diagnosis for escapes that PCRE does not support: these are
3003 \L, \l, \N, \P, \p, \U, \u, and \X.
3004
3005 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3006 reading past the end of the pattern in cases such as /[abcd/.
3007
3008 20. PCRE was getting more memory than necessary for patterns with classes that
3009 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3010
3011 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3012 compiling PCRE for use with Virtual Pascal.
3013
3014 22. Small fix to the Makefile to make it work properly if the build is done
3015 outside the source tree.
3016
3017 23. Added a new extension: a condition to go with recursion. If a conditional
3018 subpattern starts with (?(R) the "true" branch is used if recursion has
3019 happened, whereas the "false" branch is used only at the top level.
3020
3021 24. When there was a very long string of literal characters (over 255 bytes
3022 without UTF support, over 250 bytes with UTF support), the computation of how
3023 much memory was required could be incorrect, leading to segfaults or other
3024 strange effects.
3025
3026 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3027 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3028 there was a subsequent back reference to those brackets. This meant that, for
3029 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3030 possible to check for precisely this case. All we can do is abandon the
3031 optimization if .* occurs inside capturing brackets when there are any back
3032 references whatsoever. (See below for a better fix that came later.)
3033
3034 26. The handling of the optimization for finding the first character of a
3035 non-anchored pattern, and for finding a character that is required later in the
3036 match were failing in some cases. This didn't break the matching; it just
3037 failed to optimize when it could. The way this is done has been re-implemented.
3038
3039 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3040
3041 28. Added a new feature that provides some of the functionality that Perl
3042 provides with (?{...}). The facility is termed a "callout". The way it is done
3043 in PCRE is for the caller to provide an optional function, by setting
3044 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3045 global variable. By default it is unset, which disables all calling out. To get
3046 the function called, the regex must include (?C) at appropriate points. This
3047 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3048 This provides a means of identifying different callout points. When PCRE
3049 reaches such a point in the regex, if pcre_callout has been set, the external
3050 function is called. It is provided with data in a structure called
3051 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3052 matching continues; if it returns a non-zero value, the match at the current
3053 point fails. However, backtracking will occur if possible. [This was changed
3054 later and other features added - see item 49 below.]
3055
3056 29. pcretest is upgraded to test the callout functionality. It provides a
3057 callout function that displays information. By default, it shows the start of
3058 the match and the current position in the text. There are some new data escapes
3059 to vary what happens:
3060
3061 \C+ in addition, show current contents of captured substrings
3062 \C- do not supply a callout function
3063 \C!n return 1 when callout number n is reached
3064 \C!n!m return 1 when callout number n is reached for the mth time
3065
3066 30. If pcregrep was called with the -l option and just a single file name, it
3067 output "<stdin>" if a match was found, instead of the file name.
3068
3069 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3070 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3071 pcre_exec(). This saves a malloc/free per call. The default value of
3072 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3073 when configuring.
3074
3075 32. The default maximum size of a compiled pattern is 64K. There have been a
3076 few cases of people hitting this limit. The code now uses macros to handle the
3077 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3078 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3079 configuring. Tests 2 and 5 work only with 2-byte links because they output
3080 debugging information about compiled patterns.
3081
3082 33. Internal code re-arrangements:
3083
3084 (a) Moved the debugging function for printing out a compiled regex into
3085 its own source file (printint.c) and used #include to pull it into
3086 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3087 separate copies.
3088
3089 (b) Defined the list of op-code names for debugging as a macro in
3090 internal.h so that it is next to the definition of the opcodes.
3091
3092 (c) Defined a table of op-code lengths for simpler skipping along compiled
3093 code. This is again a macro in internal.h so that it is next to the
3094 definition of the opcodes.
3095
3096 34. Added support for recursive calls to individual subpatterns, along the
3097 lines of Robin Houston's patch (but implemented somewhat differently).
3098
3099 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3100 allow it to read and process whole directories in Win32. This code was
3101 contributed by Lionel Fourquaux; it has not been tested by me.
3102
3103 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3104 used to name a group. Names consist of alphanumerics and underscores, and must
3105 be unique. Back references use the syntax (?P=name) and recursive calls use
3106 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3107 numbers. The function pcre_fullinfo() can be used after compilation to extract
3108 a name/number map. There are three relevant calls:
3109
3110 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3111 PCRE_INFO_NAMECOUNT yields the number of entries
3112 PCRE_INFO_NAMETABLE yields a pointer to the map.
3113
3114 The map is a vector of fixed-size entries. The size of each entry depends on
3115 the length of the longest name used. The first two bytes of each entry are the
3116 group number, most significant byte first. There follows the corresponding
3117 name, zero terminated. The names are in alphabetical order.
3118
3119 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3120 case instead of 255. Making it the same both with and without UTF-8 support
3121 means that the same test output works with both.
3122
3123 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3124 calling malloc() with a zero argument.
3125
3126 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3127 optimization. I've improved things by keeping a bitmap of backreferences with
3128 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3129 fact referenced, the optimization can be applied. It is unlikely that a
3130 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3131 the match to follow \n) will appear inside brackets with a number greater than
3132 31, but if it does, any back reference > 31 suppresses the optimization.
3133
3134 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3135 of disabling numbered capturing parentheses. Any opening parenthesis that is
3136 not followed by ? behaves as if it were followed by ?: but named parentheses
3137 can still be used for capturing (and they will acquire numbers in the usual
3138 way).
3139
3140 41. Redesigned the return codes from the match() function into yes/no/error so
3141 that errors can be passed back from deep inside the nested calls. A malloc
3142 failure while inside a recursive subpattern call now causes the
3143 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3144
3145 42. It is now possible to set a limit on the number of times the match()
3146 function is called in a call to pcre_exec(). This facility makes it possible to
3147 limit the amount of recursion and backtracking, though not in a directly
3148 obvious way, because the match() function is used in a number of different
3149 circumstances. The count starts from zero for each position in the subject
3150 string (for non-anchored patterns). The default limit is, for compatibility, a
3151 large number, namely 10 000 000. You can change this in two ways:
3152
3153 (a) When configuring PCRE before making, you can use --with-match-limit=n
3154 to set a default value for the compiled library.
3155
3156 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3157 a different value is set. See 45 below.
3158
3159 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3160
3161 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3162 of things that can be changed at compile time. The first argument specifies
3163 what is wanted and the second points to where the information is to be placed.
3164 The current list of available information is:
3165
3166 PCRE_CONFIG_UTF8
3167
3168 The output is an integer that is set to one if UTF-8 support is available;
3169 otherwise it is set to zero.
3170
3171 PCRE_CONFIG_NEWLINE
3172
3173 The output is an integer that it set to the value of the code that is used for
3174 newline. It is either LF (10) or CR (13).
3175
3176 PCRE_CONFIG_LINK_SIZE
3177
3178 The output is an integer that contains the number of bytes used for internal
3179 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3180
3181 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3182
3183 The output is an integer that contains the threshold above which the POSIX
3184 interface uses malloc() for output vectors. See item 31 above.
3185
3186 PCRE_CONFIG_MATCH_LIMIT
3187
3188 The output is an unsigned integer that contains the default limit of the number
3189 of match() calls in a pcre_exec() execution. See 42 above.
3190
3191 44. pcretest has been upgraded by the addition of the -C option. This causes it
3192 to extract all the available output from the new pcre_config() function, and to
3193 output it. The program then exits immediately.
3194
3195 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3196 order to support additional features. One way would have been to define
3197 pcre_exec2() (for example) with extra arguments, but this would not have been
3198 extensible, and would also have required all calls to the original function to
3199 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3200 is used for passing in "extra" data from pcre_study().
3201
3202 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3203 contains the following fields:
3204
3205 flags a bitmap indicating which of the following fields are set
3206 study_data opaque data from pcre_study()
3207 match_limit a way of specifying a limit on match() calls for a specific
3208 call to pcre_exec()
3209 callout_data data for callouts (see 49 below)
3210
3211 The flag bits are also defined in pcre.h, and are
3212
3213 PCRE_EXTRA_STUDY_DATA
3214 PCRE_EXTRA_MATCH_LIMIT
3215 PCRE_EXTRA_CALLOUT_DATA
3216
3217 The pcre_study() function now returns one of these new pcre_extra blocks, with
3218 the actual study data pointed to by the study_data field, and the
3219 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3220 before. That is, this change is entirely upwards-compatible and requires no
3221 change to existing code.
3222
3223 If you want to pass in additional data to pcre_exec(), you can either place it
3224 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3225 block.
3226
3227 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3228 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3229 times with different match limits, until it finds the minimum value needed for
3230 pcre_exec() to complete. The value is then output. This can be instructive; for
3231 most simple matches the number is quite small, but for pathological cases it
3232 gets very large very quickly.
3233
3234 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3235 returns the size of the data block pointed to by the study_data field in a
3236 pcre_extra block, that is, the value that was passed as the argument to
3237 pcre_malloc() when PCRE was getting memory in which to place the information
3238 created by pcre_study(). The fourth argument should point to a size_t variable.
3239 pcretest has been extended so that this information is shown after a successful
3240 pcre_study() call when information about the compiled regex is being displayed.
3241
3242 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3243 because what follows is always an absolute path. (Later: it turns out that this
3244 is more than cosmetic for MinGW, because it doesn't like empty path
3245 components.)
3246
3247 49. Some changes have been made to the callout feature (see 28 above):
3248
3249 (i) A callout function now has three choices for what it returns:
3250
3251 0 => success, carry on matching
3252 > 0 => failure at this point, but backtrack if possible
3253 < 0 => serious error, return this value from pcre_exec()
3254
3255 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3256 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3257 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3258 use by callout functions. It will never be used by PCRE itself.
3259
3260 (ii) The pcre_extra structure (see 45 above) has a void * field called
3261 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3262 pcre_callout_block structure has a field of the same name. The contents of
3263 the field passed in the pcre_extra structure are passed to the callout
3264 function in the corresponding field in the callout block. This makes it
3265 easier to use the same callout-containing regex from multiple threads. For
3266 testing, the pcretest program has a new data escape
3267
3268 \C*n pass the number n (may be negative) as callout_data
3269
3270 If the callout function in pcretest receives a non-zero value as
3271 callout_data, it returns that value.
3272
3273 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3274 there were some redundant $(CFLAGS) in commands that are now specified as
3275 $(LINK), which already includes $(CFLAGS).
3276
3277 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3278 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3279 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3280 one-byte characters throughout. Note that case-insensitive matching applies
3281 only to characters whose values are less than 256. PCRE doesn't support the
3282 notion of cases for higher-valued characters.
3283
3284 (i) A character class whose characters are all within 0-255 is handled as
3285 a bit map, and the map is inverted for negative classes. Previously, a
3286 character > 255 always failed to match such a class; however it should
3287 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3288
3289 (ii) A negated character class with a single character < 255 is coded as
3290 "not this character" (OP_NOT). This wasn't working properly when the test
3291 character was multibyte, either singly or repeated.
3292
3293 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3294 mode, for example: \x{100}{2,3}.
3295
3296 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3297 singly or repeated) now correctly test multibyte characters. However,
3298 PCRE doesn't recognize any characters with values greater than 255 as
3299 digits, spaces, or word characters. Such characters always match \D, \S,
3300 and \W, and never match \d, \s, or \w.
3301
3302 (v) Classes may now contain characters and character ranges with values
3303 greater than 255. For example: [ab\x{100}-\x{400}].
3304
3305 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3306 PCRE in UTF-8 mode.
3307
3308 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3309 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3310 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3311 value.)
3312
3313 53. The single man page has become too large. I have therefore split it up into
3314 a number of separate man pages. These also give rise to individual HTML pages;
3315 these are now put in a separate directory, and there is an index.html page that
3316 lists them all. Some hyperlinking between the pages has been installed.
3317
3318 54. Added convenience functions for handling named capturing parentheses.
3319
3320 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3321 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3322 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3323 are faulted.
3324
3325 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3326 calling configure. These values are used when compiling the dftables.c program
3327 which is run to generate the source of the default character tables. They
3328 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3329 you will need to set these values.
3330
3331 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3332
3333
3334 Version 3.9 02-Jan-02
3335 ---------------------
3336
3337 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3338
3339 2. If --disable-static was given, the building process failed when trying to
3340 build pcretest and pcregrep. (For some reason it was using libtool to compile
3341 them, which is not right, as they aren't part of the library.)
3342
3343
3344 Version 3.8 18-Dec-01
3345 ---------------------
3346
3347 1. The experimental UTF-8 code was completely screwed up. It was packing the
3348 bytes in the wrong order. How dumb can you get?
3349
3350
3351 Version 3.7 29-Oct-01
3352 ---------------------
3353
3354 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3355 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3356 this didn't happen under Solaris 8, where I normally test things.
3357
3358 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3359 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3360 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3361 this fix an hour or so after the initial 3.7 release.)
3362
3363
3364 Version 3.6 23-Oct-01
3365 ---------------------
3366
3367 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3368 offsets passed as NULL with zero offset count.
3369
3370 2. The config.guess and config.sub files had not been updated when I moved to
3371 the latest autoconf.
3372
3373
3374 Version 3.5 15-Aug-01
3375 ---------------------
3376
3377 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3378 had been forgotten.
3379
3380 2. By using declared but undefined structures, we can avoid using "void"
3381 definitions in pcre.h while keeping the internal definitions of the structures
3382 private.
3383
3384 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3385 user point of view, this means that both static and shared libraries are built
3386 by default, but this can be individually controlled. More of the work of
3387 handling this static/shared cases is now inside libtool instead of PCRE's make
3388 file.
3389
3390 4. The pcretest utility is now installed along with pcregrep because it is
3391 useful for users (to test regexs) and by doing this, it automatically gets
3392 relinked by libtool. The documentation has been turned into a man page, so
3393 there are now .1, .txt, and .html versions in /doc.
3394
3395 5. Upgrades to pcregrep:
3396 (i) Added long-form option names like gnu grep.
3397 (ii) Added --help to list all options with an explanatory phrase.
3398 (iii) Added -r, --recursive to recurse into sub-directories.
3399 (iv) Added -f, --file to read patterns from a file.
3400
3401 6. pcre_exec() was referring to its "code" argument before testing that
3402 argument for NULL (and giving an error if it was NULL).
3403
3404 7. Upgraded Makefile.in to allow for compiling in a different directory from
3405 the source directory.
3406
3407 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3408 options bits, the pointer it was passed was to an int instead of to an unsigned
3409 long int. This mattered only on 64-bit systems.
3410
3411 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3412 generated) instead of pcre.in, which it its source. Also made the same change
3413 in several of the .c files.
3414
3415 10. A new release of gcc defines printf() as a macro, which broke pcretest
3416 because it had an ifdef in the middle of a string argument for printf(). Fixed
3417 by using separate calls to printf().
3418
3419 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3420 script, to force use of CR or LF instead of \n in the source. On non-Unix
3421 systems, the value can be set in config.h.
3422
3423 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3424 absolute limit. Changed the text of the error message to make this clear, and
3425 likewise updated the man page.
3426
3427 13. The limit of 99 on the number of capturing subpatterns has been removed.
3428 The new limit is 65535, which I hope will not be a "real" limit.
3429
3430
3431 Version 3.4 22-Aug-00
3432 ---------------------
3433
3434 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3435
3436 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3437
3438
3439 Version 3.3 01-Aug-00
3440 ---------------------
3441
3442 1. If an octal character was given, but the value was greater than \377, it
3443 was not getting masked to the least significant bits, as documented. This could
3444 lead to crashes in some systems.
3445
3446 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3447 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3448
3449 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3450 These just pass their arguments on to (pcre_free)(), but they are provided
3451 because some uses of PCRE bind it to non-C systems that can call its functions,
3452 but cannot call free() or pcre_free() directly.
3453
3454 4. Add "make test" as a synonym for "make check". Corrected some comments in
3455 the Makefile.
3456
3457 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3458 Makefile.
3459
3460 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3461 command called pgrep for grepping around the active processes.
3462
3463 7. Added the beginnings of support for UTF-8 character strings.
3464
3465 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3466 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3467 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3468 out for the ar command.)
3469
3470
3471 Version 3.2 12-May-00
3472 ---------------------
3473
3474 This is purely a bug fixing release.
3475
3476 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3477 of ZA. This was just one example of several cases that could provoke this bug,
3478 which was introduced by change 9 of version 2.00. The code for breaking
3479 infinite loops after an iteration that matches an empty string was't working
3480 correctly.
3481
3482 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3483 when matched against abbab (for example). After matching an empty string, it
3484 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3485 caused it to match further down the string than it should.
3486
3487 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3488 was there because it doesn't seem to be needed, and it causes trouble on some
3489 systems, as it is not a Standard C header. It has been removed.
3490
3491 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3492 were reported on the Macintosh. The changes were from
3493
3494 while ((c = *(++ptr)) != 0 && c != '\n');
3495 to
3496 while ((c = *(++ptr)) != 0 && c != '\n') ;
3497
3498 Totally extraordinary, but if that's what it takes...
3499
3500 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3501 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3502 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3503 assumes the way PCRE uses memmove() (always moving upwards).
3504
3505 6. PCRE is being used in one environment where strchr() is not available. There
3506 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3507 faster code anyway.
3508
3509
3510 Version 3.1 09-Feb-00
3511 ---------------------
3512
3513 The only change in this release is the fixing of some bugs in Makefile.in for
3514 the "install" target:
3515
3516 (1) It was failing to install pcreposix.h.
3517
3518 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3519
3520
3521 Version 3.0 01-Feb-00
3522 ---------------------
3523
3524 1. Add support for the /+ modifier to perltest (to output $` like it does in
3525 pcretest).
3526
3527 2. Add support for the /g modifier to perltest.
3528
3529 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3530 matches null strings.
3531
3532 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3533 pattern. Perl treats empty patterns specially - it reuses the most recent
3534 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3535 effect.
3536
3537 5. The POSIX interface was broken in that it was just handing over the POSIX
3538 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3539 required a bigger vector, with some working space on the end. This means that
3540 the POSIX wrapper now has to get and free some memory, and copy the results.
3541
3542 6. Added some simple autoconf support, placing the test data and the
3543 documentation in separate directories, re-organizing some of the
3544 information files, and making it build pcre-config (a GNU standard). Also added
3545 libtool support for building PCRE as a shared library, which is now the
3546 default.
3547
3548 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3549 09 are not valid octal constants. Single digits will be used for minor values
3550 less than 10.
3551
3552 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3553 existing programs that set these in the POSIX interface can use PCRE without
3554 modification.
3555
3556 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3557 return all that pcre_info() returns, plus additional data. The pcre_info()
3558 function is retained for compatibility, but is considered to be obsolete.
3559
3560 10. Added experimental recursion feature (?R) to handle one common case that
3561 Perl 5.6 will be able to do with (?p{...}).
3562
3563 11. Added support for POSIX character classes like [:alpha:], which Perl is
3564 adopting.
3565
3566
3567 Version 2.08 31-Aug-99
3568 ----------------------
3569
3570 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3571 trying to match at the startoffset position, but instead was moving forward to
3572 the next newline as if a previous match had failed.
3573
3574 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3575 and could get into a loop if a null string was matched other than at the start
3576 of the subject.
3577
3578 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3579 be distinguished at compile time, and for completeness also added PCRE_DATE.
3580
3581 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3582 in GnuWin32 environments.
3583
3584
3585 Version 2.07 29-Jul-99
3586 ----------------------
3587
3588 1. The documentation is now supplied in plain text form and HTML as well as in
3589 the form of man page sources.
3590
3591 2. C++ compilers don't like assigning (void *) values to other pointer types.
3592 In particular this affects malloc(). Although there is no problem in Standard
3593 C, I've put in casts to keep C++ compilers happy.
3594
3595 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3596 should be (const char *).
3597
3598 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3599 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3600 However, I haven't made this a standard facility. The documentation doesn't
3601 mention it, and the Makefile doesn't support it.
3602
3603 5. The Makefile now contains an "install" target, with editable destinations at
3604 the top of the file. The pcretest program is not installed.
3605
3606 6. pgrep -V now gives the PCRE version number and date.
3607
3608 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3609 causing the entire string to be ignored, instead of just the last character.
3610
3611 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3612 non-matching string, it can take a very, very long time, even for strings of
3613 quite modest length, because of the nested recursion. PCRE now does better in
3614 some of these cases. It does this by remembering the last required literal
3615 character in the pattern, and pre-searching the subject to ensure it is present
3616 before running the real match. In other words, it applies a heuristic to detect
3617 some types of certain failure quickly, and in the above example, if presented
3618 with a string that has no trailing " it gives "no match" very quickly.
3619
3620 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3621 other alternatives are tried instead.
3622
3623
3624 Version 2.06 09-Jun-99
3625 ----------------------
3626
3627 1. Change pcretest's output for amount of store used to show just the code
3628 space, because the remainder (the data block) varies in size between 32-bit and
3629 64-bit systems.
3630
3631 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3632 start matching at. This allows lookbehinds to work when searching for multiple
3633 occurrences in a string.
3634
3635 3. Added additional options to pcretest for testing multiple occurrences:
3636
3637 /+ outputs the rest of the string that follows a match
3638 /g loops for multiple occurrences, using the new startoffset argument
3639 /G loops for multiple occurrences by passing an incremented pointer
3640
3641 4. PCRE wasn't doing the "first character" optimization for patterns starting
3642 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3643 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3644 the letter 'x'. On long subject strings, this gives a significant speed-up.
3645
3646
3647 Version 2.05 21-Apr-99
3648 ----------------------
3649
3650 1. Changed the type of magic_number from int to long int so that it works
3651 properly on 16-bit systems.
3652
3653 2. Fixed a bug which caused patterns starting with .* not to work correctly
3654 when the subject string contained newline characters. PCRE was assuming
3655 anchoring for such patterns in all cases, which is not correct because .* will
3656 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3657 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3658 must be retried after every newline in the subject.
3659
3660
3661 Version 2.04 18-Feb-99
3662 ----------------------
3663
3664 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3665 computation of the store needed to hold the pattern was incorrect (too large).
3666 If such patterns were nested a few deep, this could multiply and become a real
3667 problem.
3668
3669 2. Added /M option to pcretest to show the memory requirement of a specific
3670 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3671
3672 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3673 compiled in such a way that the backtracking after subsequent failure was
3674 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3675 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3676
3677
3678 Version 2.03 02-Feb-99
3679 ----------------------
3680
3681 1. Fixed typo and small mistake in man page.
3682
3683 2. Added 4th condition (GPL supersedes if conflict) and created separate
3684 LICENCE file containing the conditions.
3685
3686 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3687 Perl, that is the internal \ allows the delimiter to be included in the
3688 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3689 the final delimiter, add \ to the end of the pattern (to test the error).
3690
3691 4. Added the convenience functions for extracting substrings after a successful
3692 match. Updated pcretest to make it able to test these functions.
3693
3694
3695 Version 2.02 14-Jan-99
3696 ----------------------
3697
3698 1. Initialized the working variables associated with each extraction so that
3699 their saving and restoring doesn't refer to uninitialized store.
3700
3701 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3702 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3703 fix the problem.
3704
3705 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3706 calls, and wasn't printing the correct value for compiling calls. Increased the
3707 default value of LOOPREPEAT, and the number of significant figures in the
3708 times.
3709
3710 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3711
3712 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3713 a building problem on Windows NT with a FAT file system.
3714
3715
3716 Version 2.01 21-Oct-98
3717 ----------------------
3718
3719 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3720 to character tables built by pcre_maketables() in the current locale. If NULL
3721 is passed, the default tables are used.
3722
3723
3724 Version 2.00 24-Sep-98
3725 ----------------------
3726
3727 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3728 it any more.
3729
3730 2. Allow quantification of (?>) groups, and make it work correctly.
3731
3732 3. The first character computation wasn't working for (?>) groups.
3733
3734 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3735 end of the subject) and add 5.005's \z, which really does match only at the
3736 very end of the subject.
3737
3738 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3739
3740 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3741 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3742 localized options. All options to pcre_study() were also removed.
3743
3744 7. Add other new features from 5.005:
3745
3746 $(?<= positive lookbehind
3747 $(?<! negative lookbehind
3748 (?imsx-imsx) added the unsetting capability
3749 such a setting is global if at outer level; local otherwise
3750 (?imsx-imsx:) non-capturing groups with option setting
3751 (?(cond)re|re) conditional pattern matching
3752
3753 A backreference to itself in a repeated group matches the previous
3754 captured string.
3755
3756 8. General tidying up of studying (both automatic and via "study")
3757 consequential on the addition of new assertions.
3758
3759 9. As in 5.005, unlimited repeated groups that could match an empty substring
3760 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3761 runtime if any iteration does actually match an empty substring.
3762
3763 10. Include the RunTest script in the distribution.
3764
3765 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3766 discrepancies, some of which were old and were also with respect to 5.004. They
3767 have now been fixed.
3768
3769
3770 Version 1.09 28-Apr-98
3771 ----------------------
3772
3773 1. A negated single character class followed by a quantifier with a minimum
3774 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3775 program crashes, or just wrong answers. This did not apply to negated classes
3776 containing more than one character, or to minima other than one.
3777
3778
3779 Version 1.08 27-Mar-98
3780 ----------------------
3781
3782 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3783
3784 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3785 latter must appear before anything that relies on it in the pattern.
3786
3787
3788 Version 1.07 16-Feb-98
3789 ----------------------
3790
3791 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3792 repeat of a potentially empty string).
3793
3794
3795 Version 1.06 23-Jan-98
3796 ----------------------
3797
3798 1. Added Markus Oberhumer's little patches for C++.
3799
3800 2. Literal strings longer than 255 characters were broken.
3801
3802
3803 Version 1.05 23-Dec-97
3804 ----------------------
3805
3806 1. Negated character classes containing more than one character were failing if
3807 PCRE_CASELESS was set at run time.
3808
3809
3810 Version 1.04 19-Dec-97
3811 ----------------------
3812
3813 1. Corrected the man page, where some "const" qualifiers had been omitted.
3814
3815 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3816 input syntax.
3817
3818 3. Fixed memory leak which occurred when a regex with back references was
3819 matched with an offsets vector that wasn't big enough. The temporary memory
3820 that is used in this case wasn't being freed if the match failed.
3821
3822 4. Tidied pcretest to ensure it frees memory that it gets.
3823
3824 5. Temporary memory was being obtained in the case where the passed offsets
3825 vector was exactly big enough.
3826
3827 6. Corrected definition of offsetof() from change 5 below.
3828
3829 7. I had screwed up change 6 below and broken the rules for the use of
3830 setjmp(). Now fixed.
3831
3832
3833 Version 1.03 18-Dec-97
3834 ----------------------
3835
3836 1. A erroneous regex with a missing opening parenthesis was correctly
3837 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3838 on some systems.
3839
3840 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3841 it was reported that one broken compiler failed on the former because "code" is
3842 also an independent variable.
3843
3844 3. The erroneous regex a[]b caused an array overrun reference.
3845
3846 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3847 fail on data ending with that character. (It was going on too far, and checking
3848 the next character, typically a binary zero.) This was specific to the
3849 optimized code for single-character negative classes.
3850
3851 5. Added a contributed patch from the TIN world which does the following:
3852
3853 + Add an undef for memmove, in case the the system defines a macro for it.
3854
3855 + Add a definition of offsetof(), in case there isn't one. (I don't know
3856 the reason behind this - offsetof() is part of the ANSI standard - but
3857 it does no harm).
3858
3859 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3860 most of the places where whitespace preceded '#'. I have given up and
3861 allowed the remaining 2 cases to be at the margin.
3862
3863 + Rename some variables in pcre to eliminate shadowing. This seems very
3864 pedantic, but does no harm, of course.
3865
3866 6. Moved the call to setjmp() into its own function, to get rid of warnings
3867 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3868
3869 7. Constructs such as \d{8,} were compiling into the equivalent of
3870 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3871 outcome, but in this particular case used more store than had been allocated,
3872 which caused the bug to be discovered because it threw up an internal error.
3873
3874 8. The debugging code in both pcre and pcretest for outputting the compiled
3875 form of a regex was going wrong in the case of back references followed by
3876 curly-bracketed repeats.
3877
3878
3879 Version 1.02 12-Dec-97
3880 ----------------------
3881
3882 1. Typos in pcre.3 and comments in the source fixed.
3883
3884 2. Applied a contributed patch to get rid of places where it used to remove
3885 'const' from variables, and fixed some signed/unsigned and uninitialized
3886 variable warnings.
3887
3888 3. Added the "runtest" target to Makefile.
3889
3890 4. Set default compiler flag to -O2 rather than just -O.
3891
3892
3893 Version 1.01 19-Nov-97
3894 ----------------------
3895
3896 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3897 like /([ab]*)*/, that is, for classes with more than one character in them.
3898
3899 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3900 as /((?>a*))*/ (a PCRE_EXTRA facility).
3901
3902
3903 Version 1.00 18-Nov-97
3904 ----------------------
3905
3906 1. Added compile-time macros to support systems such as SunOS4 which don't have
3907 memmove() or strerror() but have other things that can be used instead.
3908
3909 2. Arranged that "make clean" removes the executables.
3910
3911
3912 Version 0.99 27-Oct-97
3913 ----------------------
3914
3915 1. Fixed bug in code for optimizing classes with only one character. It was
3916 initializing a 32-byte map regardless, which could cause it to run off the end
3917 of the memory it had got.
3918
3919 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3920
3921
3922 Version 0.98 22-Oct-97
3923 ----------------------
3924
3925 1. Fixed bug in code for handling temporary memory usage when there are more
3926 back references than supplied space in the ovector. This could cause segfaults.
3927
3928
3929 Version 0.97 21-Oct-97
3930 ----------------------
3931
3932 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3933
3934 2. Optimized negated single characters not to use a bit map.
3935
3936 3. Brought error texts together as macro definitions; clarified some of them;
3937 fixed one that was wrong - it said "range out of order" when it meant "invalid
3938 escape sequence".
3939
3940 4. Changed some char * arguments to const char *.
3941
3942 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3943
3944 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3945 pcretest.
3946
3947
3948 Version 0.96 16-Oct-97
3949 ----------------------
3950
3951 1. Added a simple "pgrep" utility to the distribution.
3952
3953 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3954 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3955 where "ddd" means "one or more decimal digits".
3956
3957 3. Fixed serious bug. If a pattern had a back reference, but the call to
3958 pcre_exec() didn't supply a large enough ovector to record the related
3959 identifying subpattern, the match always failed. PCRE now remembers the number
3960 of the largest back reference, and gets some temporary memory in which to save
3961 the offsets during matching if necessary, in order to ensure that
3962 backreferences always work.
3963
3964 4. Increased the compatibility with Perl in a number of ways:
3965
3966 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3967 to request this handling. The option can be set at compile or exec time.
3968
3969 (b) $ matches before a terminating newline by default; an option
3970 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3971 mode). The option can be set at compile or exec time.
3972
3973 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3974 the same as Perl's. If the decimal number it represents is less than 10
3975 or there aren't that many previous left capturing parentheses, an octal
3976 escape is read. Inside a character class, it's always an octal escape,
3977 even if it is a single digit.
3978
3979 (d) An escaped but undefined alphabetic character is taken as a literal,
3980 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3981 escapes.
3982
3983 (e) {0} is now permitted. (The previous item is removed from the compiled
3984 pattern).
3985
3986 5. Changed all the names of code files so that the basic parts are no longer
3987 than 10 characters, and abolished the teeny "globals.c" file.
3988
3989 6. Changed the handling of character classes; they are now done with a 32-byte
3990 bit map always.
3991
3992 7. Added the -d and /D options to pcretest to make it possible to look at the
3993 internals of compilation without having to recompile pcre.
3994
3995
3996 Version 0.95 23-Sep-97
3997 ----------------------
3998
3999 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4000 \x20 at the start of a run of normal characters. These were being treated as
4001 real characters, instead of the source characters being re-checked.
4002
4003
4004 Version 0.94 18-Sep-97
4005 ----------------------
4006
4007 1. The functions are now thread-safe, with the caveat that the global variables
4008 containing pointers to malloc() and free() or alternative functions are the
4009 same for all threads.
4010
4011 2. Get pcre_study() to generate a bitmap of initial characters for non-
4012 anchored patterns when this is possible, and use it if passed to pcre_exec().
4013
4014
4015 Version 0.93 15-Sep-97
4016 ----------------------
4017
4018 1. /(b)|(:+)/ was computing an incorrect first character.
4019
4020 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4021 but not actually doing anything yet.
4022
4023 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4024 as Perl does (e.g. [-az] or [az-]).
4025
4026 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4027 all possible positions.
4028
4029 5. Split up into different modules to avoid including unneeded functions in a
4030 compiled binary. However, compile and exec are still in one module. The "study"
4031 function is split off.
4032
4033 6. The character tables are now in a separate module whose source is generated
4034 by an auxiliary program - but can then be edited by hand if required. There are
4035 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4036 toupper() in the code.
4037
4038 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4039 make them global. Abolish the function for setting them, as the caller can now
4040 set them directly.
4041
4042
4043 Version 0.92 11-Sep-97
4044 ----------------------
4045
4046 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4047 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4048
4049 2. Caseless matching was not working in character classes if the characters in
4050 the pattern were in upper case.
4051
4052 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4053
4054 4. Make PCRE_ANCHORED public and accept as a compile option.
4055
4056 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4057 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4058 pass them.
4059
4060 6. Give an error if bad option bits passed at compile or run time.
4061
4062 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4063 pcretest to cause it to pass that flag.
4064
4065 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4066 options, and the first character, if set.
4067
4068 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4069
4070
4071 Version 0.91 10-Sep-97
4072 ----------------------
4073
4074 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4075 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4076
4077 2. PCRE was looping on encountering an indefinitely repeated back reference to
4078 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4079 Perl does - treats the match as successful.
4080
4081 ****

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5