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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 548 - (show annotations) (download)
Fri Jun 25 14:42:00 2010 UTC (4 years, 3 months ago) by ph10
File size: 189526 byte(s)
Final updates for 8.10 release.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12