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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 501 - (show annotations) (download)
Sun Mar 7 11:49:54 2010 UTC (4 years, 4 months ago) by ph10
File size: 184033 byte(s)
Preparation code for future (*MARK) support.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12