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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 529 - (show annotations)
Mon May 31 17:28:08 2010 UTC (5 years ago) by ph10
File size: 187962 byte(s)
Error occurred while calculating annotation data.
Fix crash for property test in non-UTF-8 mode.
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 14. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
59 possible starting bytes for non-anchored patterns.
60
61 15. The "auto-possessify" feature of pcre_compile() now recognizes \R.
62
63 16. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
64 input, it could crash or give wrong results if characters with values
65 greater than 0xc0 were present in the subject string. (Detail: it assumed
66 UTF-8 input when processing these items.)
67
68
69
70 Version 8.02 19-Mar-2010
71 ------------------------
72
73 1. The Unicode data tables have been updated to Unicode 5.2.0.
74
75 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
76 configured.
77
78 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
79 original author of that file, following a query about its status.
80
81 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
82 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
83
84 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
85 quantifier applied to a forward-referencing subroutine call, could compile
86 incorrect code or give the error "internal error: previously-checked
87 referenced subpattern not found".
88
89 6. Both MS Visual Studio and Symbian OS have problems with initializing
90 variables to point to external functions. For these systems, therefore,
91 pcre_malloc etc. are now initialized to local functions that call the
92 relevant global functions.
93
94 7. There were two entries missing in the vectors called coptable and poptable
95 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
96 I've fixed the data, and added a kludgy way of testing at compile time that
97 the lengths are correct (equal to the number of opcodes).
98
99 8. Following on from 7, I added a similar kludge to check the length of the
100 eint vector in pcreposix.c.
101
102 9. Error texts for pcre_compile() are held as one long string to avoid too
103 much relocation at load time. To find a text, the string is searched,
104 counting zeros. There was no check for running off the end of the string,
105 which could happen if a new error number was added without updating the
106 string.
107
108 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
109
110 11. \K was not working if it appeared in an atomic group or in a group that
111 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
112 \K is "not well defined" if used in an assertion. PCRE now accepts it if
113 the assertion is positive, but not if it is negative.
114
115 12. Change 11 fortuitously reduced the size of the stack frame used in the
116 "match()" function of pcre_exec.c by one pointer. Forthcoming
117 implementation of support for (*MARK) will need an extra pointer on the
118 stack; I have reserved it now, so that the stack frame size does not
119 decrease.
120
121 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
122 item in branch that calls a recursion is a subroutine call - as in the
123 second branch in the above example - was incorrectly given the compile-
124 time error "recursive call could loop indefinitely" because pcre_compile()
125 was not correctly checking the subroutine for matching a non-empty string.
126
127 14. The checks for overrunning compiling workspace could trigger after an
128 overrun had occurred. This is a "should never occur" error, but it can be
129 triggered by pathological patterns such as hundreds of nested parentheses.
130 The checks now trigger 100 bytes before the end of the workspace.
131
132 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
133
134
135 Version 8.01 19-Jan-2010
136 ------------------------
137
138 1. If a pattern contained a conditional subpattern with only one branch (in
139 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
140 computed the wrong minimum data length (which is of course zero for such
141 subpatterns). This could cause incorrect "no match" results.
142
143 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
144 the pattern is reset in the first branch, pcre_compile() failed with
145 "internal error: code overflow at offset...". This happened only when
146 the reset was to the original external option setting. (An optimization
147 abstracts leading options settings into an external setting, which was the
148 cause of this.)
149
150 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
151 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
152 assertion pattern did not match (meaning that the assertion was true), it
153 was incorrectly treated as false if the SKIP had been reached during the
154 matching. This also applied to assertions used as conditions.
155
156 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
157 assertion subpattern, including such a pattern used as a condition,
158 unpredictable results occurred, instead of the error return
159 PCRE_ERROR_DFA_UITEM.
160
161 5. The C++ GlobalReplace function was not working like Perl for the special
162 situation when an empty string is matched. It now does the fancy magic
163 stuff that is necessary.
164
165 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
166 removed. (These were left over from very, very early versions of PCRE.)
167
168 7. Some cosmetic changes to the code to make life easier when compiling it
169 as part of something else:
170
171 (a) Change DEBUG to PCRE_DEBUG.
172
173 (b) In pcre_compile(), rename the member of the "branch_chain" structure
174 called "current" as "current_branch", to prevent a collision with the
175 Linux macro when compiled as a kernel module.
176
177 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
178 prevent a collision with the Linux macro when compiled as a kernel
179 module.
180
181 8. In pcre_compile() there are some checks for integer overflows that used to
182 cast potentially large values to (double). This has been changed to that
183 when building, a check for int64_t is made, and if it is found, it is used
184 instead, thus avoiding the use of floating point arithmetic. (There is no
185 other use of FP in PCRE.) If int64_t is not found, the fallback is to
186 double.
187
188 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
189 2005 (difference between two addresses compared to an unsigned value).
190
191 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
192 custom one, because of the following reported problem in Windows:
193
194 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
195 under Win32.
196 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
197 therefore missing the function definition.
198 - The compiler thus generates a "C" signature for the test function.
199 - The linker fails to find the "C" function.
200 - PCRE fails to configure if asked to do so against libbz2.
201
202 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
203 messages were output:
204
205 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
206 rerunning libtoolize, to keep the correct libtool macros in-tree.
207 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
208
209 I have done both of these things.
210
211 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
212 most of the time, it *can* run out if it is given a pattern that contains a
213 runaway infinite recursion. I updated the discussion in the pcrestack man
214 page.
215
216 13. Now that we have gone to the x.xx style of version numbers, the minor
217 version may start with zero. Using 08 or 09 is a bad idea because users
218 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
219 interpreted as invalid octal numbers. I've updated the previous comment in
220 configure.ac, and also added a check that gives an error if 08 or 09 are
221 used.
222
223 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
224 causing partial matching to fail when the end of the subject matched \W
225 in a UTF-8 pattern where \W was quantified with a minimum of 3.
226
227 15. There were some discrepancies between the declarations in pcre_internal.h
228 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
229 their definitions. The declarations used "const uschar *" and the
230 definitions used USPTR. Even though USPTR is normally defined as "const
231 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
232 reported that: "This difference in casting confuses some C++ compilers, for
233 example, SunCC recognizes above declarations as different functions and
234 generates broken code for hbpcre." I have changed the declarations to use
235 USPTR.
236
237 16. GNU libtool is named differently on some systems. The autogen.sh script now
238 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
239 (FreeBSD).
240
241 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
242 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
243 comment: "Figure out how to create a longlong from a string: strtoll and
244 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
245 instance, but it only takes 2 args instead of 3!"
246
247 18. A subtle bug concerned with back references has been fixed by a change of
248 specification, with a corresponding code fix. A pattern such as
249 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
250 refers, was giving matches when it shouldn't. For example, xa=xaaa would
251 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
252 same bug. Such groups have to be quantified to be useful, or contained
253 inside another quantified group. (If there's no repetition, the reference
254 can never match.) The problem arises because, having left the group and
255 moved on to the rest of the pattern, a later failure that backtracks into
256 the group uses the captured value from the final iteration of the group
257 rather than the correct earlier one. I have fixed this in PCRE by forcing
258 any group that contains a reference to itself to be an atomic group; that
259 is, there cannot be any backtracking into it once it has completed. This is
260 similar to recursive and subroutine calls.
261
262
263 Version 8.00 19-Oct-09
264 ----------------------
265
266 1. The table for translating pcre_compile() error codes into POSIX error codes
267 was out-of-date, and there was no check on the pcre_compile() error code
268 being within the table. This could lead to an OK return being given in
269 error.
270
271 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
272 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
273 in a Windows environment.
274
275 3. The pcregrep --count option prints the count for each file even when it is
276 zero, as does GNU grep. However, pcregrep was also printing all files when
277 --files-with-matches was added. Now, when both options are given, it prints
278 counts only for those files that have at least one match. (GNU grep just
279 prints the file name in this circumstance, but including the count seems
280 more useful - otherwise, why use --count?) Also ensured that the
281 combination -clh just lists non-zero counts, with no names.
282
283 4. The long form of the pcregrep -F option was incorrectly implemented as
284 --fixed_strings instead of --fixed-strings. This is an incompatible change,
285 but it seems right to fix it, and I didn't think it was worth preserving
286 the old behaviour.
287
288 5. The command line items --regex=pattern and --regexp=pattern were not
289 recognized by pcregrep, which required --regex pattern or --regexp pattern
290 (with a space rather than an '='). The man page documented the '=' forms,
291 which are compatible with GNU grep; these now work.
292
293 6. No libpcreposix.pc file was created for pkg-config; there was just
294 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
295
296 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
297 when UCP support is not needed, by modifying the Python script that
298 generates it from Unicode data files. This should not matter if the module
299 is correctly used as a library, but I received one complaint about 50K of
300 unwanted data. My guess is that the person linked everything into his
301 program rather than using a library. Anyway, it does no harm.
302
303 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
304 was a minimum greater than 1 for a wide character in a possessive
305 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
306 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
307 character. Chaos in the form of incorrect output or a compiling loop could
308 result.
309
310 9. The restrictions on what a pattern can contain when partial matching is
311 requested for pcre_exec() have been removed. All patterns can now be
312 partially matched by this function. In addition, if there are at least two
313 slots in the offset vector, the offset of the earliest inspected character
314 for the match and the offset of the end of the subject are set in them when
315 PCRE_ERROR_PARTIAL is returned.
316
317 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
318 synonymous with PCRE_PARTIAL, for backwards compatibility, and
319 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
320 and may be more useful for multi-segment matching.
321
322 11. Partial matching with pcre_exec() is now more intuitive. A partial match
323 used to be given if ever the end of the subject was reached; now it is
324 given only if matching could not proceed because another character was
325 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
326 string "Z", which now yields "no match" instead of "partial match". In the
327 case of pcre_dfa_exec(), "no match" is given if every matching path for the
328 final character ended with (*FAIL).
329
330 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
331 if the pattern had a "must contain" character that was already found in the
332 earlier partial match, unless partial matching was again requested. For
333 example, with the pattern /dog.(body)?/, the "must contain" character is
334 "g". If the first part-match was for the string "dog", restarting with
335 "sbody" failed. This bug has been fixed.
336
337 13. The string returned by pcre_dfa_exec() after a partial match has been
338 changed so that it starts at the first inspected character rather than the
339 first character of the match. This makes a difference only if the pattern
340 starts with a lookbehind assertion or \b or \B (\K is not supported by
341 pcre_dfa_exec()). It's an incompatible change, but it makes the two
342 matching functions compatible, and I think it's the right thing to do.
343
344 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
345 so that the demonstration program is easily available in environments where
346 PCRE has not been installed from source.
347
348 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
349 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
350 library.
351
352 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
353 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
354 is not the first non-POSIX option to be added. Clearly some people find
355 these options useful.
356
357 17. If a caller to the POSIX matching function regexec() passes a non-zero
358 value for nmatch with a NULL value for pmatch, the value of
359 nmatch is forced to zero.
360
361 18. RunGrepTest did not have a test for the availability of the -u option of
362 the diff command, as RunTest does. It now checks in the same way as
363 RunTest, and also checks for the -b option.
364
365 19. If an odd number of negated classes containing just a single character
366 interposed, within parentheses, between a forward reference to a named
367 subpattern and the definition of the subpattern, compilation crashed with
368 an internal error, complaining that it could not find the referenced
369 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
370 [The bug was that it was starting one character too far in when skipping
371 over the character class, thus treating the ] as data rather than
372 terminating the class. This meant it could skip too much.]
373
374 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
375 /g option in pcretest when the pattern contains \K, which makes it possible
376 to have an empty string match not at the start, even when the pattern is
377 anchored. Updated pcretest and pcredemo to use this option.
378
379 21. If the maximum number of capturing subpatterns in a recursion was greater
380 than the maximum at the outer level, the higher number was returned, but
381 with unset values at the outer level. The correct (outer level) value is
382 now given.
383
384 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
385 PCRE did not set those parentheses (unlike Perl). I have now found a way to
386 make it do so. The string so far is captured, making this feature
387 compatible with Perl.
388
389 23. The tests have been re-organized, adding tests 11 and 12, to make it
390 possible to check the Perl 5.10 features against Perl 5.10.
391
392 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
393 pattern matches a fixed length string. PCRE did not allow this; now it
394 does. Neither allows recursion.
395
396 25. I finally figured out how to implement a request to provide the minimum
397 length of subject string that was needed in order to match a given pattern.
398 (It was back references and recursion that I had previously got hung up
399 on.) This code has now been added to pcre_study(); it finds a lower bound
400 to the length of subject needed. It is not necessarily the greatest lower
401 bound, but using it to avoid searching strings that are too short does give
402 some useful speed-ups. The value is available to calling programs via
403 pcre_fullinfo().
404
405 26. While implementing 25, I discovered to my embarrassment that pcretest had
406 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
407 study optimizations had never been tested with that matching function.
408 Oops. What is worse, even when it was passed study data, there was a bug in
409 pcre_dfa_exec() that meant it never actually used it. Double oops. There
410 were also very few tests of studied patterns with pcre_dfa_exec().
411
412 27. If (?| is used to create subpatterns with duplicate numbers, they are now
413 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
414 on the other side of the coin, they are no longer allowed to have different
415 names, because these cannot be distinguished in PCRE, and this has caused
416 confusion. (This is a difference from Perl.)
417
418 28. When duplicate subpattern names are present (necessarily with different
419 numbers, as required by 27 above), and a test is made by name in a
420 conditional pattern, either for a subpattern having been matched, or for
421 recursion in such a pattern, all the associated numbered subpatterns are
422 tested, and the overall condition is true if the condition is true for any
423 one of them. This is the way Perl works, and is also more like the way
424 testing by number works.
425
426
427 Version 7.9 11-Apr-09
428 ---------------------
429
430 1. When building with support for bzlib/zlib (pcregrep) and/or readline
431 (pcretest), all targets were linked against these libraries. This included
432 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
433 libraries. This caused unwanted dependencies to be created. This problem
434 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
435 pcretest is linked with readline.
436
437 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
438 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
439 moved outside it again, because FALSE and TRUE are already defined in AIX,
440 but BOOL is not.
441
442 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
443 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
444
445 4. The pcregrep documentation said spaces were inserted as well as colons (or
446 hyphens) following file names and line numbers when outputting matching
447 lines. This is not true; no spaces are inserted. I have also clarified the
448 wording for the --colour (or --color) option.
449
450 5. In pcregrep, when --colour was used with -o, the list of matching strings
451 was not coloured; this is different to GNU grep, so I have changed it to be
452 the same.
453
454 6. When --colo(u)r was used in pcregrep, only the first matching substring in
455 each matching line was coloured. Now it goes on to look for further matches
456 of any of the test patterns, which is the same behaviour as GNU grep.
457
458 7. A pattern that could match an empty string could cause pcregrep to loop; it
459 doesn't make sense to accept an empty string match in pcregrep, so I have
460 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
461 seems to be how GNU grep behaves.
462
463 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
464 start or after a newline", because the conditional assertion was not being
465 correctly handled. The rule now is that both the assertion and what follows
466 in the first alternative must satisfy the test.
467
468 9. If auto-callout was enabled in a pattern with a conditional group whose
469 condition was an assertion, PCRE could crash during matching, both with
470 pcre_exec() and pcre_dfa_exec().
471
472 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
473 used for matching.
474
475 11. Unicode property support in character classes was not working for
476 characters (bytes) greater than 127 when not in UTF-8 mode.
477
478 12. Added the -M command line option to pcretest.
479
480 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
481
482 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
483
484 16. Added comments and documentation about mis-use of no_arg in the C++
485 wrapper.
486
487 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
488 from Martin Jerabek that uses macro names for all relevant character and
489 string constants.
490
491 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
492 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
493 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
494 these, but not everybody uses configure.
495
496 19. A conditional group that had only one branch was not being correctly
497 recognized as an item that could match an empty string. This meant that an
498 enclosing group might also not be so recognized, causing infinite looping
499 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
500 with the subject "ab", where knowledge that the repeated group can match
501 nothing is needed in order to break the loop.
502
503 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
504 exec(), but without supplying a callout function, matching went wrong.
505
506 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
507 leak if the size of the offset vector was greater than 30. When the vector
508 is smaller, the saved offsets during recursion go onto a local stack
509 vector, but for larger vectors malloc() is used. It was failing to free
510 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
511 error, in fact).
512
513 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
514 heapframe that is used only when UTF-8 support is enabled. This caused no
515 problem, but was untidy.
516
517 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
518 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
519 included within another project.
520
521 24. Steven Van Ingelgem's patches to add more options to the CMake support,
522 slightly modified by me:
523
524 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
525 not building pcregrep.
526
527 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
528 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
529
530 25. Forward references, both numeric and by name, in patterns that made use of
531 duplicate group numbers, could behave incorrectly or give incorrect errors,
532 because when scanning forward to find the reference group, PCRE was not
533 taking into account the duplicate group numbers. A pattern such as
534 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
535
536 26. Changed a few more instances of "const unsigned char *" to USPTR, making
537 the feature of a custom pointer more persuasive (as requested by a user).
538
539 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
540 pcretest.c, inside #ifndefs, because it seems they are sometimes already
541 pre-defined.
542
543 28. Added support for (*UTF8) at the start of a pattern.
544
545 29. Arrange for flags added by the "release type" setting in CMake to be shown
546 in the configuration summary.
547
548
549 Version 7.8 05-Sep-08
550 ---------------------
551
552 1. Replaced UCP searching code with optimized version as implemented for Ad
553 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
554 stage table and inline lookup instead of a function, giving speed ups of 2
555 to 5 times on some simple patterns that I tested. Permission was given to
556 distribute the MultiStage2.py script that generates the tables (it's not in
557 the tarball, but is in the Subversion repository).
558
559 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
560 scripts.
561
562 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
563 a group with a zero qualifier. The result of the study could be incorrect,
564 or the function might crash, depending on the pattern.
565
566 4. Caseless matching was not working for non-ASCII characters in back
567 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
568 It now works when Unicode Property Support is available.
569
570 5. In pcretest, an escape such as \x{de} in the data was always generating
571 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
572 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
573 truncation.
574
575 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
576
577 7. Added two (int) casts to pcregrep when printing the difference of two
578 pointers, in case they are 64-bit values.
579
580 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
581 test 2 if it fails.
582
583 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
584 and a #define of that name to empty if it is not externally set. This is to
585 allow users of MSVC to set it if necessary.
586
587 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
588 the convenience functions in the pcre_get.c source file.
589
590 11. An option change at the start of a pattern that had top-level alternatives
591 could cause overwriting and/or a crash. This command provoked a crash in
592 some environments:
593
594 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
595
596 This potential security problem was recorded as CVE-2008-2371.
597
598 12. For a pattern where the match had to start at the beginning or immediately
599 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
600 pcre_dfa_exec() could read past the end of the passed subject if there was
601 no match. To help with detecting such bugs (e.g. with valgrind), I modified
602 pcretest so that it places the subject at the end of its malloc-ed buffer.
603
604 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
605 exec() might read past the end of the data buffer in UTF-8 mode.
606
607 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
608 the data contained the byte 0x85 as part of a UTF-8 character within its
609 first line. This applied both to normal and DFA matching.
610
611 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
612 /^[^d]*?$/8 failed to match "abc".
613
614 16. Added a missing copyright notice to pcrecpp_internal.h.
615
616 17. Make it more clear in the documentation that values returned from
617 pcre_exec() in ovector are byte offsets, not character counts.
618
619 18. Tidied a few places to stop certain compilers from issuing warnings.
620
621 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
622 supplied by Stefan Weber. I made a further small update for 7.8 because
623 there is a change of source arrangements: the pcre_searchfuncs.c module is
624 replaced by pcre_ucd.c.
625
626
627 Version 7.7 07-May-08
628 ---------------------
629
630 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
631 a string to a long long, pretend we don't even have a long long." This is
632 done by checking for the strtoq, strtoll, and _strtoi64 functions.
633
634 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
635 pre-7.6 versions, which defined a global no_arg variable instead of putting
636 it in the RE class. (See also #8 below.)
637
638 3. Remove a line of dead code, identified by coverity and reported by Nuno
639 Lopes.
640
641 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
642
643 (1) The include/exclude patterns were being applied to the whole pathnames
644 of files, instead of just to the final components.
645
646 (2) If there was more than one level of directory, the subdirectories were
647 skipped unless they satisfied the include/exclude conditions. This is
648 inconsistent with GNU grep (and could even be seen as contrary to the
649 pcregrep specification - which I improved to make it absolutely clear).
650 The action now is always to scan all levels of directory, and just
651 apply the include/exclude patterns to regular files.
652
653 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
654 --exclude_dir in the tests to avoid scanning .svn directories.
655
656 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
657 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
658 doesn't support NULs in patterns.
659
660 7. Added some missing "const"s to declarations of static tables in
661 pcre_compile.c and pcre_dfa_exec.c.
662
663 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
664 caused by fix #2 above. (Subsequently also a second patch to fix the
665 first patch. And a third patch - this was a messy problem.)
666
667 9. Applied Craig's patch to remove the use of push_back().
668
669 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
670 matching function regexec().
671
672 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
673 which, however, unlike Perl's \g{...}, are subroutine calls, not back
674 references. PCRE supports relative numbers with this syntax (I don't think
675 Oniguruma does).
676
677 12. Previously, a group with a zero repeat such as (...){0} was completely
678 omitted from the compiled regex. However, this means that if the group
679 was called as a subroutine from elsewhere in the pattern, things went wrong
680 (an internal error was given). Such groups are now left in the compiled
681 pattern, with a new opcode that causes them to be skipped at execution
682 time.
683
684 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
685 to the way PCRE behaves:
686
687 (a) A lone ] character is dis-allowed (Perl treats it as data).
688
689 (b) A back reference to an unmatched subpattern matches an empty string
690 (Perl fails the current match path).
691
692 (c) A data ] in a character class must be notated as \] because if the
693 first data character in a class is ], it defines an empty class. (In
694 Perl it is not possible to have an empty class.) The empty class []
695 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
696 The negative empty class [^] matches any one character, independently
697 of the DOTALL setting.
698
699 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
700 non-existent subpattern following a character class starting with ']' and
701 containing () gave an internal compiling error instead of "reference to
702 non-existent subpattern". Fortunately, when the pattern did exist, the
703 compiled code was correct. (When scanning forwards to check for the
704 existencd of the subpattern, it was treating the data ']' as terminating
705 the class, so got the count wrong. When actually compiling, the reference
706 was subsequently set up correctly.)
707
708 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
709 it was being rejected as not supported by pcre_dfa_exec(), even though
710 other assertions are supported. I have made pcre_dfa_exec() support
711 (*FAIL).
712
713 16. The implementation of 13c above involved the invention of a new opcode,
714 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
715 cannot be changed at match time, I realized I could make a small
716 improvement to matching performance by compiling OP_ALLANY instead of
717 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
718 on the OP_ANY path.
719
720 17. Compiling pcretest on Windows with readline support failed without the
721 following two fixes: (1) Make the unistd.h include conditional on
722 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
723
724 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
725 ncurses library to be included for pcretest when ReadLine support is
726 requested, but also to allow for it to be overridden. This patch came from
727 Daniel Bergström.
728
729 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
730 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
731 any errors with the current Unicode tables. Thanks to Peter Kankowski for
732 spotting this.
733
734
735 Version 7.6 28-Jan-08
736 ---------------------
737
738 1. A character class containing a very large number of characters with
739 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
740 overflow.
741
742 2. Patch to cut out the "long long" test in pcrecpp_unittest when
743 HAVE_LONG_LONG is not defined.
744
745 3. Applied Christian Ehrlicher's patch to update the CMake build files to
746 bring them up to date and include new features. This patch includes:
747
748 - Fixed PH's badly added libz and libbz2 support.
749 - Fixed a problem with static linking.
750 - Added pcredemo. [But later removed - see 7 below.]
751 - Fixed dftables problem and added an option.
752 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
753 HAVE_LONG_LONG.
754 - Added readline support for pcretest.
755 - Added an listing of the option settings after cmake has run.
756
757 4. A user submitted a patch to Makefile that makes it easy to create
758 "pcre.dll" under mingw when using Configure/Make. I added stuff to
759 Makefile.am that cause it to include this special target, without
760 affecting anything else. Note that the same mingw target plus all
761 the other distribution libraries and programs are now supported
762 when configuring with CMake (see 6 below) instead of with
763 Configure/Make.
764
765 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
766 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
767 exported in the Windows port". It has not yet been confirmed that the patch
768 solves the problem, but it does no harm.
769
770 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
771 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
772 with CMake, and also correct the comment about stack recursion.
773
774 7. Remove the automatic building of pcredemo from the ./configure system and
775 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
776 of a program that users should build themselves after PCRE is installed, so
777 building it automatically is not really right. What is more, it gave
778 trouble in some build environments.
779
780 8. Further tidies to CMakeLists.txt from Sheri and Christian.
781
782
783 Version 7.5 10-Jan-08
784 ---------------------
785
786 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
787 values in parens when parsing an RE using the C++ wrapper."
788
789 2. Negative specials like \S did not work in character classes in UTF-8 mode.
790 Characters greater than 255 were excluded from the class instead of being
791 included.
792
793 3. The same bug as (2) above applied to negated POSIX classes such as
794 [:^space:].
795
796 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
797 defined or documented. It seems to have been a typo for PCRE_STATIC, so
798 I have changed it.
799
800 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
801 first named subpattern) and a construct such as (?&a) would reference the
802 first named subpattern whose name started with "a" (in other words, the
803 length check was missing). Both these problems are fixed. "Subpattern name
804 expected" is now given for (?&) (a zero-length name), and this patch also
805 makes it give the same error for \k'' (previously it complained that that
806 was a reference to a non-existent subpattern).
807
808 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
809 this is right because (?- can be followed by option settings as well as by
810 digits. I have, however, made the messages clearer.
811
812 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
813 than the number used in the conditional) now cause a compile-time error.
814 This is actually not compatible with Perl, which accepts such patterns, but
815 treats the conditional as always being FALSE (as PCRE used to), but it
816 seems to me that giving a diagnostic is better.
817
818 8. Change "alphameric" to the more common word "alphanumeric" in comments
819 and messages.
820
821 9. Fix two occurrences of "backslash" in comments that should have been
822 "backspace".
823
824 10. Remove two redundant lines of code that can never be obeyed (their function
825 was moved elsewhere).
826
827 11. The program that makes PCRE's Unicode character property table had a bug
828 which caused it to generate incorrect table entries for sequences of
829 characters that have the same character type, but are in different scripts.
830 It amalgamated them into a single range, with the script of the first of
831 them. In other words, some characters were in the wrong script. There were
832 thirteen such cases, affecting characters in the following ranges:
833
834 U+002b0 - U+002c1
835 U+0060c - U+0060d
836 U+0061e - U+00612
837 U+0064b - U+0065e
838 U+0074d - U+0076d
839 U+01800 - U+01805
840 U+01d00 - U+01d77
841 U+01d9b - U+01dbf
842 U+0200b - U+0200f
843 U+030fc - U+030fe
844 U+03260 - U+0327f
845 U+0fb46 - U+0fbb1
846 U+10450 - U+1049d
847
848 12. The -o option (show only the matching part of a line) for pcregrep was not
849 compatible with GNU grep in that, if there was more than one match in a
850 line, it showed only the first of them. It now behaves in the same way as
851 GNU grep.
852
853 13. If the -o and -v options were combined for pcregrep, it printed a blank
854 line for every non-matching line. GNU grep prints nothing, and pcregrep now
855 does the same. The return code can be used to tell if there were any
856 non-matching lines.
857
858 14. Added --file-offsets and --line-offsets to pcregrep.
859
860 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
861 infinitely looping recursion. The bug was that positive lookaheads were not
862 being skipped when checking for a possible empty match (negative lookaheads
863 and both kinds of lookbehind were skipped).
864
865 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
866 inclusion of <windows.h> to before rather than after the definition of
867 INVALID_FILE_ATTRIBUTES (patch from David Byron).
868
869 17. Specifying a possessive quantifier with a specific limit for a Unicode
870 character property caused pcre_compile() to compile bad code, which led at
871 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
872 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
873 caused the error; without that there was no problem.
874
875 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
876
877 19. Added --enable-pcretest-libreadline.
878
879 20. In pcrecpp.cc, the variable 'count' was incremented twice in
880 RE::GlobalReplace(). As a result, the number of replacements returned was
881 double what it should be. I removed one of the increments, but Craig sent a
882 later patch that removed the other one (the right fix) and added unit tests
883 that check the return values (which was not done before).
884
885 21. Several CMake things:
886
887 (1) Arranged that, when cmake is used on Unix, the libraries end up with
888 the names libpcre and libpcreposix, not just pcre and pcreposix.
889
890 (2) The above change means that pcretest and pcregrep are now correctly
891 linked with the newly-built libraries, not previously installed ones.
892
893 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
894
895 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
896 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
897 UTF-8 newline character). The key issue is that the pattern starts .*;
898 this means that the match must be either at the beginning, or after a
899 newline. The bug was in the code for advancing after a failed match and
900 checking that the new position followed a newline. It was not taking
901 account of UTF-8 characters correctly.
902
903 23. PCRE was behaving differently from Perl in the way it recognized POSIX
904 character classes. PCRE was not treating the sequence [:...:] as a
905 character class unless the ... were all letters. Perl, however, seems to
906 allow any characters between [: and :], though of course it rejects as
907 unknown any "names" that contain non-letters, because all the known class
908 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
909 for example, whereas PCRE did not - it did not recognize a POSIX character
910 class. This seemed a bit dangerous, so the code has been changed to be
911 closer to Perl. The behaviour is not identical to Perl, because PCRE will
912 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
913 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
914 Perl does, and where it didn't before.
915
916 24. Rewrite so as to remove the single use of %n from pcregrep because in some
917 Windows environments %n is disabled by default.
918
919
920 Version 7.4 21-Sep-07
921 ---------------------
922
923 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
924 means that a class such as [\s] counted as "explicit reference to CR or
925 LF". That isn't really right - the whole point of the change was to try to
926 help when there was an actual mention of one of the two characters. So now
927 the change happens only if \r or \n (or a literal CR or LF) character is
928 encountered.
929
930 2. The 32-bit options word was also used for 6 internal flags, but the numbers
931 of both had grown to the point where there were only 3 bits left.
932 Fortunately, there was spare space in the data structure, and so I have
933 moved the internal flags into a new 16-bit field to free up more option
934 bits.
935
936 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
937 but did not set the internal JCHANGED flag - either of these is enough to
938 control the way the "get" function works - but the PCRE_INFO_JCHANGED
939 facility is supposed to tell if (?J) was ever used, so now (?J) at the
940 start sets both bits.
941
942 4. Added options (at build time, compile time, exec time) to change \R from
943 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
944
945 5. doc/pcresyntax.html was missing from the distribution.
946
947 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
948 compatibility, even though it is no longer used.
949
950 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
951 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
952 windows.h file is present (where different names are used). [This was
953 reversed later after testing - see 16 below.]
954
955 8. Changed all #include <config.h> to #include "config.h". There were also
956 some further <pcre.h> cases that I changed to "pcre.h".
957
958 9. When pcregrep was used with the --colour option, it missed the line ending
959 sequence off the lines that it output.
960
961 10. It was pointed out to me that arrays of string pointers cause lots of
962 relocations when a shared library is dynamically loaded. A technique of
963 using a single long string with a table of offsets can drastically reduce
964 these. I have refactored PCRE in four places to do this. The result is
965 dramatic:
966
967 Originally: 290
968 After changing UCP table: 187
969 After changing error message table: 43
970 After changing table of "verbs" 36
971 After changing table of Posix names 22
972
973 Thanks to the folks working on Gregex for glib for this insight.
974
975 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
976 unicode-properties was also set.
977
978 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
979
980 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
981 checked only for CRLF.
982
983 14. Added casts to pcretest.c to avoid compiler warnings.
984
985 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
986
987 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
988 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
989 entirely. This removes changes made in 7 above.
990
991 17. The CMake files have been updated, and there is now more information about
992 building with CMake in the NON-UNIX-USE document.
993
994
995 Version 7.3 28-Aug-07
996 ---------------------
997
998 1. In the rejigging of the build system that eventually resulted in 7.1, the
999 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1000 brackets there is not right, since it causes compilers to look for an
1001 installed pcre.h, not the version that is in the source that is being
1002 compiled (which of course may be different). I have changed it back to:
1003
1004 #include "pcre.h"
1005
1006 I have a vague recollection that the change was concerned with compiling in
1007 different directories, but in the new build system, that is taken care of
1008 by the VPATH setting the Makefile.
1009
1010 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1011 when the subject happened to end in the byte 0x85 (e.g. if the last
1012 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1013 characters but of course it shouldn't be taken as a newline when it is part
1014 of another character. The bug was that, for an unlimited repeat of . in
1015 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1016 characters when looking for a newline.
1017
1018 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1019
1020 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1021 in debug output.
1022
1023 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1024 long printing in the pcrecpp unittest when running under MinGW.
1025
1026 6. ESC_K was left out of the EBCDIC table.
1027
1028 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1029 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1030 limit also applies to "virtual nesting" when a pattern is recursive, and in
1031 this case 1000 isn't so big. I have been able to remove this limit at the
1032 expense of backing off one optimization in certain circumstances. Normally,
1033 when pcre_exec() would call its internal match() function recursively and
1034 immediately return the result unconditionally, it uses a "tail recursion"
1035 feature to save stack. However, when a subpattern that can match an empty
1036 string has an unlimited repetition quantifier, it no longer makes this
1037 optimization. That gives it a stack frame in which to save the data for
1038 checking that an empty string has been matched. Previously this was taken
1039 from the 1000-entry workspace that had been reserved. So now there is no
1040 explicit limit, but more stack is used.
1041
1042 8. Applied Daniel's patches to solve problems with the import/export magic
1043 syntax that is required for Windows, and which was going wrong for the
1044 pcreposix and pcrecpp parts of the library. These were overlooked when this
1045 problem was solved for the main library.
1046
1047 9. There were some crude static tests to avoid integer overflow when computing
1048 the size of patterns that contain repeated groups with explicit upper
1049 limits. As the maximum quantifier is 65535, the maximum group length was
1050 set at 30,000 so that the product of these two numbers did not overflow a
1051 32-bit integer. However, it turns out that people want to use groups that
1052 are longer than 30,000 bytes (though not repeat them that many times).
1053 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1054 made it possible to implement the integer overflow checks in a much more
1055 dynamic way, which I have now done. The artificial limitation on group
1056 length has been removed - we now have only the limit on the total length of
1057 the compiled pattern, which depends on the LINK_SIZE setting.
1058
1059 10. Fixed a bug in the documentation for get/copy named substring when
1060 duplicate names are permitted. If none of the named substrings are set, the
1061 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1062 empty string.
1063
1064 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1065 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1066 because the ] is interpreted as the first data character and the
1067 terminating ] is not found. PCRE has been made compatible with Perl in this
1068 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1069 cause memory overwriting.
1070
1071 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1072 string has been matched (to stop an infinite loop). It was not recognizing
1073 a conditional subpattern that could match an empty string if that
1074 subpattern was within another subpattern. For example, it looped when
1075 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1076 condition was not nested. This bug has been fixed.
1077
1078 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1079 past the start of the subject in the presence of bytes with the top bit
1080 set, for example "\x8aBCD".
1081
1082 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1083 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1084
1085 14. Optimized (?!) to (*FAIL).
1086
1087 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1088 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1089 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1090 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1091 does: it's just the validity check that is more restrictive.
1092
1093 16. Inserted checks for integer overflows during escape sequence (backslash)
1094 processing, and also fixed erroneous offset values for syntax errors during
1095 backslash processing.
1096
1097 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1098 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1099
1100 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1101 caused an overrun.
1102
1103 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1104 something other than just ASCII characters) inside a group that had an
1105 unlimited repeat caused a loop at compile time (while checking to see
1106 whether the group could match an empty string).
1107
1108 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1109 [\P{Any}] did so. (Error in the code for printing property names.)
1110
1111 21. An orphan \E inside a character class could cause a crash.
1112
1113 22. A repeated capturing bracket such as (A)? could cause a wild memory
1114 reference during compilation.
1115
1116 23. There are several functions in pcre_compile() that scan along a compiled
1117 expression for various reasons (e.g. to see if it's fixed length for look
1118 behind). There were bugs in these functions when a repeated \p or \P was
1119 present in the pattern. These operators have additional parameters compared
1120 with \d, etc, and these were not being taken into account when moving along
1121 the compiled data. Specifically:
1122
1123 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1124 length.
1125
1126 (b) An item such as \pL+ within a repeated group could cause crashes or
1127 loops.
1128
1129 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1130 "reference to non-existent subpattern" error.
1131
1132 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1133
1134 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1135 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1136
1137 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1138
1139 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1140 character were causing crashes (broken optimization).
1141
1142 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1143 \p or \P) caused a compile-time loop.
1144
1145 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1146 break. For example, the unstudied pattern [\r\n]A does not match the string
1147 "\r\nA" because change 7.0/46 below moves the current point on by two
1148 characters after failing to match at the start. However, the pattern \nA
1149 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1150 the same is true. There doesn't seem any very clean way out of this, but
1151 what I have chosen to do makes the common cases work: PCRE now takes note
1152 of whether there can be an explicit match for \r or \n anywhere in the
1153 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1154 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1155 pattern has explicit CR or LF references.
1156
1157 29. Added (*CR) etc for changing newline setting at start of pattern.
1158
1159
1160 Version 7.2 19-Jun-07
1161 ---------------------
1162
1163 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1164 which is apparently normally available under Windows.
1165
1166 2. Re-jig the pcregrep tests with different newline settings in an attempt
1167 to make them independent of the local environment's newline setting.
1168
1169 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1170
1171 4. Some of the "internals" tests were previously cut out when the link size
1172 was not 2, because the output contained actual offsets. The recent new
1173 "Z" feature of pcretest means that these can be cut out, making the tests
1174 usable with all link sizes.
1175
1176 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1177 stack recursion. This gives a massive performance boost under BSD, but just
1178 a small improvement under Linux. However, it saves one field in the frame
1179 in all cases.
1180
1181 6. Added more features from the forthcoming Perl 5.10:
1182
1183 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1184 recursion call. It refers to the nth most recently opened parentheses.
1185
1186 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1187 to be opened parentheses.
1188
1189 (c) Conditions that refer to capturing parentheses can be specified
1190 relatively, for example, (?(-2)... or (?(+3)...
1191
1192 (d) \K resets the start of the current match so that everything before
1193 is not part of it.
1194
1195 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1196
1197 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1198 reference syntax.
1199
1200 (g) (?| introduces a group in which the numbering of parentheses in each
1201 alternative starts with the same number.
1202
1203 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1204
1205 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1206 PCRE_INFO_JCHANGED.
1207
1208 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1209 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1210 for detecting groups that can match an empty string.
1211
1212 9. A pattern with a very large number of alternatives (more than several
1213 hundred) was running out of internal workspace during the pre-compile
1214 phase, where pcre_compile() figures out how much memory will be needed. A
1215 bit of new cunning has reduced the workspace needed for groups with
1216 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1217 workspace instead of running out of the 4096 that are available.
1218
1219 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1220
1221 11. Applied patch from Google to remove an optimization that didn't quite work.
1222 The report of the bug said:
1223
1224 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1225 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1226 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1227
1228 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1229 it matched the wrong number of bytes.
1230
1231
1232 Version 7.1 24-Apr-07
1233 ---------------------
1234
1235 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1236 that is more "standard", making use of automake and other Autotools. There
1237 is some re-arrangement of the files and adjustment of comments consequent
1238 on this.
1239
1240 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1241 for recursive directory scanning broke on some systems because the files
1242 are not scanned in any specific order and on different systems the order
1243 was different. A call to "sort" has been inserted into RunGrepTest for the
1244 approprate test as a short-term fix. In the longer term there may be an
1245 alternative.
1246
1247 3. I had an email from Eric Raymond about problems translating some of PCRE's
1248 man pages to HTML (despite the fact that I distribute HTML pages, some
1249 people do their own conversions for various reasons). The problems
1250 concerned the use of low-level troff macros .br and .in. I have therefore
1251 removed all such uses from the man pages (some were redundant, some could
1252 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1253 HTML has been updated to handle .nf/.fi and to complain if it encounters
1254 .br or .in.
1255
1256 4. Updated comments in configure.ac that get placed in config.h.in and also
1257 arranged for config.h to be included in the distribution, with the name
1258 config.h.generic, for the benefit of those who have to compile without
1259 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1260
1261 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1262 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1263 makevp.bat for the current PCRE, using the additional files
1264 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1265
1266 6. A Windows user reported a minor discrepancy with test 2, which turned out
1267 to be caused by a trailing space on an input line that had got lost in his
1268 copy. The trailing space was an accident, so I've just removed it.
1269
1270 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1271 that is needed.
1272
1273 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1274 as "const" (a) because they are and (b) because it helps the PHP
1275 maintainers who have recently made a script to detect big data structures
1276 in the php code that should be moved to the .rodata section. I remembered
1277 to update Builducptable as well, so it won't revert if ucptable.h is ever
1278 re-created.
1279
1280 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1281 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1282 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1283 support is not required. This saves 1.5-2K of code, which is important in
1284 some applications.
1285
1286 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1287 so as not to refer to the tables, even though these functions will never be
1288 called when UTF-8 support is disabled. Otherwise there are problems with a
1289 shared library.
1290
1291 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1292
1293 (a) It was defining its arguments as char * instead of void *.
1294
1295 (b) It was assuming that all moves were upwards in memory; this was true
1296 a long time ago when I wrote it, but is no longer the case.
1297
1298 The emulated memove() is provided for those environments that have neither
1299 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1300 is clearly not the case, as these two bugs were recently reported.
1301
1302 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1303 and Detrail to create the HTML documentation, the .txt form of the man
1304 pages, and it removes trailing spaces from listed files. It also creates
1305 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1306 case, it wraps all the #defines with #ifndefs. This script should be run
1307 before "make dist".
1308
1309 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1310 with Unicode property support.
1311
1312 (a) For a maximizing quantifier, if the two different cases of the
1313 character were of different lengths in their UTF-8 codings (there are
1314 some cases like this - I found 11), and the matching function had to
1315 back up over a mixture of the two cases, it incorrectly assumed they
1316 were both the same length.
1317
1318 (b) When PCRE was configured to use the heap rather than the stack for
1319 recursion during matching, it was not correctly preserving the data for
1320 the other case of a UTF-8 character when checking ahead for a match
1321 while processing a minimizing repeat. If the check also involved
1322 matching a wide character, but failed, corruption could cause an
1323 erroneous result when trying to check for a repeat of the original
1324 character.
1325
1326 13. Some tidying changes to the testing mechanism:
1327
1328 (a) The RunTest script now detects the internal link size and whether there
1329 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1330 values substituted by "configure". (The RunGrepTest script already did
1331 this for UTF-8.) The configure.ac script no longer substitutes the
1332 relevant variables.
1333
1334 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1335 with length and offset values. This means that the output is different
1336 for different internal link sizes. Test 2 is skipped for link sizes
1337 other than 2 because of this, bypassing the problem. Unfortunately,
1338 there was also a test in test 3 (the locale tests) that used /B and
1339 failed for link sizes other than 2. Rather than cut the whole test out,
1340 I have added a new /Z option to pcretest that replaces the length and
1341 offset values with spaces. This is now used to make test 3 independent
1342 of link size. (Test 2 will be tidied up later.)
1343
1344 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1345 segmentation fault instead of returning the appropriate error message.
1346
1347 15. In multiline mode when the newline sequence was set to "any", the pattern
1348 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1349 This doesn't seem right; it now treats the CRLF combination as the line
1350 ending, and so does not match in that case. It's only a pattern such as ^$
1351 that would hit this one: something like ^ABC$ would have failed after \r
1352 and then tried again after \r\n.
1353
1354 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1355 in an attempt to make files that differ only in their line terminators
1356 compare equal. This works on Linux.
1357
1358 17. Under certain error circumstances pcregrep might try to free random memory
1359 as it exited. This is now fixed, thanks to valgrind.
1360
1361 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1362 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1363 was because its rules for how to advance for /g after matching an empty
1364 string at the end of a line did not allow for this case. They now check for
1365 it specially.
1366
1367 20. pcretest is supposed to handle patterns and data of any length, by
1368 extending its buffers when necessary. It was getting this wrong when the
1369 buffer for a data line had to be extended.
1370
1371 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1372 CRLF as a newline sequence.
1373
1374 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1375 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1376 I have nevertheless tidied it up.
1377
1378 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1379
1380 24. Added a man page for pcre-config.
1381
1382
1383 Version 7.0 19-Dec-06
1384 ---------------------
1385
1386 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1387 moving to gcc 4.1.1.
1388
1389 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1390 sys/time.h, which is documented as needed for this function. It doesn't
1391 seem to matter on Linux, but it showed up on some releases of OS X.
1392
1393 3. It seems that there are systems where bytes whose values are greater than
1394 127 match isprint() in the "C" locale. The "C" locale should be the
1395 default when a C program starts up. In most systems, only ASCII printing
1396 characters match isprint(). This difference caused the output from pcretest
1397 to vary, making some of the tests fail. I have changed pcretest so that:
1398
1399 (a) When it is outputting text in the compiled version of a pattern, bytes
1400 other than 32-126 are always shown as hex escapes.
1401
1402 (b) When it is outputting text that is a matched part of a subject string,
1403 it does the same, unless a different locale has been set for the match
1404 (using the /L modifier). In this case, it uses isprint() to decide.
1405
1406 4. Fixed a major bug that caused incorrect computation of the amount of memory
1407 required for a compiled pattern when options that changed within the
1408 pattern affected the logic of the preliminary scan that determines the
1409 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1410 that the computed length was too small. The symptoms of this bug were
1411 either the PCRE error "internal error: code overflow" from pcre_compile(),
1412 or a glibc crash with a message such as "pcretest: free(): invalid next
1413 size (fast)". Examples of patterns that provoked this bug (shown in
1414 pcretest format) are:
1415
1416 /(?-x: )/x
1417 /(?x)(?-x: \s*#\s*)/
1418 /((?i)[\x{c0}])/8
1419 /(?i:[\x{c0}])/8
1420
1421 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1422 is now done differently.
1423
1424 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1425 wrapper classes; (b) implement a new function in the C++ scanner that is
1426 more efficient than the old way of doing things because it avoids levels of
1427 recursion in the regex matching; (c) add a paragraph to the documentation
1428 for the FullMatch() function.
1429
1430 6. The escape sequence \n was being treated as whatever was defined as
1431 "newline". Not only was this contrary to the documentation, which states
1432 that \n is character 10 (hex 0A), but it also went horribly wrong when
1433 "newline" was defined as CRLF. This has been fixed.
1434
1435 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1436 was being set to -1 for the "end of line" case (supposedly a value that no
1437 character can have). Though this value is never used (the check for end of
1438 line is "zero bytes in current character"), it caused compiler complaints.
1439 I've changed it to 0xffffffff.
1440
1441 8. In pcre_version.c, the version string was being built by a sequence of
1442 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1443 string (as it is for production releases) called a macro with an empty
1444 argument. The C standard says the result of this is undefined. The gcc
1445 compiler treats it as an empty string (which was what was wanted) but it is
1446 reported that Visual C gives an error. The source has been hacked around to
1447 avoid this problem.
1448
1449 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1450 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1451 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1452 of them did).
1453
1454 10. Originally, pcretest opened its input and output without "b"; then I was
1455 told that "b" was needed in some environments, so it was added for release
1456 5.0 to both the input and output. (It makes no difference on Unix-like
1457 systems.) Later I was told that it is wrong for the input on Windows. I've
1458 now abstracted the modes into two macros, to make it easier to fiddle with
1459 them, and removed "b" from the input mode under Windows.
1460
1461 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1462
1463 12. Added -help and --help to pcretest as an official way of being reminded
1464 of the options.
1465
1466 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1467 and pcrecpp.cc because they annoy compilers at high warning levels.
1468
1469 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1470
1471 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1472 scripts are not C programs :-) and which was not noticed because it works
1473 on Linux.
1474
1475 16. pcretest is supposed to handle any length of pattern and data line (as one
1476 line or as a continued sequence of lines) by extending its input buffer if
1477 necessary. This feature was broken for very long pattern lines, leading to
1478 a string of junk being passed to pcre_compile() if the pattern was longer
1479 than about 50K.
1480
1481 17. I have done a major re-factoring of the way pcre_compile() computes the
1482 amount of memory needed for a compiled pattern. Previously, there was code
1483 that made a preliminary scan of the pattern in order to do this. That was
1484 OK when PCRE was new, but as the facilities have expanded, it has become
1485 harder and harder to keep it in step with the real compile phase, and there
1486 have been a number of bugs (see for example, 4 above). I have now found a
1487 cunning way of running the real compile function in a "fake" mode that
1488 enables it to compute how much memory it would need, while actually only
1489 ever using a few hundred bytes of working memory and without too many
1490 tests of the mode. This should make future maintenance and development
1491 easier. A side effect of this work is that the limit of 200 on the nesting
1492 depth of parentheses has been removed (though this was never a serious
1493 limitation, I suspect). However, there is a downside: pcre_compile() now
1494 runs more slowly than before (30% or more, depending on the pattern). I
1495 hope this isn't a big issue. There is no effect on runtime performance.
1496
1497 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1498 newline (only possible for the last line of a file) and it was a
1499 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1500
1501 19. Added additional timing features to pcretest. (1) The -tm option now times
1502 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1503 separate command line item, by a number that specifies the number of
1504 repeats to use when timing. The default is 50000; this gives better
1505 precision, but takes uncomfortably long for very large patterns.
1506
1507 20. Extended pcre_study() to be more clever in cases where a branch of a
1508 subpattern has no definite first character. For example, (a*|b*)[cd] would
1509 previously give no result from pcre_study(). Now it recognizes that the
1510 first character must be a, b, c, or d.
1511
1512 21. There was an incorrect error "recursive call could loop indefinitely" if
1513 a subpattern (or the entire pattern) that was being tested for matching an
1514 empty string contained only one non-empty item after a nested subpattern.
1515 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1516 incorrectly, because the \d was being skipped in the check.
1517
1518 22. The pcretest program now has a new pattern option /B and a command line
1519 option -b, which is equivalent to adding /B to every pattern. This causes
1520 it to show the compiled bytecode, without the additional information that
1521 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1522 is the same as /B/I).
1523
1524 23. A new optimization is now able automatically to treat some sequences such
1525 as a*b as a*+b. More specifically, if something simple (such as a character
1526 or a simple class like \d) has an unlimited quantifier, and is followed by
1527 something that cannot possibly match the quantified thing, the quantifier
1528 is automatically "possessified".
1529
1530 24. A recursive reference to a subpattern whose number was greater than 39
1531 went wrong under certain circumstances in UTF-8 mode. This bug could also
1532 have affected the operation of pcre_study().
1533
1534 25. Realized that a little bit of performance could be had by replacing
1535 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1536
1537 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1538
1539 27. Possessive quantifiers such as a++ were previously implemented by turning
1540 them into atomic groups such as ($>a+). Now they have their own opcodes,
1541 which improves performance. This includes the automatically created ones
1542 from 23 above.
1543
1544 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1545 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1546 the first matched character to be a colon. This applied both to named and
1547 numbered groups.
1548
1549 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1550
1551 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1552 building PCRE on Windows easier, so I have included it in the distribution.
1553
1554 31. There is now a check in pcretest against a ridiculously large number being
1555 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1556 loop, the loop is abandoned.
1557
1558 32. Forward references to subpatterns in conditions such as (?(2)...) where
1559 subpattern 2 is defined later cause pcre_compile() to search forwards in
1560 the pattern for the relevant set of parentheses. This search went wrong
1561 when there were unescaped parentheses in a character class, parentheses
1562 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1563
1564 33. "Subroutine" calls and backreferences were previously restricted to
1565 referencing subpatterns earlier in the regex. This restriction has now
1566 been removed.
1567
1568 34. Added a number of extra features that are going to be in Perl 5.10. On the
1569 whole, these are just syntactic alternatives for features that PCRE had
1570 previously implemented using the Python syntax or my own invention. The
1571 other formats are all retained for compatibility.
1572
1573 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1574 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1575 also .NET compatible.
1576
1577 (b) A recursion or subroutine call to a named group can now be defined as
1578 (?&name) as well as (?P>name).
1579
1580 (c) A backreference to a named group can now be defined as \k<name> or
1581 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1582 5.10, are also .NET compatible.
1583
1584 (d) A conditional reference to a named group can now use the syntax
1585 (?(<name>) or (?('name') as well as (?(name).
1586
1587 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1588 groups (named and numbered) that are never evaluated inline, but can be
1589 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1590 is always false. There may be only one alternative in such a group.
1591
1592 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1593 as the simple (?(R). The condition is true only if the most recent
1594 recursion is that of the given number or name. It does not search out
1595 through the entire recursion stack.
1596
1597 (g) The escape \gN or \g{N} has been added, where N is a positive or
1598 negative number, specifying an absolute or relative reference.
1599
1600 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1601 some "unreachable code" warnings.
1602
1603 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1604 things, this adds five new scripts.
1605
1606 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1607 There were also incompatibilities regarding the handling of \Q..\E inside
1608 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1609 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1610
1611 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1612 matches an empty string, and forcibly breaks the loop. There were bugs in
1613 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1614 against aaaa the result was just "a" rather than "aaaa", for example. Two
1615 separate and independent bugs (that affected different cases) have been
1616 fixed.
1617
1618 39. Refactored the code to abolish the use of different opcodes for small
1619 capturing bracket numbers. This is a tidy that I avoided doing when I
1620 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1621 The new approach is not only tidier, it makes it possible to reduce the
1622 memory needed to fix the previous bug (38).
1623
1624 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1625 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1626 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1627 mode.
1628
1629 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1630 report.
1631
1632 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1633 copy construction and assignment in the C++ wrapper.
1634
1635 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1636 couple of bugs that could have given wrong results in the "--newline=crlf"
1637 case.
1638
1639 44. Added a number of casts and did some reorganization of signed/unsigned int
1640 variables following suggestions from Dair Grant. Also renamed the variable
1641 "this" as "item" because it is a C++ keyword.
1642
1643 45. Arranged for dftables to add
1644
1645 #include "pcre_internal.h"
1646
1647 to pcre_chartables.c because without it, gcc 4.x may remove the array
1648 definition from the final binary if PCRE is built into a static library and
1649 dead code stripping is activated.
1650
1651 46. For an unanchored pattern, if a match attempt fails at the start of a
1652 newline sequence, and the newline setting is CRLF or ANY, and the next two
1653 characters are CRLF, advance by two characters instead of one.
1654
1655
1656 Version 6.7 04-Jul-06
1657 ---------------------
1658
1659 1. In order to handle tests when input lines are enormously long, pcretest has
1660 been re-factored so that it automatically extends its buffers when
1661 necessary. The code is crude, but this _is_ just a test program. The
1662 default size has been increased from 32K to 50K.
1663
1664 2. The code in pcre_study() was using the value of the re argument before
1665 testing it for NULL. (Of course, in any sensible call of the function, it
1666 won't be NULL.)
1667
1668 3. The memmove() emulation function in pcre_internal.h, which is used on
1669 systems that lack both memmove() and bcopy() - that is, hardly ever -
1670 was missing a "static" storage class specifier.
1671
1672 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1673 containing an extended class (one that cannot be represented by a bitmap
1674 because it contains high-valued characters or Unicode property items, e.g.
1675 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1676 pattern, but PCRE should not loop if you do not (it no longer does).
1677 [Detail: two cases were found: (a) a repeated subpattern containing an
1678 extended class; (b) a recursive reference to a subpattern that followed a
1679 previous extended class. It wasn't skipping over the extended class
1680 correctly when UTF-8 mode was not set.]
1681
1682 5. A negated single-character class was not being recognized as fixed-length
1683 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1684 compile error "lookbehind assertion is not fixed length".
1685
1686 6. The RunPerlTest auxiliary script was showing an unexpected difference
1687 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1688 write a Perl script that can interpret lines of an input file either as
1689 byte characters or as UTF-8, which is what "perltest" was being required to
1690 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1691 can't do is switch easily at run time between having the "use utf8;" pragma
1692 or not. In the end, I fudged it by using the RunPerlTest script to insert
1693 "use utf8;" explicitly for the UTF-8 tests.
1694
1695 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1696 the end of the subject string, contrary to the documentation and to what
1697 Perl does. This was true of both matching functions. Now it matches only at
1698 the start of the subject and immediately after *internal* newlines.
1699
1700 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1701 a pointer to an int instead of a pointer to an unsigned long int. This
1702 caused problems on 64-bit systems.
1703
1704 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1705 instance of the 'standard' template library not being so standard".
1706
1707 10. There was no check on the number of named subpatterns nor the maximum
1708 length of a subpattern name. The product of these values is used to compute
1709 the size of the memory block for a compiled pattern. By supplying a very
1710 long subpattern name and a large number of named subpatterns, the size
1711 computation could be caused to overflow. This is now prevented by limiting
1712 the length of names to 32 characters, and the number of named subpatterns
1713 to 10,000.
1714
1715 11. Subpatterns that are repeated with specific counts have to be replicated in
1716 the compiled pattern. The size of memory for this was computed from the
1717 length of the subpattern and the repeat count. The latter is limited to
1718 65535, but there was no limit on the former, meaning that integer overflow
1719 could in principle occur. The compiled length of a repeated subpattern is
1720 now limited to 30,000 bytes in order to prevent this.
1721
1722 12. Added the optional facility to have named substrings with the same name.
1723
1724 13. Added the ability to use a named substring as a condition, using the
1725 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1726 are numbers (not recommended). Forward references are permitted.
1727
1728 14. Added forward references in named backreferences (if you see what I mean).
1729
1730 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1731 pattern could run off the end of the subject. For example, the pattern
1732 "(?s)(.{1,5})"8 did this with the subject "ab".
1733
1734 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1735 PCRE_CASELESS was set when matching characters that were quantified with ?
1736 or *.
1737
1738 17. A character class other than a single negated character that had a minimum
1739 but no maximum quantifier - for example [ab]{6,} - was not handled
1740 correctly by pce_dfa_exec(). It would match only one character.
1741
1742 18. A valid (though odd) pattern that looked like a POSIX character
1743 class but used an invalid character after [ (for example [[,abc,]]) caused
1744 pcre_compile() to give the error "Failed: internal error: code overflow" or
1745 in some cases to crash with a glibc free() error. This could even happen if
1746 the pattern terminated after [[ but there just happened to be a sequence of
1747 letters, a binary zero, and a closing ] in the memory that followed.
1748
1749 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1750 over the years. Originally (before any Unicode support), just the bottom 8
1751 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1752 output from "man perlunicode" includes this:
1753
1754 The regular expression compiler produces polymorphic opcodes. That
1755 is, the pattern adapts to the data and automatically switches to
1756 the Unicode character scheme when presented with Unicode data--or
1757 instead uses a traditional byte scheme when presented with byte
1758 data.
1759
1760 Sadly, a wide octal escape does not cause a switch, and in a string with
1761 no other multibyte characters, these octal escapes are treated as before.
1762 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1763 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1764 Unicode string.
1765
1766 I have not perpetrated such confusion in PCRE. Up till now, it took just
1767 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1768 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1769 translate to the appropriate multibyte character.
1770
1771 29. Applied some refactoring to reduce the number of warnings from Microsoft
1772 and Borland compilers. This has included removing the fudge introduced
1773 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1774 a warning about an unused variable.
1775
1776 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1777 characters since release 4.0, because Perl (from release 5.004) does not.
1778 [Or at least, is documented not to: some releases seem to be in conflict
1779 with the documentation.] However, when a pattern was studied with
1780 pcre_study() and all its branches started with \s, PCRE still included VT
1781 as a possible starting character. Of course, this did no harm; it just
1782 caused an unnecessary match attempt.
1783
1784 22. Removed a now-redundant internal flag bit that recorded the fact that case
1785 dependency changed within the pattern. This was once needed for "required
1786 byte" processing, but is no longer used. This recovers a now-scarce options
1787 bit. Also moved the least significant internal flag bit to the most-
1788 significant bit of the word, which was not previously used (hangover from
1789 the days when it was an int rather than a uint) to free up another bit for
1790 the future.
1791
1792 23. Added support for CRLF line endings as well as CR and LF. As well as the
1793 default being selectable at build time, it can now be changed at runtime
1794 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1795 specify that it is scanning data with non-default line endings.
1796
1797 24. Changed the definition of CXXLINK to make it agree with the definition of
1798 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1799
1800 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1801 recursions. This makes a big different to stack usage for some patterns.
1802
1803 26. If a subpattern containing a named recursion or subroutine reference such
1804 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1805 the space required for the compiled pattern went wrong and gave too small a
1806 value. Depending on the environment, this could lead to "Failed: internal
1807 error: code overflow at offset 49" or "glibc detected double free or
1808 corruption" errors.
1809
1810 27. Applied patches from Google (a) to support the new newline modes and (b) to
1811 advance over multibyte UTF-8 characters in GlobalReplace.
1812
1813 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1814 difference for some implementation of PCRE in some Windows version.
1815
1816 29. Added some extra testing facilities to pcretest:
1817
1818 \q<number> in a data line sets the "match limit" value
1819 \Q<number> in a data line sets the "match recursion limt" value
1820 -S <number> sets the stack size, where <number> is in megabytes
1821
1822 The -S option isn't available for Windows.
1823
1824
1825 Version 6.6 06-Feb-06
1826 ---------------------
1827
1828 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1829 in pcreposix.h. I have copied the definition from pcre.h.
1830
1831 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1832 because pcre.h is no longer a built file.
1833
1834 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1835 not normally included in the compiled code.
1836
1837
1838 Version 6.5 01-Feb-06
1839 ---------------------
1840
1841 1. When using the partial match feature with pcre_dfa_exec(), it was not
1842 anchoring the second and subsequent partial matches at the new starting
1843 point. This could lead to incorrect results. For example, with the pattern
1844 /1234/, partially matching against "123" and then "a4" gave a match.
1845
1846 2. Changes to pcregrep:
1847
1848 (a) All non-match returns from pcre_exec() were being treated as failures
1849 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1850 error message is output. Some extra information is given for the
1851 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1852 probably the only errors that are likely to be caused by users (by
1853 specifying a regex that has nested indefinite repeats, for instance).
1854 If there are more than 20 of these errors, pcregrep is abandoned.
1855
1856 (b) A binary zero was treated as data while matching, but terminated the
1857 output line if it was written out. This has been fixed: binary zeroes
1858 are now no different to any other data bytes.
1859
1860 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1861 used to set a locale for matching. The --locale=xxxx long option has
1862 been added (no short equivalent) to specify a locale explicitly on the
1863 pcregrep command, overriding the environment variables.
1864
1865 (d) When -B was used with -n, some line numbers in the output were one less
1866 than they should have been.
1867
1868 (e) Added the -o (--only-matching) option.
1869
1870 (f) If -A or -C was used with -c (count only), some lines of context were
1871 accidentally printed for the final match.
1872
1873 (g) Added the -H (--with-filename) option.
1874
1875 (h) The combination of options -rh failed to suppress file names for files
1876 that were found from directory arguments.
1877
1878 (i) Added the -D (--devices) and -d (--directories) options.
1879
1880 (j) Added the -F (--fixed-strings) option.
1881
1882 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1883
1884 (l) Added the --colo(u)r option.
1885
1886 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1887 is not present by default.
1888
1889 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1890 items such as (?R) or (?1), when the recursion could match a number of
1891 alternatives. If it matched one of the alternatives, but subsequently,
1892 outside the recursion, there was a failure, the code tried to back up into
1893 the recursion. However, because of the way PCRE is implemented, this is not
1894 possible, and the result was an incorrect result from the match.
1895
1896 In order to prevent this happening, the specification of recursion has
1897 been changed so that all such subpatterns are automatically treated as
1898 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1899
1900 4. I had overlooked the fact that, in some locales, there are characters for
1901 which isalpha() is true but neither isupper() nor islower() are true. In
1902 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1903 and ordfeminine) are like this. This affected the treatment of \w and \W
1904 when they appeared in character classes, but not when they appeared outside
1905 a character class. The bit map for "word" characters is now created
1906 separately from the results of isalnum() instead of just taking it from the
1907 upper, lower, and digit maps. (Plus the underscore character, of course.)
1908
1909 5. The above bug also affected the handling of POSIX character classes such as
1910 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1911 permanent tables. Instead, the bit maps for such a class were previously
1912 created as the appropriate unions of the upper, lower, and digit bitmaps.
1913 Now they are created by subtraction from the [[:word:]] class, which has
1914 its own bitmap.
1915
1916 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1917 It is created by subtracting the vertical space characters (\x09, \x0a,
1918 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1919 subtraction was done in the overall bitmap for a character class, meaning
1920 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1921 be recognized. This bug has been fixed.
1922
1923 7. Patches from the folks at Google:
1924
1925 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1926 real life, but is still worth protecting against".
1927
1928 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1929 regular expressions".
1930
1931 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1932 have it.
1933
1934 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1935 "configure" and the latter not, in order to fix a problem somebody had
1936 with compiling the Arg class on HP-UX.
1937
1938 (e) Improve the error-handling of the C++ wrapper a little bit.
1939
1940 (f) New tests for checking recursion limiting.
1941
1942 8. The pcre_memmove() function, which is used only if the environment does not
1943 have a standard memmove() function (and is therefore rarely compiled),
1944 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1945 returning a result (though PCRE never actually uses the result).
1946
1947 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1948 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1949 returned instead of calling malloc() with an overflowing number that would
1950 most likely cause subsequent chaos.
1951
1952 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1953
1954 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1955 with this option is matched, the nmatch and pmatch options of regexec() are
1956 ignored.
1957
1958 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1959 provided in case anyone wants to the the POSIX interface with UTF-8
1960 strings.
1961
1962 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1963 C++ linking (needed for some HP-UX environments).
1964
1965 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1966 (unused parameter) and in the pcre_printint() function (omitted "default"
1967 switch label when the default is to do nothing).
1968
1969 15. Added some code to make it possible, when PCRE is compiled as a C++
1970 library, to replace subject pointers for pcre_exec() with a smart pointer
1971 class, thus making it possible to process discontinuous strings.
1972
1973 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1974 much the same function. They were added by different people who were trying
1975 to make PCRE easy to compile on non-Unix systems. It has been suggested
1976 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1977 for compiling on Windows systems. I have therefore replaced it with
1978 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1979 defaults to "extern" for C or "extern C" for C++, which works fine on
1980 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1981 SCOPE with something explicit in config.h. In addition:
1982
1983 (a) pcreposix.h still had just "extern" instead of either of these macros;
1984 I have replaced it with PCRE_DATA_SCOPE.
1985
1986 (b) Functions such as _pcre_xclass(), which are internal to the library,
1987 but external in the C sense, all had PCRE_EXPORT in their definitions.
1988 This is apparently wrong for the Windows case, so I have removed it.
1989 (It makes no difference on Unix-like systems.)
1990
1991 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1992 of recursive calls to match(). This is different to MATCH_LIMIT because
1993 that limits the total number of calls to match(), not all of which increase
1994 the depth of recursion. Limiting the recursion depth limits the amount of
1995 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1996 when PCRE is compiled, and changed at run time. A patch from Google adds
1997 this functionality to the C++ interface.
1998
1999 18. Changes to the handling of Unicode character properties:
2000
2001 (a) Updated the table to Unicode 4.1.0.
2002
2003 (b) Recognize characters that are not in the table as "Cn" (undefined).
2004
2005 (c) I revised the way the table is implemented to a much improved format
2006 which includes recognition of ranges. It now supports the ranges that
2007 are defined in UnicodeData.txt, and it also amalgamates other
2008 characters into ranges. This has reduced the number of entries in the
2009 table from around 16,000 to around 3,000, thus reducing its size
2010 considerably. I realized I did not need to use a tree structure after
2011 all - a binary chop search is just as efficient. Having reduced the
2012 number of entries, I extended their size from 6 bytes to 8 bytes to
2013 allow for more data.
2014
2015 (d) Added support for Unicode script names via properties such as \p{Han}.
2016
2017 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2018 matching that character.
2019
2020 20. When matching a repeated Unicode property with a minimum greater than zero,
2021 (for example \pL{2,}), PCRE could look past the end of the subject if it
2022 reached it while seeking the minimum number of characters. This could
2023 happen only if some of the characters were more than one byte long, because
2024 there is a check for at least the minimum number of bytes.
2025
2026 21. Refactored the implementation of \p and \P so as to be more general, to
2027 allow for more different types of property in future. This has changed the
2028 compiled form incompatibly. Anybody with saved compiled patterns that use
2029 \p or \P will have to recompile them.
2030
2031 22. Added "Any" and "L&" to the supported property types.
2032
2033 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2034 but give a compile time error if the value is greater than 0xff.
2035
2036 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2037 accidentally not being installed or uninstalled.
2038
2039 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2040 made were to insert the current release number. This seemed silly, because
2041 it made things harder for people building PCRE on systems that don't run
2042 "configure". I have turned pcre.h into a distributed file, no longer built
2043 by "configure", with the version identification directly included. There is
2044 no longer a pcre.h.in file.
2045
2046 However, this change necessitated a change to the pcre-config script as
2047 well. It is built from pcre-config.in, and one of the substitutions was the
2048 release number. I have updated configure.ac so that ./configure now finds
2049 the release number by grepping pcre.h.
2050
2051 26. Added the ability to run the tests under valgrind.
2052
2053
2054 Version 6.4 05-Sep-05
2055 ---------------------
2056
2057 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2058 "--" to be printed when multiple files were scanned, even when none of the
2059 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2060 consider it to be a bug, and have restored the previous behaviour.
2061
2062 2. A couple of code tidies to get rid of compiler warnings.
2063
2064 3. The pcretest program used to cheat by referring to symbols in the library
2065 whose names begin with _pcre_. These are internal symbols that are not
2066 really supposed to be visible externally, and in some environments it is
2067 possible to suppress them. The cheating is now confined to including
2068 certain files from the library's source, which is a bit cleaner.
2069
2070 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2071 file's purpose clearer.
2072
2073 5. Reorganized pcre_ucp_findchar().
2074
2075
2076 Version 6.3 15-Aug-05
2077 ---------------------
2078
2079 1. The file libpcre.pc.in did not have general read permission in the tarball.
2080
2081 2. There were some problems when building without C++ support:
2082
2083 (a) If C++ support was not built, "make install" and "make test" still
2084 tried to test it.
2085
2086 (b) There were problems when the value of CXX was explicitly set. Some
2087 changes have been made to try to fix these, and ...
2088
2089 (c) --disable-cpp can now be used to explicitly disable C++ support.
2090
2091 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2092 backslash in a target when C++ was disabled. This confuses some
2093 versions of "make", apparently. Using an intermediate variable solves
2094 this. (Same for CPP_LOBJ.)
2095
2096 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2097 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2098 necessary on certain architectures.
2099
2100 4. Added a setting of -export-symbols-regex to the link command to remove
2101 those symbols that are exported in the C sense, but actually are local
2102 within the library, and not documented. Their names all begin with
2103 "_pcre_". This is not a perfect job, because (a) we have to except some
2104 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2105 available (and never for static libraries). I have made a note to try to
2106 find a way round (a) in the future.
2107
2108
2109 Version 6.2 01-Aug-05
2110 ---------------------
2111
2112 1. There was no test for integer overflow of quantifier values. A construction
2113 such as {1111111111111111} would give undefined results. What is worse, if
2114 a minimum quantifier for a parenthesized subpattern overflowed and became
2115 negative, the calculation of the memory size went wrong. This could have
2116 led to memory overwriting.
2117
2118 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2119
2120 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2121 operating environments where this matters.
2122
2123 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2124 PCRE options from within the C++ wrapper.
2125
2126 5. Named capturing subpatterns were not being correctly counted when a pattern
2127 was compiled. This caused two problems: (a) If there were more than 100
2128 such subpatterns, the calculation of the memory needed for the whole
2129 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2130 back references of the form \12, where the number was greater than 9, were
2131 not recognized as back references, even though there were sufficient
2132 previous subpatterns.
2133
2134 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2135 versions of gcc, e.g. 2.95.4.
2136
2137
2138 Version 6.1 21-Jun-05
2139 ---------------------
2140
2141 1. There was one reference to the variable "posix" in pcretest.c that was not
2142 surrounded by "#if !defined NOPOSIX".
2143
2144 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2145 the cross-check on the old pcre_info() function, for the benefit of the
2146 cut-down version of PCRE that is currently imported into Exim.
2147
2148 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2149 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2150 patterns that start (?i) but I don't think that matters. The use of (?i) is
2151 just an example; this all applies to the other options as well.
2152
2153 4. Since libtool seems to echo the compile commands it is issuing, the output
2154 from "make" can be reduced a bit by putting "@" in front of each libtool
2155 compile command.
2156
2157 5. Patch from the folks at Google for configure.in to be a bit more thorough
2158 in checking for a suitable C++ installation before trying to compile the
2159 C++ stuff. This should fix a reported problem when a compiler was present,
2160 but no suitable headers.
2161
2162 6. The man pages all had just "PCRE" as their title. I have changed them to
2163 be the relevant file name. I have also arranged that these names are
2164 retained in the file doc/pcre.txt, which is a concatenation in text format
2165 of all the man pages except the little individual ones for each function.
2166
2167 7. The NON-UNIX-USE file had not been updated for the different set of source
2168 files that come with release 6. I also added a few comments about the C++
2169 wrapper.
2170
2171
2172 Version 6.0 07-Jun-05
2173 ---------------------
2174
2175 1. Some minor internal re-organization to help with my DFA experiments.
2176
2177 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2178 didn't matter for the library itself when fully configured, but did matter
2179 when compiling without UCP support, or within Exim, where the ucp files are
2180 not imported.
2181
2182 3. Refactoring of the library code to split up the various functions into
2183 different source modules. The addition of the new DFA matching code (see
2184 below) to a single monolithic source would have made it really too
2185 unwieldy, quite apart from causing all the code to be include in a
2186 statically linked application, when only some functions are used. This is
2187 relevant even without the DFA addition now that patterns can be compiled in
2188 one application and matched in another.
2189
2190 The downside of splitting up is that there have to be some external
2191 functions and data tables that are used internally in different modules of
2192 the library but which are not part of the API. These have all had their
2193 names changed to start with "_pcre_" so that they are unlikely to clash
2194 with other external names.
2195
2196 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2197 a different (DFA) algorithm. Although it is slower than the original
2198 function, it does have some advantages for certain types of matching
2199 problem.
2200
2201 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2202 including restarting after a partial match.
2203
2204 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2205 defined when compiling for Windows was sent to me. I have put it into the
2206 code, though I have no means of testing or verifying it.
2207
2208 7. Added the pcre_refcount() auxiliary function.
2209
2210 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2211 match before or at the first newline in the subject string. In pcretest,
2212 the /f option on a pattern can be used to set this.
2213
2214 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2215 would behave wrongly. This has been present in PCRE since release 4.0.
2216
2217 10. A number of changes to the pcregrep command:
2218
2219 (a) Refactored how -x works; insert ^(...)$ instead of setting
2220 PCRE_ANCHORED and checking the length, in preparation for adding
2221 something similar for -w.
2222
2223 (b) Added the -w (match as a word) option.
2224
2225 (c) Refactored the way lines are read and buffered so as to have more
2226 than one at a time available.
2227
2228 (d) Implemented a pcregrep test script.
2229
2230 (e) Added the -M (multiline match) option. This allows patterns to match
2231 over several lines of the subject. The buffering ensures that at least
2232 8K, or the rest of the document (whichever is the shorter) is available
2233 for matching (and similarly the previous 8K for lookbehind assertions).
2234
2235 (f) Changed the --help output so that it now says
2236
2237 -w, --word-regex(p)
2238
2239 instead of two lines, one with "regex" and the other with "regexp"
2240 because that confused at least one person since the short forms are the
2241 same. (This required a bit of code, as the output is generated
2242 automatically from a table. It wasn't just a text change.)
2243
2244 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2245 option but starts with a hyphen. Could be a pattern or a path name
2246 starting with a hyphen, for instance.
2247
2248 (h) "-" can be given as a file name to represent stdin.
2249
2250 (i) When file names are being printed, "(standard input)" is used for
2251 the standard input, for compatibility with GNU grep. Previously
2252 "<stdin>" was used.
2253
2254 (j) The option --label=xxx can be used to supply a name to be used for
2255 stdin when file names are being printed. There is no short form.
2256
2257 (k) Re-factored the options decoding logic because we are going to add
2258 two more options that take data. Such options can now be given in four
2259 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2260
2261 (l) Added the -A, -B, and -C options for requesting that lines of context
2262 around matches be printed.
2263
2264 (m) Added the -L option to print the names of files that do not contain
2265 any matching lines, that is, the complement of -l.
2266
2267 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2268 continue to scan other files.
2269
2270 (o) The -s option was incorrectly implemented. For compatibility with other
2271 greps, it now suppresses the error message for a non-existent or non-
2272 accessible file (but not the return code). There is a new option called
2273 -q that suppresses the output of matching lines, which was what -s was
2274 previously doing.
2275
2276 (p) Added --include and --exclude options to specify files for inclusion
2277 and exclusion when recursing.
2278
2279 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2280 Hopefully, it now does.
2281
2282 12. Missing cast in pcre_study().
2283
2284 13. Added an "uninstall" target to the makefile.
2285
2286 14. Replaced "extern" in the function prototypes in Makefile.in with
2287 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2288 world, but is set differently for Windows.
2289
2290 15. Added a second compiling function called pcre_compile2(). The only
2291 difference is that it has an extra argument, which is a pointer to an
2292 integer error code. When there is a compile-time failure, this is set
2293 non-zero, in addition to the error test pointer being set to point to an
2294 error message. The new argument may be NULL if no error number is required
2295 (but then you may as well call pcre_compile(), which is now just a
2296 wrapper). This facility is provided because some applications need a
2297 numeric error indication, but it has also enabled me to tidy up the way
2298 compile-time errors are handled in the POSIX wrapper.
2299
2300 16. Added VPATH=.libs to the makefile; this should help when building with one
2301 prefix path and installing with another. (Or so I'm told by someone who
2302 knows more about this stuff than I do.)
2303
2304 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2305 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2306 match everything, including newlines. This is not POSIX-compatible, but
2307 somebody wanted the feature. From pcretest it can be activated by using
2308 both the P and the s flags.
2309
2310 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2311
2312 19. libpcre.pc was being incorrectly installed as executable.
2313
2314 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2315 it now also looks for '\r' so that it will work unmodified on Windows.
2316
2317 21. Added Google's contributed C++ wrapper to the distribution.
2318
2319 22. Added some untidy missing memory free() calls in pcretest, to keep
2320 Electric Fence happy when testing.
2321
2322
2323
2324 Version 5.0 13-Sep-04
2325 ---------------------
2326
2327 1. Internal change: literal characters are no longer packed up into items
2328 containing multiple characters in a single byte-string. Each character
2329 is now matched using a separate opcode. However, there may be more than one
2330 byte in the character in UTF-8 mode.
2331
2332 2. The pcre_callout_block structure has two new fields: pattern_position and
2333 next_item_length. These contain the offset in the pattern to the next match
2334 item, and its length, respectively.
2335
2336 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2337 insertion of callouts before each pattern item. Added the /C option to
2338 pcretest to make use of this.
2339
2340 4. On the advice of a Windows user, the lines
2341
2342 #if defined(_WIN32) || defined(WIN32)
2343 _setmode( _fileno( stdout ), 0x8000 );
2344 #endif /* defined(_WIN32) || defined(WIN32) */
2345
2346 have been added to the source of pcretest. This apparently does useful
2347 magic in relation to line terminators.
2348
2349 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2350 for the benefit of those environments where the "b" makes a difference.
2351
2352 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2353 to know about it. I have put a hack into configure.in that adds in code
2354 to set GCC=yes if CC=icc. This seems to end up at a point in the
2355 generated configure script that is early enough to affect the setting of
2356 compiler options, which is what is needed, but I have no means of testing
2357 whether it really works. (The user who reported this had patched the
2358 generated configure script, which of course I cannot do.)
2359
2360 LATER: After change 22 below (new libtool files), the configure script
2361 seems to know about icc (and also ecc). Therefore, I have commented out
2362 this hack in configure.in.
2363
2364 7. Added support for pkg-config (2 patches were sent in).
2365
2366 8. Negated POSIX character classes that used a combination of internal tables
2367 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2368 [[:^ascii]]. Typically, they would match almost any characters. The other
2369 POSIX classes were not broken in this way.
2370
2371 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2372 to find the match, as PCRE was deluded into thinking that the match had to
2373 start at the start point or following a newline. The same bug applied to
2374 patterns with negative forward assertions or any backward assertions
2375 preceding ".*" at the start, unless the pattern required a fixed first
2376 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2377
2378 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2379 starting at the last subject character, bytes beyond the end of the subject
2380 string were read.
2381
2382 11. Renamed the variable "class" as "classbits" to make life easier for C++
2383 users. (Previously there was a macro definition, but it apparently wasn't
2384 enough.)
2385
2386 12. Added the new field "tables" to the extra data so that tables can be passed
2387 in at exec time, or the internal tables can be re-selected. This allows
2388 a compiled regex to be saved and re-used at a later time by a different
2389 program that might have everything at different addresses.
2390
2391 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2392 -R library as well as a -L library.
2393
2394 14. The debugging options of pcretest (-d on the command line or D on a
2395 pattern) showed incorrect output for anything following an extended class
2396 that contained multibyte characters and which was followed by a quantifier.
2397
2398 15. Added optional support for general category Unicode character properties
2399 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2400 support. It adds about 90K to the size of the library. The meanings of the
2401 inbuilt class escapes such as \d and \s have NOT been changed.
2402
2403 16. Updated pcredemo.c to include calls to free() to release the memory for the
2404 compiled pattern.
2405
2406 17. The generated file chartables.c was being created in the source directory
2407 instead of in the building directory. This caused the build to fail if the
2408 source directory was different from the building directory, and was
2409 read-only.
2410
2411 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2412 file. No doubt somebody will tell me if they don't make sense... Also added
2413 Dan Mooney's comments about building on OpenVMS.
2414
2415 19. Added support for partial matching via the PCRE_PARTIAL option for
2416 pcre_exec() and the \P data escape in pcretest.
2417
2418 20. Extended pcretest with 3 new pattern features:
2419
2420 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2421 write the compiled pattern to the file whose name is "rest-of-line".
2422 This is a straight binary dump of the data, with the saved pointer to
2423 the character tables forced to be NULL. The study data, if any, is
2424 written too. After writing, pcretest reads a new pattern.
2425
2426 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2427 compiled pattern from the given file. There must not be any
2428 occurrences of "<" in the file name (pretty unlikely); if there are,
2429 pcretest will instead treat the initial "<" as a pattern delimiter.
2430 After reading in the pattern, pcretest goes on to read data lines as
2431 usual.
2432
2433 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2434 and 16-bit fields in a compiled pattern, to simulate a pattern that
2435 was compiled on a host of opposite endianness.
2436
2437 21. The pcre-exec() function can now cope with patterns that were compiled on
2438 hosts of opposite endianness, with this restriction:
2439
2440 As for any compiled expression that is saved and used later, the tables
2441 pointer field cannot be preserved; the extra_data field in the arguments
2442 to pcre_exec() should be used to pass in a tables address if a value
2443 other than the default internal tables were used at compile time.
2444
2445 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2446 now diagnosed as an error. Previously, most of the time, a negative number
2447 would have been treated as zero, but if in addition "ovector" was passed as
2448 NULL, a crash could occur.
2449
2450 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2451 new versions from the libtool 1.5 distribution (the last one is a copy of
2452 a file called libtool.m4). This seems to have fixed the need to patch
2453 "configure" to support Darwin 1.3 (which I used to do). However, I still
2454 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2455 workstation).
2456
2457 24. Changed the PCRE licence to be the more standard "BSD" licence.
2458
2459
2460 Version 4.5 01-Dec-03
2461 ---------------------
2462
2463 1. There has been some re-arrangement of the code for the match() function so
2464 that it can be compiled in a version that does not call itself recursively.
2465 Instead, it keeps those local variables that need separate instances for
2466 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2467 needs to "recurse". Keeping track of where control must go is done by means
2468 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2469 hide most of the details from the main code, and operates only if
2470 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2471 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2472 operating.
2473
2474 To make it easier for callers to provide specially tailored get/free
2475 functions for this usage, two new functions, pcre_stack_malloc, and
2476 pcre_stack_free, are used. They are always called in strict stacking order,
2477 and the size of block requested is always the same.
2478
2479 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2480 PCRE has been compiled to use the stack or the heap for recursion. The
2481 -C option of pcretest uses this to show which version is compiled.
2482
2483 A new data escape \S, is added to pcretest; it causes the amounts of store
2484 obtained and freed by both kinds of malloc/free at match time to be added
2485 to the output.
2486
2487 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2488 what's available on my current Linux desktop machine.
2489
2490 3. When matching a UTF-8 string, the test for a valid string at the start has
2491 been extended. If start_offset is not zero, PCRE now checks that it points
2492 to a byte that is the start of a UTF-8 character. If not, it returns
2493 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2494 this is necessary because there may be backward assertions in the pattern.
2495 When matching the same subject several times, it may save resources to use
2496 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2497
2498 4. The code for checking the validity of UTF-8 strings has been tightened so
2499 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2500 containing "overlong sequences".
2501
2502 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2503 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2504 should have been "&", but it just so happened that all the cases this let
2505 through by mistake were picked up later in the function.
2506
2507 6. I had used a variable called "isblank" - this is a C99 function, causing
2508 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2509
2510 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2511 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2512 so the version is shown; (c) stop "make test" echoing "./RunTest".
2513
2514 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2515
2516 9. The prototype for memmove() for systems that don't have it was using
2517 size_t, but the inclusion of the header that defines size_t was later. I've
2518 moved the #includes for the C headers earlier to avoid this.
2519
2520 10. Added some adjustments to the code to make it easier to compiler on certain
2521 special systems:
2522
2523 (a) Some "const" qualifiers were missing.
2524 (b) Added the macro EXPORT before all exported functions; by default this
2525 is defined to be empty.
2526 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2527 that it reads its output file name as an argument instead of writing
2528 to the standard output and assuming this can be redirected.
2529
2530 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2531 class containing characters with values greater than 255, PCRE compilation
2532 went into a loop.
2533
2534 12. A recursive reference to a subpattern that was within another subpattern
2535 that had a minimum quantifier of zero caused PCRE to crash. For example,
2536 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2537 recursion. If the recursively-called subpattern itself had a zero repeat,
2538 that was OK.
2539
2540 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2541 buffer into which it was copied (for escape processing) was still set at
2542 1024, so long lines caused crashes.
2543
2544 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2545 "internal error: code overflow...". This applied to any character class
2546 that was followed by a possessive quantifier.
2547
2548 15. Modified the Makefile to add libpcre.la as a prerequisite for
2549 libpcreposix.la because I was told this is needed for a parallel build to
2550 work.
2551
2552 16. If a pattern that contained .* following optional items at the start was
2553 studied, the wrong optimizing data was generated, leading to matching
2554 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2555 matching string must start with a or b or c. The correct conclusion for
2556 this pattern is that a match can start with any character.
2557
2558
2559 Version 4.4 13-Aug-03
2560 ---------------------
2561
2562 1. In UTF-8 mode, a character class containing characters with values between
2563 127 and 255 was not handled correctly if the compiled pattern was studied.
2564 In fixing this, I have also improved the studying algorithm for such
2565 classes (slightly).
2566
2567 2. Three internal functions had redundant arguments passed to them. Removal
2568 might give a very teeny performance improvement.
2569
2570 3. Documentation bug: the value of the capture_top field in a callout is *one
2571 more than* the number of the hightest numbered captured substring.
2572
2573 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2574 in incorrectly linking with a previously installed version. They now link
2575 explicitly with libpcre.la.
2576
2577 5. configure.in no longer needs to recognize Cygwin specially.
2578
2579 6. A problem in pcre.in for Windows platforms is fixed.
2580
2581 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2582 pcretest, it used to include the size of the study block as part of its
2583 output. Unfortunately, the structure contains a field that has a different
2584 size on different hardware architectures. This meant that the tests that
2585 showed this size failed. As the block is currently always of a fixed size,
2586 this information isn't actually particularly useful in pcretest output, so
2587 I have just removed it.
2588
2589 8. Three pre-processor statements accidentally did not start in column 1.
2590 Sadly, there are *still* compilers around that complain, even though
2591 standard C has not required this for well over a decade. Sigh.
2592
2593 9. In pcretest, the code for checking callouts passed small integers in the
2594 callout_data field, which is a void * field. However, some picky compilers
2595 complained about the casts involved for this on 64-bit systems. Now
2596 pcretest passes the address of the small integer instead, which should get
2597 rid of the warnings.
2598
2599 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2600 both compile and run time, and gives an error if an invalid UTF-8 sequence
2601 is found. There is a option for disabling this check in cases where the
2602 string is known to be correct and/or the maximum performance is wanted.
2603
2604 11. In response to a bug report, I changed one line in Makefile.in from
2605
2606 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2607 to
2608 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2609
2610 to look similar to other lines, but I have no way of telling whether this
2611 is the right thing to do, as I do not use Windows. No doubt I'll get told
2612 if it's wrong...
2613
2614
2615 Version 4.3 21-May-03
2616 ---------------------
2617
2618 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2619 Makefile.
2620
2621 2. Some refactoring to improve the quality of the code:
2622
2623 (i) The utf8_table... variables are now declared "const".
2624
2625 (ii) The code for \cx, which used the "case flipping" table to upper case
2626 lower case letters, now just substracts 32. This is ASCII-specific,
2627 but the whole concept of \cx is ASCII-specific, so it seems
2628 reasonable.
2629
2630 (iii) PCRE was using its character types table to recognize decimal and
2631 hexadecimal digits in the pattern. This is silly, because it handles
2632 only 0-9, a-f, and A-F, but the character types table is locale-
2633 specific, which means strange things might happen. A private
2634 table is now used for this - though it costs 256 bytes, a table is
2635 much faster than multiple explicit tests. Of course, the standard
2636 character types table is still used for matching digits in subject
2637 strings against \d.
2638
2639 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2640 ending in _t are). So I've renamed it as ESC_tee.
2641
2642 3. The first argument for regexec() in the POSIX wrapper should have been
2643 defined as "const".
2644
2645 4. Changed pcretest to use malloc() for its buffers so that they can be
2646 Electric Fenced for debugging.
2647
2648 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2649 to read one or more bytes before the start of the subject string. Often this
2650 had no effect on PCRE's behaviour, but in some circumstances it could
2651 provoke a segmentation fault.
2652
2653 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2654 to try to read one or more bytes before the start of the subject string.
2655
2656 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2657 UTF-8 support could misbehave in various ways if the subject string
2658 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2659 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2660 back over UTF-8 characters.)
2661
2662
2663 Version 4.2 14-Apr-03
2664 ---------------------
2665
2666 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2667
2668 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2669 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2670 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2671 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2672 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2673 and BUILD_EXEEXT
2674 Note: automatic setting of the BUILD variables is not yet working
2675 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2676 compile-time but not at link-time
2677 [LINK]: use for linking executables only
2678 make different versions for Windows and non-Windows
2679 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2680 libraries
2681 [LINK_FOR_BUILD]: new variable
2682 [OBJEXT]: use throughout
2683 [EXEEXT]: use throughout
2684 <winshared>: new target
2685 <wininstall>: new target
2686 <dftables.o>: use native compiler
2687 <dftables>: use native linker
2688 <install>: handle Windows platform correctly
2689 <clean>: ditto
2690 <check>: ditto
2691 copy DLL to top builddir before testing
2692
2693 As part of these changes, -no-undefined was removed again. This was reported
2694 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2695 in any case.
2696
2697 3. Some tidies to get rid of compiler warnings:
2698
2699 . In the match_data structure, match_limit was an unsigned long int, whereas
2700 match_call_count was an int. I've made them both unsigned long ints.
2701
2702 . In pcretest the fact that a const uschar * doesn't automatically cast to
2703 a void * provoked a warning.
2704
2705 . Turning on some more compiler warnings threw up some "shadow" variables
2706 and a few more missing casts.
2707
2708 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2709 option, a class that contained a single character with a value between 128
2710 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2711
2712 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2713 option, a class that contained several characters, but with at least one
2714 whose value was between 128 and 255 caused PCRE to crash.
2715
2716
2717 Version 4.1 12-Mar-03
2718 ---------------------
2719
2720 1. Compiling with gcc -pedantic found a couple of places where casts were
2721 needed, and a string in dftables.c that was longer than standard compilers are
2722 required to support.
2723
2724 2. Compiling with Sun's compiler found a few more places where the code could
2725 be tidied up in order to avoid warnings.
2726
2727 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2728 first of these names is deprecated in the latest Autoconf in favour of the name
2729 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2730 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2731 analogy I have changed it to CFLAGS_FOR_BUILD.
2732
2733 4. Added -no-undefined to the linking command in the Makefile, because this is
2734 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2735 linking step for the pcreposix library.
2736
2737 5. PCRE was failing to diagnose the case of two named groups with the same
2738 name.
2739
2740 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2741 literal character that is needed in the subject for a match, and scans along to
2742 ensure that it is present before embarking on the full matching process. This
2743 saves time in cases of nested unlimited repeats that are never going to match.
2744 Problem: the scan can take a lot of time if the subject is very long (e.g.
2745 megabytes), thus penalizing straightforward matches. It is now done only if the
2746 amount of subject to be scanned is less than 1000 bytes.
2747
2748 7. A lesser problem with the same optimization is that it was recording the
2749 first character of an anchored pattern as "needed", thus provoking a search
2750 right along the subject, even when the first match of the pattern was going to
2751 fail. The "needed" character is now not set for anchored patterns, unless it
2752 follows something in the pattern that is of non-fixed length. Thus, it still
2753 fulfils its original purpose of finding quick non-matches in cases of nested
2754 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2755
2756
2757 Version 4.0 17-Feb-03
2758 ---------------------
2759
2760 1. If a comment in an extended regex that started immediately after a meta-item
2761 extended to the end of string, PCRE compiled incorrect data. This could lead to
2762 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2763
2764 2. Moved to autoconf 2.53 and libtool 1.4.2.
2765
2766 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2767 the special perltest8 script is no longer needed - all the tests can be run
2768 from a single perltest script.
2769
2770 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2771 by \s. It has now been removed in PCRE. This means it isn't recognized as
2772 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2773 class [:space:] *does* include VT, thereby creating a mess.
2774
2775 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2776 space and tab.
2777
2778 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2779 its new features into the main test script, reducing the number of scripts.
2780
2781 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2782 were backward compatible, and made the (?i) apply to the whole pattern, as if
2783 /i were given. Now it behaves more logically, and applies the option setting
2784 only to what follows. PCRE has been changed to follow suit. However, if it
2785 finds options settings right at the start of the pattern, it extracts them into
2786 the global options, as before. Thus, they show up in the info data.
2787
2788 8. Added support for the \Q...\E escape sequence. Characters in between are
2789 treated as literals. This is slightly different from Perl in that $ and @ are
2790 also handled as literals inside the quotes. In Perl, they will cause variable
2791 interpolation. Note the following examples:
2792
2793 Pattern PCRE matches Perl matches
2794
2795 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2796 \Qabc\$xyz\E abc\$xyz abc\$xyz
2797 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2798
2799 For compatibility with Perl, \Q...\E sequences are recognized inside character
2800 classes as well as outside them.
2801
2802 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2803 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2804 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2805 signed/unsigned warnings.
2806
2807 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2808 option for pcretest, so I've replaced it by a simple function that does just
2809 that job.
2810
2811 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2812 "pcregrep -".
2813
2814 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2815 Java package. This provides some syntactic sugar for simple cases of what my
2816 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2817 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2818 item, you can use this simplified notation. Note that only makes sense with
2819 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2820 greediness, whatever the setting of the PCRE_UNGREEDY option.
2821
2822 13. A change of greediness default within a pattern was not taking effect at
2823 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2824 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2825 was abstracted outside.
2826
2827 14. PCRE now supports the \G assertion. It is true when the current matching
2828 position is at the start point of the match. This differs from \A when the
2829 starting offset is non-zero. Used with the /g option of pcretest (or similar
2830 code), it works in the same way as it does for Perl's /g option. If all
2831 alternatives of a regex begin with \G, the expression is anchored to the start
2832 match position, and the "anchored" flag is set in the compiled expression.
2833
2834 15. Some bugs concerning the handling of certain option changes within patterns
2835 have been fixed. These applied to options other than (?ims). For example,
2836 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2837 been the other way round. Some of this was related to change 7 above.
2838
2839 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2840 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2841 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2842 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2843
2844 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2845 mode. Unlike ".", it always matches newline, whatever the setting of
2846 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2847 assertions. Perl allows it, but it doesn't (in general) work because it can't
2848 calculate the length of the lookbehind. At least, that's the case for Perl
2849 5.8.0 - I've been told they are going to document that it doesn't work in
2850 future.
2851
2852 18. Added an error diagnosis for escapes that PCRE does not support: these are
2853 \L, \l, \N, \P, \p, \U, \u, and \X.
2854
2855 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2856 reading past the end of the pattern in cases such as /[abcd/.
2857
2858 20. PCRE was getting more memory than necessary for patterns with classes that
2859 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2860
2861 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2862 compiling PCRE for use with Virtual Pascal.
2863
2864 22. Small fix to the Makefile to make it work properly if the build is done
2865 outside the source tree.
2866
2867 23. Added a new extension: a condition to go with recursion. If a conditional
2868 subpattern starts with (?(R) the "true" branch is used if recursion has
2869 happened, whereas the "false" branch is used only at the top level.
2870
2871 24. When there was a very long string of literal characters (over 255 bytes
2872 without UTF support, over 250 bytes with UTF support), the computation of how
2873 much memory was required could be incorrect, leading to segfaults or other
2874 strange effects.
2875
2876 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2877 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2878 there was a subsequent back reference to those brackets. This meant that, for
2879 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2880 possible to check for precisely this case. All we can do is abandon the
2881 optimization if .* occurs inside capturing brackets when there are any back
2882 references whatsoever. (See below for a better fix that came later.)
2883
2884 26. The handling of the optimization for finding the first character of a
2885 non-anchored pattern, and for finding a character that is required later in the
2886 match were failing in some cases. This didn't break the matching; it just
2887 failed to optimize when it could. The way this is done has been re-implemented.
2888
2889 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2890
2891 28. Added a new feature that provides some of the functionality that Perl
2892 provides with (?{...}). The facility is termed a "callout". The way it is done
2893 in PCRE is for the caller to provide an optional function, by setting
2894 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2895 global variable. By default it is unset, which disables all calling out. To get
2896 the function called, the regex must include (?C) at appropriate points. This
2897 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2898 This provides a means of identifying different callout points. When PCRE
2899 reaches such a point in the regex, if pcre_callout has been set, the external
2900 function is called. It is provided with data in a structure called
2901 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2902 matching continues; if it returns a non-zero value, the match at the current
2903 point fails. However, backtracking will occur if possible. [This was changed
2904 later and other features added - see item 49 below.]
2905
2906 29. pcretest is upgraded to test the callout functionality. It provides a
2907 callout function that displays information. By default, it shows the start of
2908 the match and the current position in the text. There are some new data escapes
2909 to vary what happens:
2910
2911 \C+ in addition, show current contents of captured substrings
2912 \C- do not supply a callout function
2913 \C!n return 1 when callout number n is reached
2914 \C!n!m return 1 when callout number n is reached for the mth time
2915
2916 30. If pcregrep was called with the -l option and just a single file name, it
2917 output "<stdin>" if a match was found, instead of the file name.
2918
2919 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2920 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2921 pcre_exec(). This saves a malloc/free per call. The default value of
2922 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2923 when configuring.
2924
2925 32. The default maximum size of a compiled pattern is 64K. There have been a
2926 few cases of people hitting this limit. The code now uses macros to handle the
2927 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2928 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2929 configuring. Tests 2 and 5 work only with 2-byte links because they output
2930 debugging information about compiled patterns.
2931
2932 33. Internal code re-arrangements:
2933
2934 (a) Moved the debugging function for printing out a compiled regex into
2935 its own source file (printint.c) and used #include to pull it into
2936 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2937 separate copies.
2938
2939 (b) Defined the list of op-code names for debugging as a macro in
2940 internal.h so that it is next to the definition of the opcodes.
2941
2942 (c) Defined a table of op-code lengths for simpler skipping along compiled
2943 code. This is again a macro in internal.h so that it is next to the
2944 definition of the opcodes.
2945
2946 34. Added support for recursive calls to individual subpatterns, along the
2947 lines of Robin Houston's patch (but implemented somewhat differently).
2948
2949 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2950 allow it to read and process whole directories in Win32. This code was
2951 contributed by Lionel Fourquaux; it has not been tested by me.
2952
2953 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2954 used to name a group. Names consist of alphanumerics and underscores, and must
2955 be unique. Back references use the syntax (?P=name) and recursive calls use
2956 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2957 numbers. The function pcre_fullinfo() can be used after compilation to extract
2958 a name/number map. There are three relevant calls:
2959
2960 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2961 PCRE_INFO_NAMECOUNT yields the number of entries
2962 PCRE_INFO_NAMETABLE yields a pointer to the map.
2963
2964 The map is a vector of fixed-size entries. The size of each entry depends on
2965 the length of the longest name used. The first two bytes of each entry are the
2966 group number, most significant byte first. There follows the corresponding
2967 name, zero terminated. The names are in alphabetical order.
2968
2969 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2970 case instead of 255. Making it the same both with and without UTF-8 support
2971 means that the same test output works with both.
2972
2973 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2974 calling malloc() with a zero argument.
2975
2976 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2977 optimization. I've improved things by keeping a bitmap of backreferences with
2978 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2979 fact referenced, the optimization can be applied. It is unlikely that a
2980 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2981 the match to follow \n) will appear inside brackets with a number greater than
2982 31, but if it does, any back reference > 31 suppresses the optimization.
2983
2984 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2985 of disabling numbered capturing parentheses. Any opening parenthesis that is
2986 not followed by ? behaves as if it were followed by ?: but named parentheses
2987 can still be used for capturing (and they will acquire numbers in the usual
2988 way).
2989
2990 41. Redesigned the return codes from the match() function into yes/no/error so
2991 that errors can be passed back from deep inside the nested calls. A malloc
2992 failure while inside a recursive subpattern call now causes the
2993 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2994
2995 42. It is now possible to set a limit on the number of times the match()
2996 function is called in a call to pcre_exec(). This facility makes it possible to
2997 limit the amount of recursion and backtracking, though not in a directly
2998 obvious way, because the match() function is used in a number of different
2999 circumstances. The count starts from zero for each position in the subject
3000 string (for non-anchored patterns). The default limit is, for compatibility, a
3001 large number, namely 10 000 000. You can change this in two ways:
3002
3003 (a) When configuring PCRE before making, you can use --with-match-limit=n
3004 to set a default value for the compiled library.
3005
3006 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3007 a different value is set. See 45 below.
3008
3009 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3010
3011 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3012 of things that can be changed at compile time. The first argument specifies
3013 what is wanted and the second points to where the information is to be placed.
3014 The current list of available information is:
3015
3016 PCRE_CONFIG_UTF8
3017
3018 The output is an integer that is set to one if UTF-8 support is available;
3019 otherwise it is set to zero.
3020
3021 PCRE_CONFIG_NEWLINE
3022
3023 The output is an integer that it set to the value of the code that is used for
3024 newline. It is either LF (10) or CR (13).
3025
3026 PCRE_CONFIG_LINK_SIZE
3027
3028 The output is an integer that contains the number of bytes used for internal
3029 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3030
3031 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3032
3033 The output is an integer that contains the threshold above which the POSIX
3034 interface uses malloc() for output vectors. See item 31 above.
3035
3036 PCRE_CONFIG_MATCH_LIMIT
3037
3038 The output is an unsigned integer that contains the default limit of the number
3039 of match() calls in a pcre_exec() execution. See 42 above.
3040
3041 44. pcretest has been upgraded by the addition of the -C option. This causes it
3042 to extract all the available output from the new pcre_config() function, and to
3043 output it. The program then exits immediately.
3044
3045 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3046 order to support additional features. One way would have been to define
3047 pcre_exec2() (for example) with extra arguments, but this would not have been
3048 extensible, and would also have required all calls to the original function to
3049 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3050 is used for passing in "extra" data from pcre_study().
3051
3052 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3053 contains the following fields:
3054
3055 flags a bitmap indicating which of the following fields are set
3056 study_data opaque data from pcre_study()
3057 match_limit a way of specifying a limit on match() calls for a specific
3058 call to pcre_exec()
3059 callout_data data for callouts (see 49 below)
3060
3061 The flag bits are also defined in pcre.h, and are
3062
3063 PCRE_EXTRA_STUDY_DATA
3064 PCRE_EXTRA_MATCH_LIMIT
3065 PCRE_EXTRA_CALLOUT_DATA
3066
3067 The pcre_study() function now returns one of these new pcre_extra blocks, with
3068 the actual study data pointed to by the study_data field, and the
3069 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3070 before. That is, this change is entirely upwards-compatible and requires no
3071 change to existing code.
3072
3073 If you want to pass in additional data to pcre_exec(), you can either place it
3074 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3075 block.
3076
3077 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3078 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3079 times with different match limits, until it finds the minimum value needed for
3080 pcre_exec() to complete. The value is then output. This can be instructive; for
3081 most simple matches the number is quite small, but for pathological cases it
3082 gets very large very quickly.
3083
3084 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3085 returns the size of the data block pointed to by the study_data field in a
3086 pcre_extra block, that is, the value that was passed as the argument to
3087 pcre_malloc() when PCRE was getting memory in which to place the information
3088 created by pcre_study(). The fourth argument should point to a size_t variable.
3089 pcretest has been extended so that this information is shown after a successful
3090 pcre_study() call when information about the compiled regex is being displayed.
3091
3092 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3093 because what follows is always an absolute path. (Later: it turns out that this
3094 is more than cosmetic for MinGW, because it doesn't like empty path
3095 components.)
3096
3097 49. Some changes have been made to the callout feature (see 28 above):
3098
3099 (i) A callout function now has three choices for what it returns:
3100
3101 0 => success, carry on matching
3102 > 0 => failure at this point, but backtrack if possible
3103 < 0 => serious error, return this value from pcre_exec()
3104
3105 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3106 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3107 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3108 use by callout functions. It will never be used by PCRE itself.
3109
3110 (ii) The pcre_extra structure (see 45 above) has a void * field called
3111 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3112 pcre_callout_block structure has a field of the same name. The contents of
3113 the field passed in the pcre_extra structure are passed to the callout
3114 function in the corresponding field in the callout block. This makes it
3115 easier to use the same callout-containing regex from multiple threads. For
3116 testing, the pcretest program has a new data escape
3117
3118 \C*n pass the number n (may be negative) as callout_data
3119
3120 If the callout function in pcretest receives a non-zero value as
3121 callout_data, it returns that value.
3122
3123 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3124 there were some redundant $(CFLAGS) in commands that are now specified as
3125 $(LINK), which already includes $(CFLAGS).
3126
3127 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3128 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3129 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3130 one-byte characters throughout. Note that case-insensitive matching applies
3131 only to characters whose values are less than 256. PCRE doesn't support the
3132 notion of cases for higher-valued characters.
3133
3134 (i) A character class whose characters are all within 0-255 is handled as
3135 a bit map, and the map is inverted for negative classes. Previously, a
3136 character > 255 always failed to match such a class; however it should
3137 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3138
3139 (ii) A negated character class with a single character < 255 is coded as
3140 "not this character" (OP_NOT). This wasn't working properly when the test
3141 character was multibyte, either singly or repeated.
3142
3143 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3144 mode, for example: \x{100}{2,3}.
3145
3146 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3147 singly or repeated) now correctly test multibyte characters. However,
3148 PCRE doesn't recognize any characters with values greater than 255 as
3149 digits, spaces, or word characters. Such characters always match \D, \S,
3150 and \W, and never match \d, \s, or \w.
3151
3152 (v) Classes may now contain characters and character ranges with values
3153 greater than 255. For example: [ab\x{100}-\x{400}].
3154
3155 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3156 PCRE in UTF-8 mode.
3157
3158 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3159 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3160 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3161 value.)
3162
3163 53. The single man page has become too large. I have therefore split it up into
3164 a number of separate man pages. These also give rise to individual HTML pages;
3165 these are now put in a separate directory, and there is an index.html page that
3166 lists them all. Some hyperlinking between the pages has been installed.
3167
3168 54. Added convenience functions for handling named capturing parentheses.
3169
3170 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3171 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3172 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3173 are faulted.
3174
3175 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3176 calling configure. These values are used when compiling the dftables.c program
3177 which is run to generate the source of the default character tables. They
3178 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3179 you will need to set these values.
3180
3181 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3182
3183
3184 Version 3.9 02-Jan-02
3185 ---------------------
3186
3187 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3188
3189 2. If --disable-static was given, the building process failed when trying to
3190 build pcretest and pcregrep. (For some reason it was using libtool to compile
3191 them, which is not right, as they aren't part of the library.)
3192
3193
3194 Version 3.8 18-Dec-01
3195 ---------------------
3196
3197 1. The experimental UTF-8 code was completely screwed up. It was packing the
3198 bytes in the wrong order. How dumb can you get?
3199
3200
3201 Version 3.7 29-Oct-01
3202 ---------------------
3203
3204 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3205 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3206 this didn't happen under Solaris 8, where I normally test things.
3207
3208 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3209 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3210 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3211 this fix an hour or so after the initial 3.7 release.)
3212
3213
3214 Version 3.6 23-Oct-01
3215 ---------------------
3216
3217 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3218 offsets passed as NULL with zero offset count.
3219
3220 2. The config.guess and config.sub files had not been updated when I moved to
3221 the latest autoconf.
3222
3223
3224 Version 3.5 15-Aug-01
3225 ---------------------
3226
3227 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3228 had been forgotten.
3229
3230 2. By using declared but undefined structures, we can avoid using "void"
3231 definitions in pcre.h while keeping the internal definitions of the structures
3232 private.
3233
3234 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3235 user point of view, this means that both static and shared libraries are built
3236 by default, but this can be individually controlled. More of the work of
3237 handling this static/shared cases is now inside libtool instead of PCRE's make
3238 file.
3239
3240 4. The pcretest utility is now installed along with pcregrep because it is
3241 useful for users (to test regexs) and by doing this, it automatically gets
3242 relinked by libtool. The documentation has been turned into a man page, so
3243 there are now .1, .txt, and .html versions in /doc.
3244
3245 5. Upgrades to pcregrep:
3246 (i) Added long-form option names like gnu grep.
3247 (ii) Added --help to list all options with an explanatory phrase.
3248 (iii) Added -r, --recursive to recurse into sub-directories.
3249 (iv) Added -f, --file to read patterns from a file.
3250
3251 6. pcre_exec() was referring to its "code" argument before testing that
3252 argument for NULL (and giving an error if it was NULL).
3253
3254 7. Upgraded Makefile.in to allow for compiling in a different directory from
3255 the source directory.
3256
3257 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3258 options bits, the pointer it was passed was to an int instead of to an unsigned
3259 long int. This mattered only on 64-bit systems.
3260
3261 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3262 generated) instead of pcre.in, which it its source. Also made the same change
3263 in several of the .c files.
3264
3265 10. A new release of gcc defines printf() as a macro, which broke pcretest
3266 because it had an ifdef in the middle of a string argument for printf(). Fixed
3267 by using separate calls to printf().
3268
3269 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3270 script, to force use of CR or LF instead of \n in the source. On non-Unix
3271 systems, the value can be set in config.h.
3272
3273 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3274 absolute limit. Changed the text of the error message to make this clear, and
3275 likewise updated the man page.
3276
3277 13. The limit of 99 on the number of capturing subpatterns has been removed.
3278 The new limit is 65535, which I hope will not be a "real" limit.
3279
3280
3281 Version 3.4 22-Aug-00
3282 ---------------------
3283
3284 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3285
3286 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3287
3288
3289 Version 3.3 01-Aug-00
3290 ---------------------
3291
3292 1. If an octal character was given, but the value was greater than \377, it
3293 was not getting masked to the least significant bits, as documented. This could
3294 lead to crashes in some systems.
3295
3296 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3297 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3298
3299 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3300 These just pass their arguments on to (pcre_free)(), but they are provided
3301 because some uses of PCRE bind it to non-C systems that can call its functions,
3302 but cannot call free() or pcre_free() directly.
3303
3304 4. Add "make test" as a synonym for "make check". Corrected some comments in
3305 the Makefile.
3306
3307 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3308 Makefile.
3309
3310 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3311 command called pgrep for grepping around the active processes.
3312
3313 7. Added the beginnings of support for UTF-8 character strings.
3314
3315 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3316 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3317 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3318 out for the ar command.)
3319
3320
3321 Version 3.2 12-May-00
3322 ---------------------
3323
3324 This is purely a bug fixing release.
3325
3326 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3327 of ZA. This was just one example of several cases that could provoke this bug,
3328 which was introduced by change 9 of version 2.00. The code for breaking
3329 infinite loops after an iteration that matches an empty string was't working
3330 correctly.
3331
3332 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3333 when matched against abbab (for example). After matching an empty string, it
3334 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3335 caused it to match further down the string than it should.
3336
3337 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3338 was there because it doesn't seem to be needed, and it causes trouble on some
3339 systems, as it is not a Standard C header. It has been removed.
3340
3341 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3342 were reported on the Macintosh. The changes were from
3343
3344 while ((c = *(++ptr)) != 0 && c != '\n');
3345 to
3346 while ((c = *(++ptr)) != 0 && c != '\n') ;
3347
3348 Totally extraordinary, but if that's what it takes...
3349
3350 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3351 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3352 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3353 assumes the way PCRE uses memmove() (always moving upwards).
3354
3355 6. PCRE is being used in one environment where strchr() is not available. There
3356 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3357 faster code anyway.
3358
3359
3360 Version 3.1 09-Feb-00
3361 ---------------------
3362
3363 The only change in this release is the fixing of some bugs in Makefile.in for
3364 the "install" target:
3365
3366 (1) It was failing to install pcreposix.h.
3367
3368 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3369
3370
3371 Version 3.0 01-Feb-00
3372 ---------------------
3373
3374 1. Add support for the /+ modifier to perltest (to output $` like it does in
3375 pcretest).
3376
3377 2. Add support for the /g modifier to perltest.
3378
3379 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3380 matches null strings.
3381
3382 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3383 pattern. Perl treats empty patterns specially - it reuses the most recent
3384 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3385 effect.
3386
3387 5. The POSIX interface was broken in that it was just handing over the POSIX
3388 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3389 required a bigger vector, with some working space on the end. This means that
3390 the POSIX wrapper now has to get and free some memory, and copy the results.
3391
3392 6. Added some simple autoconf support, placing the test data and the
3393 documentation in separate directories, re-organizing some of the
3394 information files, and making it build pcre-config (a GNU standard). Also added
3395 libtool support for building PCRE as a shared library, which is now the
3396 default.
3397
3398 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3399 09 are not valid octal constants. Single digits will be used for minor values
3400 less than 10.
3401
3402 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3403 existing programs that set these in the POSIX interface can use PCRE without
3404 modification.
3405
3406 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3407 return all that pcre_info() returns, plus additional data. The pcre_info()
3408 function is retained for compatibility, but is considered to be obsolete.
3409
3410 10. Added experimental recursion feature (?R) to handle one common case that
3411 Perl 5.6 will be able to do with (?p{...}).
3412
3413 11. Added support for POSIX character classes like [:alpha:], which Perl is
3414 adopting.
3415
3416
3417 Version 2.08 31-Aug-99
3418 ----------------------
3419
3420 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3421 trying to match at the startoffset position, but instead was moving forward to
3422 the next newline as if a previous match had failed.
3423
3424 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3425 and could get into a loop if a null string was matched other than at the start
3426 of the subject.
3427
3428 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3429 be distinguished at compile time, and for completeness also added PCRE_DATE.
3430
3431 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3432 in GnuWin32 environments.
3433
3434
3435 Version 2.07 29-Jul-99
3436 ----------------------
3437
3438 1. The documentation is now supplied in plain text form and HTML as well as in
3439 the form of man page sources.
3440
3441 2. C++ compilers don't like assigning (void *) values to other pointer types.
3442 In particular this affects malloc(). Although there is no problem in Standard
3443 C, I've put in casts to keep C++ compilers happy.
3444
3445 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3446 should be (const char *).
3447
3448 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3449 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3450 However, I haven't made this a standard facility. The documentation doesn't
3451 mention it, and the Makefile doesn't support it.
3452
3453 5. The Makefile now contains an "install" target, with editable destinations at
3454 the top of the file. The pcretest program is not installed.
3455
3456 6. pgrep -V now gives the PCRE version number and date.
3457
3458 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3459 causing the entire string to be ignored, instead of just the last character.
3460
3461 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3462 non-matching string, it can take a very, very long time, even for strings of
3463 quite modest length, because of the nested recursion. PCRE now does better in
3464 some of these cases. It does this by remembering the last required literal
3465 character in the pattern, and pre-searching the subject to ensure it is present
3466 before running the real match. In other words, it applies a heuristic to detect
3467 some types of certain failure quickly, and in the above example, if presented
3468 with a string that has no trailing " it gives "no match" very quickly.
3469
3470 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3471 other alternatives are tried instead.
3472
3473
3474 Version 2.06 09-Jun-99
3475 ----------------------
3476
3477 1. Change pcretest's output for amount of store used to show just the code
3478 space, because the remainder (the data block) varies in size between 32-bit and
3479 64-bit systems.
3480
3481 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3482 start matching at. This allows lookbehinds to work when searching for multiple
3483 occurrences in a string.
3484
3485 3. Added additional options to pcretest for testing multiple occurrences:
3486
3487 /+ outputs the rest of the string that follows a match
3488 /g loops for multiple occurrences, using the new startoffset argument
3489 /G loops for multiple occurrences by passing an incremented pointer
3490
3491 4. PCRE wasn't doing the "first character" optimization for patterns starting
3492 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3493 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3494 the letter 'x'. On long subject strings, this gives a significant speed-up.
3495
3496
3497 Version 2.05 21-Apr-99
3498 ----------------------
3499
3500 1. Changed the type of magic_number from int to long int so that it works
3501 properly on 16-bit systems.
3502
3503 2. Fixed a bug which caused patterns starting with .* not to work correctly
3504 when the subject string contained newline characters. PCRE was assuming
3505 anchoring for such patterns in all cases, which is not correct because .* will
3506 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3507 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3508 must be retried after every newline in the subject.
3509
3510
3511 Version 2.04 18-Feb-99
3512 ----------------------
3513
3514 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3515 computation of the store needed to hold the pattern was incorrect (too large).
3516 If such patterns were nested a few deep, this could multiply and become a real
3517 problem.
3518
3519 2. Added /M option to pcretest to show the memory requirement of a specific
3520 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3521
3522 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3523 compiled in such a way that the backtracking after subsequent failure was
3524 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3525 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3526
3527
3528 Version 2.03 02-Feb-99
3529 ----------------------
3530
3531 1. Fixed typo and small mistake in man page.
3532
3533 2. Added 4th condition (GPL supersedes if conflict) and created separate
3534 LICENCE file containing the conditions.
3535
3536 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3537 Perl, that is the internal \ allows the delimiter to be included in the
3538 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3539 the final delimiter, add \ to the end of the pattern (to test the error).
3540
3541 4. Added the convenience functions for extracting substrings after a successful
3542 match. Updated pcretest to make it able to test these functions.
3543
3544
3545 Version 2.02 14-Jan-99
3546 ----------------------
3547
3548 1. Initialized the working variables associated with each extraction so that
3549 their saving and restoring doesn't refer to uninitialized store.
3550
3551 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3552 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3553 fix the problem.
3554
3555 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3556 calls, and wasn't printing the correct value for compiling calls. Increased the
3557 default value of LOOPREPEAT, and the number of significant figures in the
3558 times.
3559
3560 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3561
3562 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3563 a building problem on Windows NT with a FAT file system.
3564
3565
3566 Version 2.01 21-Oct-98
3567 ----------------------
3568
3569 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3570 to character tables built by pcre_maketables() in the current locale. If NULL
3571 is passed, the default tables are used.
3572
3573
3574 Version 2.00 24-Sep-98
3575 ----------------------
3576
3577 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3578 it any more.
3579
3580 2. Allow quantification of (?>) groups, and make it work correctly.
3581
3582 3. The first character computation wasn't working for (?>) groups.
3583
3584 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3585 end of the subject) and add 5.005's \z, which really does match only at the
3586 very end of the subject.
3587
3588 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3589
3590 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3591 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3592 localized options. All options to pcre_study() were also removed.
3593
3594 7. Add other new features from 5.005:
3595
3596 $(?<= positive lookbehind
3597 $(?<! negative lookbehind
3598 (?imsx-imsx) added the unsetting capability
3599 such a setting is global if at outer level; local otherwise
3600 (?imsx-imsx:) non-capturing groups with option setting
3601 (?(cond)re|re) conditional pattern matching
3602
3603 A backreference to itself in a repeated group matches the previous
3604 captured string.
3605
3606 8. General tidying up of studying (both automatic and via "study")
3607 consequential on the addition of new assertions.
3608
3609 9. As in 5.005, unlimited repeated groups that could match an empty substring
3610 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3611 runtime if any iteration does actually match an empty substring.
3612
3613 10. Include the RunTest script in the distribution.
3614
3615 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3616 discrepancies, some of which were old and were also with respect to 5.004. They
3617 have now been fixed.
3618
3619
3620 Version 1.09 28-Apr-98
3621 ----------------------
3622
3623 1. A negated single character class followed by a quantifier with a minimum
3624 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3625 program crashes, or just wrong answers. This did not apply to negated classes
3626 containing more than one character, or to minima other than one.
3627
3628
3629 Version 1.08 27-Mar-98
3630 ----------------------
3631
3632 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3633
3634 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3635 latter must appear before anything that relies on it in the pattern.
3636
3637
3638 Version 1.07 16-Feb-98
3639 ----------------------
3640
3641 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3642 repeat of a potentially empty string).
3643
3644
3645 Version 1.06 23-Jan-98
3646 ----------------------
3647
3648 1. Added Markus Oberhumer's little patches for C++.
3649
3650 2. Literal strings longer than 255 characters were broken.
3651
3652
3653 Version 1.05 23-Dec-97
3654 ----------------------
3655
3656 1. Negated character classes containing more than one character were failing if
3657 PCRE_CASELESS was set at run time.
3658
3659
3660 Version 1.04 19-Dec-97
3661 ----------------------
3662
3663 1. Corrected the man page, where some "const" qualifiers had been omitted.
3664
3665 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3666 input syntax.
3667
3668 3. Fixed memory leak which occurred when a regex with back references was
3669 matched with an offsets vector that wasn't big enough. The temporary memory
3670 that is used in this case wasn't being freed if the match failed.
3671
3672 4. Tidied pcretest to ensure it frees memory that it gets.
3673
3674 5. Temporary memory was being obtained in the case where the passed offsets
3675 vector was exactly big enough.
3676
3677 6. Corrected definition of offsetof() from change 5 below.
3678
3679 7. I had screwed up change 6 below and broken the rules for the use of
3680 setjmp(). Now fixed.
3681
3682
3683 Version 1.03 18-Dec-97
3684 ----------------------
3685
3686 1. A erroneous regex with a missing opening parenthesis was correctly
3687 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3688 on some systems.
3689
3690 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3691 it was reported that one broken compiler failed on the former because "code" is
3692 also an independent variable.
3693
3694 3. The erroneous regex a[]b caused an array overrun reference.
3695
3696 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3697 fail on data ending with that character. (It was going on too far, and checking
3698 the next character, typically a binary zero.) This was specific to the
3699 optimized code for single-character negative classes.
3700
3701 5. Added a contributed patch from the TIN world which does the following:
3702
3703 + Add an undef for memmove, in case the the system defines a macro for it.
3704
3705 + Add a definition of offsetof(), in case there isn't one. (I don't know
3706 the reason behind this - offsetof() is part of the ANSI standard - but
3707 it does no harm).
3708
3709 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3710 most of the places where whitespace preceded '#'. I have given up and
3711 allowed the remaining 2 cases to be at the margin.
3712
3713 + Rename some variables in pcre to eliminate shadowing. This seems very
3714 pedantic, but does no harm, of course.
3715
3716 6. Moved the call to setjmp() into its own function, to get rid of warnings
3717 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3718
3719 7. Constructs such as \d{8,} were compiling into the equivalent of
3720 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3721 outcome, but in this particular case used more store than had been allocated,
3722 which caused the bug to be discovered because it threw up an internal error.
3723
3724 8. The debugging code in both pcre and pcretest for outputting the compiled
3725 form of a regex was going wrong in the case of back references followed by
3726 curly-bracketed repeats.
3727
3728
3729 Version 1.02 12-Dec-97
3730 ----------------------
3731
3732 1. Typos in pcre.3 and comments in the source fixed.
3733
3734 2. Applied a contributed patch to get rid of places where it used to remove
3735 'const' from variables, and fixed some signed/unsigned and uninitialized
3736 variable warnings.
3737
3738 3. Added the "runtest" target to Makefile.
3739
3740 4. Set default compiler flag to -O2 rather than just -O.
3741
3742
3743 Version 1.01 19-Nov-97
3744 ----------------------
3745
3746 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3747 like /([ab]*)*/, that is, for classes with more than one character in them.
3748
3749 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3750 as /((?>a*))*/ (a PCRE_EXTRA facility).
3751
3752
3753 Version 1.00 18-Nov-97
3754 ----------------------
3755
3756 1. Added compile-time macros to support systems such as SunOS4 which don't have
3757 memmove() or strerror() but have other things that can be used instead.
3758
3759 2. Arranged that "make clean" removes the executables.
3760
3761
3762 Version 0.99 27-Oct-97
3763 ----------------------
3764
3765 1. Fixed bug in code for optimizing classes with only one character. It was
3766 initializing a 32-byte map regardless, which could cause it to run off the end
3767 of the memory it had got.
3768
3769 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3770
3771
3772 Version 0.98 22-Oct-97
3773 ----------------------
3774
3775 1. Fixed bug in code for handling temporary memory usage when there are more
3776 back references than supplied space in the ovector. This could cause segfaults.
3777
3778
3779 Version 0.97 21-Oct-97
3780 ----------------------
3781
3782 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3783
3784 2. Optimized negated single characters not to use a bit map.
3785
3786 3. Brought error texts together as macro definitions; clarified some of them;
3787 fixed one that was wrong - it said "range out of order" when it meant "invalid
3788 escape sequence".
3789
3790 4. Changed some char * arguments to const char *.
3791
3792 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3793
3794 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3795 pcretest.
3796
3797
3798 Version 0.96 16-Oct-97
3799 ----------------------
3800
3801 1. Added a simple "pgrep" utility to the distribution.
3802
3803 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3804 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3805 where "ddd" means "one or more decimal digits".
3806
3807 3. Fixed serious bug. If a pattern had a back reference, but the call to
3808 pcre_exec() didn't supply a large enough ovector to record the related
3809 identifying subpattern, the match always failed. PCRE now remembers the number
3810 of the largest back reference, and gets some temporary memory in which to save
3811 the offsets during matching if necessary, in order to ensure that
3812 backreferences always work.
3813
3814 4. Increased the compatibility with Perl in a number of ways:
3815
3816 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3817 to request this handling. The option can be set at compile or exec time.
3818
3819 (b) $ matches before a terminating newline by default; an option
3820 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3821 mode). The option can be set at compile or exec time.
3822
3823 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3824 the same as Perl's. If the decimal number it represents is less than 10
3825 or there aren't that many previous left capturing parentheses, an octal
3826 escape is read. Inside a character class, it's always an octal escape,
3827 even if it is a single digit.
3828
3829 (d) An escaped but undefined alphabetic character is taken as a literal,
3830 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3831 escapes.
3832
3833 (e) {0} is now permitted. (The previous item is removed from the compiled
3834 pattern).
3835
3836 5. Changed all the names of code files so that the basic parts are no longer
3837 than 10 characters, and abolished the teeny "globals.c" file.
3838
3839 6. Changed the handling of character classes; they are now done with a 32-byte
3840 bit map always.
3841
3842 7. Added the -d and /D options to pcretest to make it possible to look at the
3843 internals of compilation without having to recompile pcre.
3844
3845
3846 Version 0.95 23-Sep-97
3847 ----------------------
3848
3849 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3850 \x20 at the start of a run of normal characters. These were being treated as
3851 real characters, instead of the source characters being re-checked.
3852
3853
3854 Version 0.94 18-Sep-97
3855 ----------------------
3856
3857 1. The functions are now thread-safe, with the caveat that the global variables
3858 containing pointers to malloc() and free() or alternative functions are the
3859 same for all threads.
3860
3861 2. Get pcre_study() to generate a bitmap of initial characters for non-
3862 anchored patterns when this is possible, and use it if passed to pcre_exec().
3863
3864
3865 Version 0.93 15-Sep-97
3866 ----------------------
3867
3868 1. /(b)|(:+)/ was computing an incorrect first character.
3869
3870 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3871 but not actually doing anything yet.
3872
3873 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3874 as Perl does (e.g. [-az] or [az-]).
3875
3876 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3877 all possible positions.
3878
3879 5. Split up into different modules to avoid including unneeded functions in a
3880 compiled binary. However, compile and exec are still in one module. The "study"
3881 function is split off.
3882
3883 6. The character tables are now in a separate module whose source is generated
3884 by an auxiliary program - but can then be edited by hand if required. There are
3885 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3886 toupper() in the code.
3887
3888 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3889 make them global. Abolish the function for setting them, as the caller can now
3890 set them directly.
3891
3892
3893 Version 0.92 11-Sep-97
3894 ----------------------
3895
3896 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3897 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3898
3899 2. Caseless matching was not working in character classes if the characters in
3900 the pattern were in upper case.
3901
3902 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3903
3904 4. Make PCRE_ANCHORED public and accept as a compile option.
3905
3906 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3907 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3908 pass them.
3909
3910 6. Give an error if bad option bits passed at compile or run time.
3911
3912 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3913 pcretest to cause it to pass that flag.
3914
3915 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3916 options, and the first character, if set.
3917
3918 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3919
3920
3921 Version 0.91 10-Sep-97
3922 ----------------------
3923
3924 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3925 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3926
3927 2. PCRE was looping on encountering an indefinitely repeated back reference to
3928 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3929 Perl does - treats the match as successful.
3930
3931 ****

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5