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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 523 - (show annotations) (download)
Sun May 23 18:48:54 2010 UTC (3 years, 10 months ago) by ph10
File size: 187464 byte(s)
Fix pcre_study() bug for \p{Nd}* type items at pattern start.

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

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