ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 519 - (show annotations)
Fri May 21 16:43:17 2010 UTC (11 years, 4 months ago) by ph10
File size: 186489 byte(s)
Added --line-buffered to pcregrep.
1 ChangeLog for PCRE
2 ------------------
4 Version 8.10 03 May-2010
5 ------------------------
7 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
10 2. (*ACCEPT) was not working when inside an atomic group.
12 3. Inside a character class, \B is treated as a literal by default, but
13 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
14 causes the error). The code is unchanged, but I tidied the documentation.
16 4. Inside a character class, PCRE always treated \R and \X as literals,
17 whereas Perl faults them if its -w option is set. I have changed PCRE so
18 that it faults them when PCRE_EXTRA is set.
20 5. Added support for \N, which always matches any character other than
21 newline. (It is the same as "." when PCRE_DOTALL is not set.)
23 6. When compiling pcregrep with newer versions of gcc which may have
24 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
25 declared with attribute warn_unused_result" were given. Just casting the
26 result to (void) does not stop the warnings; a more elaborate fudge is
27 needed. I've used a macro to implement this.
29 7. Minor change to pcretest.c to avoid a compiler warning.
31 8. Added four artifical Unicode properties to help with an option to make
32 \s etc use properties (see next item). The new properties are: Xan
33 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
35 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
36 use Unicode properties. (*UCP) at the start of a pattern can be used to set
37 this option. Modified pcretest to add /W to test this facility. Added
38 REG_UCP to make it available via the POSIX interface.
40 10. Added --line-buffered to pcregrep.
43 Version 8.02 19-Mar-2010
44 ------------------------
46 1. The Unicode data tables have been updated to Unicode 5.2.0.
48 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
49 configured.
51 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
52 original author of that file, following a query about its status.
54 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
55 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
57 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
58 quantifier applied to a forward-referencing subroutine call, could compile
59 incorrect code or give the error "internal error: previously-checked
60 referenced subpattern not found".
62 6. Both MS Visual Studio and Symbian OS have problems with initializing
63 variables to point to external functions. For these systems, therefore,
64 pcre_malloc etc. are now initialized to local functions that call the
65 relevant global functions.
67 7. There were two entries missing in the vectors called coptable and poptable
68 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
69 I've fixed the data, and added a kludgy way of testing at compile time that
70 the lengths are correct (equal to the number of opcodes).
72 8. Following on from 7, I added a similar kludge to check the length of the
73 eint vector in pcreposix.c.
75 9. Error texts for pcre_compile() are held as one long string to avoid too
76 much relocation at load time. To find a text, the string is searched,
77 counting zeros. There was no check for running off the end of the string,
78 which could happen if a new error number was added without updating the
79 string.
81 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
83 11. \K was not working if it appeared in an atomic group or in a group that
84 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
85 \K is "not well defined" if used in an assertion. PCRE now accepts it if
86 the assertion is positive, but not if it is negative.
88 12. Change 11 fortuitously reduced the size of the stack frame used in the
89 "match()" function of pcre_exec.c by one pointer. Forthcoming
90 implementation of support for (*MARK) will need an extra pointer on the
91 stack; I have reserved it now, so that the stack frame size does not
92 decrease.
94 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
95 item in branch that calls a recursion is a subroutine call - as in the
96 second branch in the above example - was incorrectly given the compile-
97 time error "recursive call could loop indefinitely" because pcre_compile()
98 was not correctly checking the subroutine for matching a non-empty string.
100 14. The checks for overrunning compiling workspace could trigger after an
101 overrun had occurred. This is a "should never occur" error, but it can be
102 triggered by pathological patterns such as hundreds of nested parentheses.
103 The checks now trigger 100 bytes before the end of the workspace.
105 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
108 Version 8.01 19-Jan-2010
109 ------------------------
111 1. If a pattern contained a conditional subpattern with only one branch (in
112 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
113 computed the wrong minimum data length (which is of course zero for such
114 subpatterns). This could cause incorrect "no match" results.
116 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
117 the pattern is reset in the first branch, pcre_compile() failed with
118 "internal error: code overflow at offset...". This happened only when
119 the reset was to the original external option setting. (An optimization
120 abstracts leading options settings into an external setting, which was the
121 cause of this.)
123 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
124 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
125 assertion pattern did not match (meaning that the assertion was true), it
126 was incorrectly treated as false if the SKIP had been reached during the
127 matching. This also applied to assertions used as conditions.
129 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
130 assertion subpattern, including such a pattern used as a condition,
131 unpredictable results occurred, instead of the error return
134 5. The C++ GlobalReplace function was not working like Perl for the special
135 situation when an empty string is matched. It now does the fancy magic
136 stuff that is necessary.
138 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
139 removed. (These were left over from very, very early versions of PCRE.)
141 7. Some cosmetic changes to the code to make life easier when compiling it
142 as part of something else:
144 (a) Change DEBUG to PCRE_DEBUG.
146 (b) In pcre_compile(), rename the member of the "branch_chain" structure
147 called "current" as "current_branch", to prevent a collision with the
148 Linux macro when compiled as a kernel module.
150 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
151 prevent a collision with the Linux macro when compiled as a kernel
152 module.
154 8. In pcre_compile() there are some checks for integer overflows that used to
155 cast potentially large values to (double). This has been changed to that
156 when building, a check for int64_t is made, and if it is found, it is used
157 instead, thus avoiding the use of floating point arithmetic. (There is no
158 other use of FP in PCRE.) If int64_t is not found, the fallback is to
159 double.
161 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
162 2005 (difference between two addresses compared to an unsigned value).
164 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
165 custom one, because of the following reported problem in Windows:
167 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
168 under Win32.
169 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
170 therefore missing the function definition.
171 - The compiler thus generates a "C" signature for the test function.
172 - The linker fails to find the "C" function.
173 - PCRE fails to configure if asked to do so against libbz2.
175 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
176 messages were output:
178 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
179 rerunning libtoolize, to keep the correct libtool macros in-tree.
180 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
182 I have done both of these things.
184 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
185 most of the time, it *can* run out if it is given a pattern that contains a
186 runaway infinite recursion. I updated the discussion in the pcrestack man
187 page.
189 13. Now that we have gone to the x.xx style of version numbers, the minor
190 version may start with zero. Using 08 or 09 is a bad idea because users
191 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
192 interpreted as invalid octal numbers. I've updated the previous comment in
193 configure.ac, and also added a check that gives an error if 08 or 09 are
194 used.
196 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
197 causing partial matching to fail when the end of the subject matched \W
198 in a UTF-8 pattern where \W was quantified with a minimum of 3.
200 15. There were some discrepancies between the declarations in pcre_internal.h
201 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
202 their definitions. The declarations used "const uschar *" and the
203 definitions used USPTR. Even though USPTR is normally defined as "const
204 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
205 reported that: "This difference in casting confuses some C++ compilers, for
206 example, SunCC recognizes above declarations as different functions and
207 generates broken code for hbpcre." I have changed the declarations to use
208 USPTR.
210 16. GNU libtool is named differently on some systems. The autogen.sh script now
211 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
212 (FreeBSD).
214 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
215 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
216 comment: "Figure out how to create a longlong from a string: strtoll and
217 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
218 instance, but it only takes 2 args instead of 3!"
220 18. A subtle bug concerned with back references has been fixed by a change of
221 specification, with a corresponding code fix. A pattern such as
222 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
223 refers, was giving matches when it shouldn't. For example, xa=xaaa would
224 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
225 same bug. Such groups have to be quantified to be useful, or contained
226 inside another quantified group. (If there's no repetition, the reference
227 can never match.) The problem arises because, having left the group and
228 moved on to the rest of the pattern, a later failure that backtracks into
229 the group uses the captured value from the final iteration of the group
230 rather than the correct earlier one. I have fixed this in PCRE by forcing
231 any group that contains a reference to itself to be an atomic group; that
232 is, there cannot be any backtracking into it once it has completed. This is
233 similar to recursive and subroutine calls.
236 Version 8.00 19-Oct-09
237 ----------------------
239 1. The table for translating pcre_compile() error codes into POSIX error codes
240 was out-of-date, and there was no check on the pcre_compile() error code
241 being within the table. This could lead to an OK return being given in
242 error.
244 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
245 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
246 in a Windows environment.
248 3. The pcregrep --count option prints the count for each file even when it is
249 zero, as does GNU grep. However, pcregrep was also printing all files when
250 --files-with-matches was added. Now, when both options are given, it prints
251 counts only for those files that have at least one match. (GNU grep just
252 prints the file name in this circumstance, but including the count seems
253 more useful - otherwise, why use --count?) Also ensured that the
254 combination -clh just lists non-zero counts, with no names.
256 4. The long form of the pcregrep -F option was incorrectly implemented as
257 --fixed_strings instead of --fixed-strings. This is an incompatible change,
258 but it seems right to fix it, and I didn't think it was worth preserving
259 the old behaviour.
261 5. The command line items --regex=pattern and --regexp=pattern were not
262 recognized by pcregrep, which required --regex pattern or --regexp pattern
263 (with a space rather than an '='). The man page documented the '=' forms,
264 which are compatible with GNU grep; these now work.
266 6. No libpcreposix.pc file was created for pkg-config; there was just
267 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
269 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
270 when UCP support is not needed, by modifying the Python script that
271 generates it from Unicode data files. This should not matter if the module
272 is correctly used as a library, but I received one complaint about 50K of
273 unwanted data. My guess is that the person linked everything into his
274 program rather than using a library. Anyway, it does no harm.
276 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
277 was a minimum greater than 1 for a wide character in a possessive
278 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
279 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
280 character. Chaos in the form of incorrect output or a compiling loop could
281 result.
283 9. The restrictions on what a pattern can contain when partial matching is
284 requested for pcre_exec() have been removed. All patterns can now be
285 partially matched by this function. In addition, if there are at least two
286 slots in the offset vector, the offset of the earliest inspected character
287 for the match and the offset of the end of the subject are set in them when
288 PCRE_ERROR_PARTIAL is returned.
290 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
291 synonymous with PCRE_PARTIAL, for backwards compatibility, and
292 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
293 and may be more useful for multi-segment matching.
295 11. Partial matching with pcre_exec() is now more intuitive. A partial match
296 used to be given if ever the end of the subject was reached; now it is
297 given only if matching could not proceed because another character was
298 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
299 string "Z", which now yields "no match" instead of "partial match". In the
300 case of pcre_dfa_exec(), "no match" is given if every matching path for the
301 final character ended with (*FAIL).
303 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
304 if the pattern had a "must contain" character that was already found in the
305 earlier partial match, unless partial matching was again requested. For
306 example, with the pattern /dog.(body)?/, the "must contain" character is
307 "g". If the first part-match was for the string "dog", restarting with
308 "sbody" failed. This bug has been fixed.
310 13. The string returned by pcre_dfa_exec() after a partial match has been
311 changed so that it starts at the first inspected character rather than the
312 first character of the match. This makes a difference only if the pattern
313 starts with a lookbehind assertion or \b or \B (\K is not supported by
314 pcre_dfa_exec()). It's an incompatible change, but it makes the two
315 matching functions compatible, and I think it's the right thing to do.
317 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
318 so that the demonstration program is easily available in environments where
319 PCRE has not been installed from source.
321 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
322 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
323 library.
325 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
326 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
327 is not the first non-POSIX option to be added. Clearly some people find
328 these options useful.
330 17. If a caller to the POSIX matching function regexec() passes a non-zero
331 value for nmatch with a NULL value for pmatch, the value of
332 nmatch is forced to zero.
334 18. RunGrepTest did not have a test for the availability of the -u option of
335 the diff command, as RunTest does. It now checks in the same way as
336 RunTest, and also checks for the -b option.
338 19. If an odd number of negated classes containing just a single character
339 interposed, within parentheses, between a forward reference to a named
340 subpattern and the definition of the subpattern, compilation crashed with
341 an internal error, complaining that it could not find the referenced
342 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
343 [The bug was that it was starting one character too far in when skipping
344 over the character class, thus treating the ] as data rather than
345 terminating the class. This meant it could skip too much.]
347 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
348 /g option in pcretest when the pattern contains \K, which makes it possible
349 to have an empty string match not at the start, even when the pattern is
350 anchored. Updated pcretest and pcredemo to use this option.
352 21. If the maximum number of capturing subpatterns in a recursion was greater
353 than the maximum at the outer level, the higher number was returned, but
354 with unset values at the outer level. The correct (outer level) value is
355 now given.
357 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
358 PCRE did not set those parentheses (unlike Perl). I have now found a way to
359 make it do so. The string so far is captured, making this feature
360 compatible with Perl.
362 23. The tests have been re-organized, adding tests 11 and 12, to make it
363 possible to check the Perl 5.10 features against Perl 5.10.
365 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
366 pattern matches a fixed length string. PCRE did not allow this; now it
367 does. Neither allows recursion.
369 25. I finally figured out how to implement a request to provide the minimum
370 length of subject string that was needed in order to match a given pattern.
371 (It was back references and recursion that I had previously got hung up
372 on.) This code has now been added to pcre_study(); it finds a lower bound
373 to the length of subject needed. It is not necessarily the greatest lower
374 bound, but using it to avoid searching strings that are too short does give
375 some useful speed-ups. The value is available to calling programs via
376 pcre_fullinfo().
378 26. While implementing 25, I discovered to my embarrassment that pcretest had
379 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
380 study optimizations had never been tested with that matching function.
381 Oops. What is worse, even when it was passed study data, there was a bug in
382 pcre_dfa_exec() that meant it never actually used it. Double oops. There
383 were also very few tests of studied patterns with pcre_dfa_exec().
385 27. If (?| is used to create subpatterns with duplicate numbers, they are now
386 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
387 on the other side of the coin, they are no longer allowed to have different
388 names, because these cannot be distinguished in PCRE, and this has caused
389 confusion. (This is a difference from Perl.)
391 28. When duplicate subpattern names are present (necessarily with different
392 numbers, as required by 27 above), and a test is made by name in a
393 conditional pattern, either for a subpattern having been matched, or for
394 recursion in such a pattern, all the associated numbered subpatterns are
395 tested, and the overall condition is true if the condition is true for any
396 one of them. This is the way Perl works, and is also more like the way
397 testing by number works.
400 Version 7.9 11-Apr-09
401 ---------------------
403 1. When building with support for bzlib/zlib (pcregrep) and/or readline
404 (pcretest), all targets were linked against these libraries. This included
405 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
406 libraries. This caused unwanted dependencies to be created. This problem
407 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
408 pcretest is linked with readline.
410 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
411 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
412 moved outside it again, because FALSE and TRUE are already defined in AIX,
413 but BOOL is not.
415 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
416 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
418 4. The pcregrep documentation said spaces were inserted as well as colons (or
419 hyphens) following file names and line numbers when outputting matching
420 lines. This is not true; no spaces are inserted. I have also clarified the
421 wording for the --colour (or --color) option.
423 5. In pcregrep, when --colour was used with -o, the list of matching strings
424 was not coloured; this is different to GNU grep, so I have changed it to be
425 the same.
427 6. When --colo(u)r was used in pcregrep, only the first matching substring in
428 each matching line was coloured. Now it goes on to look for further matches
429 of any of the test patterns, which is the same behaviour as GNU grep.
431 7. A pattern that could match an empty string could cause pcregrep to loop; it
432 doesn't make sense to accept an empty string match in pcregrep, so I have
433 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
434 seems to be how GNU grep behaves.
436 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
437 start or after a newline", because the conditional assertion was not being
438 correctly handled. The rule now is that both the assertion and what follows
439 in the first alternative must satisfy the test.
441 9. If auto-callout was enabled in a pattern with a conditional group whose
442 condition was an assertion, PCRE could crash during matching, both with
443 pcre_exec() and pcre_dfa_exec().
445 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
446 used for matching.
448 11. Unicode property support in character classes was not working for
449 characters (bytes) greater than 127 when not in UTF-8 mode.
451 12. Added the -M command line option to pcretest.
453 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
455 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
457 16. Added comments and documentation about mis-use of no_arg in the C++
458 wrapper.
460 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
461 from Martin Jerabek that uses macro names for all relevant character and
462 string constants.
464 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
465 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
466 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
467 these, but not everybody uses configure.
469 19. A conditional group that had only one branch was not being correctly
470 recognized as an item that could match an empty string. This meant that an
471 enclosing group might also not be so recognized, causing infinite looping
472 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
473 with the subject "ab", where knowledge that the repeated group can match
474 nothing is needed in order to break the loop.
476 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
477 exec(), but without supplying a callout function, matching went wrong.
479 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
480 leak if the size of the offset vector was greater than 30. When the vector
481 is smaller, the saved offsets during recursion go onto a local stack
482 vector, but for larger vectors malloc() is used. It was failing to free
483 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
484 error, in fact).
486 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
487 heapframe that is used only when UTF-8 support is enabled. This caused no
488 problem, but was untidy.
490 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
491 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
492 included within another project.
494 24. Steven Van Ingelgem's patches to add more options to the CMake support,
495 slightly modified by me:
497 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
498 not building pcregrep.
500 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
501 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
503 25. Forward references, both numeric and by name, in patterns that made use of
504 duplicate group numbers, could behave incorrectly or give incorrect errors,
505 because when scanning forward to find the reference group, PCRE was not
506 taking into account the duplicate group numbers. A pattern such as
507 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
509 26. Changed a few more instances of "const unsigned char *" to USPTR, making
510 the feature of a custom pointer more persuasive (as requested by a user).
512 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
513 pcretest.c, inside #ifndefs, because it seems they are sometimes already
514 pre-defined.
516 28. Added support for (*UTF8) at the start of a pattern.
518 29. Arrange for flags added by the "release type" setting in CMake to be shown
519 in the configuration summary.
522 Version 7.8 05-Sep-08
523 ---------------------
525 1. Replaced UCP searching code with optimized version as implemented for Ad
526 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
527 stage table and inline lookup instead of a function, giving speed ups of 2
528 to 5 times on some simple patterns that I tested. Permission was given to
529 distribute the MultiStage2.py script that generates the tables (it's not in
530 the tarball, but is in the Subversion repository).
532 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
533 scripts.
535 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
536 a group with a zero qualifier. The result of the study could be incorrect,
537 or the function might crash, depending on the pattern.
539 4. Caseless matching was not working for non-ASCII characters in back
540 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
541 It now works when Unicode Property Support is available.
543 5. In pcretest, an escape such as \x{de} in the data was always generating
544 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
545 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
546 truncation.
548 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
550 7. Added two (int) casts to pcregrep when printing the difference of two
551 pointers, in case they are 64-bit values.
553 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
554 test 2 if it fails.
556 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
557 and a #define of that name to empty if it is not externally set. This is to
558 allow users of MSVC to set it if necessary.
560 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
561 the convenience functions in the pcre_get.c source file.
563 11. An option change at the start of a pattern that had top-level alternatives
564 could cause overwriting and/or a crash. This command provoked a crash in
565 some environments:
567 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
569 This potential security problem was recorded as CVE-2008-2371.
571 12. For a pattern where the match had to start at the beginning or immediately
572 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
573 pcre_dfa_exec() could read past the end of the passed subject if there was
574 no match. To help with detecting such bugs (e.g. with valgrind), I modified
575 pcretest so that it places the subject at the end of its malloc-ed buffer.
577 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
578 exec() might read past the end of the data buffer in UTF-8 mode.
580 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
581 the data contained the byte 0x85 as part of a UTF-8 character within its
582 first line. This applied both to normal and DFA matching.
584 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
585 /^[^d]*?$/8 failed to match "abc".
587 16. Added a missing copyright notice to pcrecpp_internal.h.
589 17. Make it more clear in the documentation that values returned from
590 pcre_exec() in ovector are byte offsets, not character counts.
592 18. Tidied a few places to stop certain compilers from issuing warnings.
594 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
595 supplied by Stefan Weber. I made a further small update for 7.8 because
596 there is a change of source arrangements: the pcre_searchfuncs.c module is
597 replaced by pcre_ucd.c.
600 Version 7.7 07-May-08
601 ---------------------
603 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
604 a string to a long long, pretend we don't even have a long long." This is
605 done by checking for the strtoq, strtoll, and _strtoi64 functions.
607 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
608 pre-7.6 versions, which defined a global no_arg variable instead of putting
609 it in the RE class. (See also #8 below.)
611 3. Remove a line of dead code, identified by coverity and reported by Nuno
612 Lopes.
614 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
616 (1) The include/exclude patterns were being applied to the whole pathnames
617 of files, instead of just to the final components.
619 (2) If there was more than one level of directory, the subdirectories were
620 skipped unless they satisfied the include/exclude conditions. This is
621 inconsistent with GNU grep (and could even be seen as contrary to the
622 pcregrep specification - which I improved to make it absolutely clear).
623 The action now is always to scan all levels of directory, and just
624 apply the include/exclude patterns to regular files.
626 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
627 --exclude_dir in the tests to avoid scanning .svn directories.
629 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
630 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
631 doesn't support NULs in patterns.
633 7. Added some missing "const"s to declarations of static tables in
634 pcre_compile.c and pcre_dfa_exec.c.
636 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
637 caused by fix #2 above. (Subsequently also a second patch to fix the
638 first patch. And a third patch - this was a messy problem.)
640 9. Applied Craig's patch to remove the use of push_back().
642 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
643 matching function regexec().
645 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
646 which, however, unlike Perl's \g{...}, are subroutine calls, not back
647 references. PCRE supports relative numbers with this syntax (I don't think
648 Oniguruma does).
650 12. Previously, a group with a zero repeat such as (...){0} was completely
651 omitted from the compiled regex. However, this means that if the group
652 was called as a subroutine from elsewhere in the pattern, things went wrong
653 (an internal error was given). Such groups are now left in the compiled
654 pattern, with a new opcode that causes them to be skipped at execution
655 time.
657 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
658 to the way PCRE behaves:
660 (a) A lone ] character is dis-allowed (Perl treats it as data).
662 (b) A back reference to an unmatched subpattern matches an empty string
663 (Perl fails the current match path).
665 (c) A data ] in a character class must be notated as \] because if the
666 first data character in a class is ], it defines an empty class. (In
667 Perl it is not possible to have an empty class.) The empty class []
668 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
669 The negative empty class [^] matches any one character, independently
670 of the DOTALL setting.
672 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
673 non-existent subpattern following a character class starting with ']' and
674 containing () gave an internal compiling error instead of "reference to
675 non-existent subpattern". Fortunately, when the pattern did exist, the
676 compiled code was correct. (When scanning forwards to check for the
677 existencd of the subpattern, it was treating the data ']' as terminating
678 the class, so got the count wrong. When actually compiling, the reference
679 was subsequently set up correctly.)
681 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
682 it was being rejected as not supported by pcre_dfa_exec(), even though
683 other assertions are supported. I have made pcre_dfa_exec() support
684 (*FAIL).
686 16. The implementation of 13c above involved the invention of a new opcode,
687 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
688 cannot be changed at match time, I realized I could make a small
689 improvement to matching performance by compiling OP_ALLANY instead of
690 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
691 on the OP_ANY path.
693 17. Compiling pcretest on Windows with readline support failed without the
694 following two fixes: (1) Make the unistd.h include conditional on
695 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
697 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
698 ncurses library to be included for pcretest when ReadLine support is
699 requested, but also to allow for it to be overridden. This patch came from
700 Daniel Bergström.
702 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
703 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
704 any errors with the current Unicode tables. Thanks to Peter Kankowski for
705 spotting this.
708 Version 7.6 28-Jan-08
709 ---------------------
711 1. A character class containing a very large number of characters with
712 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
713 overflow.
715 2. Patch to cut out the "long long" test in pcrecpp_unittest when
716 HAVE_LONG_LONG is not defined.
718 3. Applied Christian Ehrlicher's patch to update the CMake build files to
719 bring them up to date and include new features. This patch includes:
721 - Fixed PH's badly added libz and libbz2 support.
722 - Fixed a problem with static linking.
723 - Added pcredemo. [But later removed - see 7 below.]
724 - Fixed dftables problem and added an option.
725 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
727 - Added readline support for pcretest.
728 - Added an listing of the option settings after cmake has run.
730 4. A user submitted a patch to Makefile that makes it easy to create
731 "pcre.dll" under mingw when using Configure/Make. I added stuff to
732 Makefile.am that cause it to include this special target, without
733 affecting anything else. Note that the same mingw target plus all
734 the other distribution libraries and programs are now supported
735 when configuring with CMake (see 6 below) instead of with
736 Configure/Make.
738 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
739 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
740 exported in the Windows port". It has not yet been confirmed that the patch
741 solves the problem, but it does no harm.
743 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
744 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
745 with CMake, and also correct the comment about stack recursion.
747 7. Remove the automatic building of pcredemo from the ./configure system and
748 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
749 of a program that users should build themselves after PCRE is installed, so
750 building it automatically is not really right. What is more, it gave
751 trouble in some build environments.
753 8. Further tidies to CMakeLists.txt from Sheri and Christian.
756 Version 7.5 10-Jan-08
757 ---------------------
759 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
760 values in parens when parsing an RE using the C++ wrapper."
762 2. Negative specials like \S did not work in character classes in UTF-8 mode.
763 Characters greater than 255 were excluded from the class instead of being
764 included.
766 3. The same bug as (2) above applied to negated POSIX classes such as
767 [:^space:].
769 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
770 defined or documented. It seems to have been a typo for PCRE_STATIC, so
771 I have changed it.
773 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
774 first named subpattern) and a construct such as (?&a) would reference the
775 first named subpattern whose name started with "a" (in other words, the
776 length check was missing). Both these problems are fixed. "Subpattern name
777 expected" is now given for (?&) (a zero-length name), and this patch also
778 makes it give the same error for \k'' (previously it complained that that
779 was a reference to a non-existent subpattern).
781 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
782 this is right because (?- can be followed by option settings as well as by
783 digits. I have, however, made the messages clearer.
785 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
786 than the number used in the conditional) now cause a compile-time error.
787 This is actually not compatible with Perl, which accepts such patterns, but
788 treats the conditional as always being FALSE (as PCRE used to), but it
789 seems to me that giving a diagnostic is better.
791 8. Change "alphameric" to the more common word "alphanumeric" in comments
792 and messages.
794 9. Fix two occurrences of "backslash" in comments that should have been
795 "backspace".
797 10. Remove two redundant lines of code that can never be obeyed (their function
798 was moved elsewhere).
800 11. The program that makes PCRE's Unicode character property table had a bug
801 which caused it to generate incorrect table entries for sequences of
802 characters that have the same character type, but are in different scripts.
803 It amalgamated them into a single range, with the script of the first of
804 them. In other words, some characters were in the wrong script. There were
805 thirteen such cases, affecting characters in the following ranges:
807 U+002b0 - U+002c1
808 U+0060c - U+0060d
809 U+0061e - U+00612
810 U+0064b - U+0065e
811 U+0074d - U+0076d
812 U+01800 - U+01805
813 U+01d00 - U+01d77
814 U+01d9b - U+01dbf
815 U+0200b - U+0200f
816 U+030fc - U+030fe
817 U+03260 - U+0327f
818 U+0fb46 - U+0fbb1
819 U+10450 - U+1049d
821 12. The -o option (show only the matching part of a line) for pcregrep was not
822 compatible with GNU grep in that, if there was more than one match in a
823 line, it showed only the first of them. It now behaves in the same way as
824 GNU grep.
826 13. If the -o and -v options were combined for pcregrep, it printed a blank
827 line for every non-matching line. GNU grep prints nothing, and pcregrep now
828 does the same. The return code can be used to tell if there were any
829 non-matching lines.
831 14. Added --file-offsets and --line-offsets to pcregrep.
833 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
834 infinitely looping recursion. The bug was that positive lookaheads were not
835 being skipped when checking for a possible empty match (negative lookaheads
836 and both kinds of lookbehind were skipped).
838 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
839 inclusion of <windows.h> to before rather than after the definition of
840 INVALID_FILE_ATTRIBUTES (patch from David Byron).
842 17. Specifying a possessive quantifier with a specific limit for a Unicode
843 character property caused pcre_compile() to compile bad code, which led at
844 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
845 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
846 caused the error; without that there was no problem.
848 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
850 19. Added --enable-pcretest-libreadline.
852 20. In pcrecpp.cc, the variable 'count' was incremented twice in
853 RE::GlobalReplace(). As a result, the number of replacements returned was
854 double what it should be. I removed one of the increments, but Craig sent a
855 later patch that removed the other one (the right fix) and added unit tests
856 that check the return values (which was not done before).
858 21. Several CMake things:
860 (1) Arranged that, when cmake is used on Unix, the libraries end up with
861 the names libpcre and libpcreposix, not just pcre and pcreposix.
863 (2) The above change means that pcretest and pcregrep are now correctly
864 linked with the newly-built libraries, not previously installed ones.
868 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
869 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
870 UTF-8 newline character). The key issue is that the pattern starts .*;
871 this means that the match must be either at the beginning, or after a
872 newline. The bug was in the code for advancing after a failed match and
873 checking that the new position followed a newline. It was not taking
874 account of UTF-8 characters correctly.
876 23. PCRE was behaving differently from Perl in the way it recognized POSIX
877 character classes. PCRE was not treating the sequence [:...:] as a
878 character class unless the ... were all letters. Perl, however, seems to
879 allow any characters between [: and :], though of course it rejects as
880 unknown any "names" that contain non-letters, because all the known class
881 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
882 for example, whereas PCRE did not - it did not recognize a POSIX character
883 class. This seemed a bit dangerous, so the code has been changed to be
884 closer to Perl. The behaviour is not identical to Perl, because PCRE will
885 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
886 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
887 Perl does, and where it didn't before.
889 24. Rewrite so as to remove the single use of %n from pcregrep because in some
890 Windows environments %n is disabled by default.
893 Version 7.4 21-Sep-07
894 ---------------------
896 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
897 means that a class such as [\s] counted as "explicit reference to CR or
898 LF". That isn't really right - the whole point of the change was to try to
899 help when there was an actual mention of one of the two characters. So now
900 the change happens only if \r or \n (or a literal CR or LF) character is
901 encountered.
903 2. The 32-bit options word was also used for 6 internal flags, but the numbers
904 of both had grown to the point where there were only 3 bits left.
905 Fortunately, there was spare space in the data structure, and so I have
906 moved the internal flags into a new 16-bit field to free up more option
907 bits.
909 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
910 but did not set the internal JCHANGED flag - either of these is enough to
911 control the way the "get" function works - but the PCRE_INFO_JCHANGED
912 facility is supposed to tell if (?J) was ever used, so now (?J) at the
913 start sets both bits.
915 4. Added options (at build time, compile time, exec time) to change \R from
916 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
918 5. doc/pcresyntax.html was missing from the distribution.
920 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
921 compatibility, even though it is no longer used.
923 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
924 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
925 windows.h file is present (where different names are used). [This was
926 reversed later after testing - see 16 below.]
928 8. Changed all #include <config.h> to #include "config.h". There were also
929 some further <pcre.h> cases that I changed to "pcre.h".
931 9. When pcregrep was used with the --colour option, it missed the line ending
932 sequence off the lines that it output.
934 10. It was pointed out to me that arrays of string pointers cause lots of
935 relocations when a shared library is dynamically loaded. A technique of
936 using a single long string with a table of offsets can drastically reduce
937 these. I have refactored PCRE in four places to do this. The result is
938 dramatic:
940 Originally: 290
941 After changing UCP table: 187
942 After changing error message table: 43
943 After changing table of "verbs" 36
944 After changing table of Posix names 22
946 Thanks to the folks working on Gregex for glib for this insight.
948 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
949 unicode-properties was also set.
951 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
953 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
954 checked only for CRLF.
956 14. Added casts to pcretest.c to avoid compiler warnings.
958 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
960 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
961 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
962 entirely. This removes changes made in 7 above.
964 17. The CMake files have been updated, and there is now more information about
965 building with CMake in the NON-UNIX-USE document.
968 Version 7.3 28-Aug-07
969 ---------------------
971 1. In the rejigging of the build system that eventually resulted in 7.1, the
972 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
973 brackets there is not right, since it causes compilers to look for an
974 installed pcre.h, not the version that is in the source that is being
975 compiled (which of course may be different). I have changed it back to:
977 #include "pcre.h"
979 I have a vague recollection that the change was concerned with compiling in
980 different directories, but in the new build system, that is taken care of
981 by the VPATH setting the Makefile.
983 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
984 when the subject happened to end in the byte 0x85 (e.g. if the last
985 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
986 characters but of course it shouldn't be taken as a newline when it is part
987 of another character. The bug was that, for an unlimited repeat of . in
988 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
989 characters when looking for a newline.
991 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
993 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
994 in debug output.
996 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
997 long printing in the pcrecpp unittest when running under MinGW.
999 6. ESC_K was left out of the EBCDIC table.
1001 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1002 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1003 limit also applies to "virtual nesting" when a pattern is recursive, and in
1004 this case 1000 isn't so big. I have been able to remove this limit at the
1005 expense of backing off one optimization in certain circumstances. Normally,
1006 when pcre_exec() would call its internal match() function recursively and
1007 immediately return the result unconditionally, it uses a "tail recursion"
1008 feature to save stack. However, when a subpattern that can match an empty
1009 string has an unlimited repetition quantifier, it no longer makes this
1010 optimization. That gives it a stack frame in which to save the data for
1011 checking that an empty string has been matched. Previously this was taken
1012 from the 1000-entry workspace that had been reserved. So now there is no
1013 explicit limit, but more stack is used.
1015 8. Applied Daniel's patches to solve problems with the import/export magic
1016 syntax that is required for Windows, and which was going wrong for the
1017 pcreposix and pcrecpp parts of the library. These were overlooked when this
1018 problem was solved for the main library.
1020 9. There were some crude static tests to avoid integer overflow when computing
1021 the size of patterns that contain repeated groups with explicit upper
1022 limits. As the maximum quantifier is 65535, the maximum group length was
1023 set at 30,000 so that the product of these two numbers did not overflow a
1024 32-bit integer. However, it turns out that people want to use groups that
1025 are longer than 30,000 bytes (though not repeat them that many times).
1026 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1027 made it possible to implement the integer overflow checks in a much more
1028 dynamic way, which I have now done. The artificial limitation on group
1029 length has been removed - we now have only the limit on the total length of
1030 the compiled pattern, which depends on the LINK_SIZE setting.
1032 10. Fixed a bug in the documentation for get/copy named substring when
1033 duplicate names are permitted. If none of the named substrings are set, the
1034 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1035 empty string.
1037 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1038 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1039 because the ] is interpreted as the first data character and the
1040 terminating ] is not found. PCRE has been made compatible with Perl in this
1041 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1042 cause memory overwriting.
1044 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1045 string has been matched (to stop an infinite loop). It was not recognizing
1046 a conditional subpattern that could match an empty string if that
1047 subpattern was within another subpattern. For example, it looped when
1048 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1049 condition was not nested. This bug has been fixed.
1051 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1052 past the start of the subject in the presence of bytes with the top bit
1053 set, for example "\x8aBCD".
1055 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1056 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1058 14. Optimized (?!) to (*FAIL).
1060 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1061 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1062 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1063 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1064 does: it's just the validity check that is more restrictive.
1066 16. Inserted checks for integer overflows during escape sequence (backslash)
1067 processing, and also fixed erroneous offset values for syntax errors during
1068 backslash processing.
1070 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1071 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1073 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1074 caused an overrun.
1076 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1077 something other than just ASCII characters) inside a group that had an
1078 unlimited repeat caused a loop at compile time (while checking to see
1079 whether the group could match an empty string).
1081 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1082 [\P{Any}] did so. (Error in the code for printing property names.)
1084 21. An orphan \E inside a character class could cause a crash.
1086 22. A repeated capturing bracket such as (A)? could cause a wild memory
1087 reference during compilation.
1089 23. There are several functions in pcre_compile() that scan along a compiled
1090 expression for various reasons (e.g. to see if it's fixed length for look
1091 behind). There were bugs in these functions when a repeated \p or \P was
1092 present in the pattern. These operators have additional parameters compared
1093 with \d, etc, and these were not being taken into account when moving along
1094 the compiled data. Specifically:
1096 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1097 length.
1099 (b) An item such as \pL+ within a repeated group could cause crashes or
1100 loops.
1102 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1103 "reference to non-existent subpattern" error.
1105 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1107 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1108 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1110 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1112 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1113 character were causing crashes (broken optimization).
1115 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1116 \p or \P) caused a compile-time loop.
1118 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1119 break. For example, the unstudied pattern [\r\n]A does not match the string
1120 "\r\nA" because change 7.0/46 below moves the current point on by two
1121 characters after failing to match at the start. However, the pattern \nA
1122 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1123 the same is true. There doesn't seem any very clean way out of this, but
1124 what I have chosen to do makes the common cases work: PCRE now takes note
1125 of whether there can be an explicit match for \r or \n anywhere in the
1126 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1127 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1128 pattern has explicit CR or LF references.
1130 29. Added (*CR) etc for changing newline setting at start of pattern.
1133 Version 7.2 19-Jun-07
1134 ---------------------
1136 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1137 which is apparently normally available under Windows.
1139 2. Re-jig the pcregrep tests with different newline settings in an attempt
1140 to make them independent of the local environment's newline setting.
1142 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1144 4. Some of the "internals" tests were previously cut out when the link size
1145 was not 2, because the output contained actual offsets. The recent new
1146 "Z" feature of pcretest means that these can be cut out, making the tests
1147 usable with all link sizes.
1149 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1150 stack recursion. This gives a massive performance boost under BSD, but just
1151 a small improvement under Linux. However, it saves one field in the frame
1152 in all cases.
1154 6. Added more features from the forthcoming Perl 5.10:
1156 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1157 recursion call. It refers to the nth most recently opened parentheses.
1159 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1160 to be opened parentheses.
1162 (c) Conditions that refer to capturing parentheses can be specified
1163 relatively, for example, (?(-2)... or (?(+3)...
1165 (d) \K resets the start of the current match so that everything before
1166 is not part of it.
1168 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1170 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1171 reference syntax.
1173 (g) (?| introduces a group in which the numbering of parentheses in each
1174 alternative starts with the same number.
1176 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1178 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1181 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1182 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1183 for detecting groups that can match an empty string.
1185 9. A pattern with a very large number of alternatives (more than several
1186 hundred) was running out of internal workspace during the pre-compile
1187 phase, where pcre_compile() figures out how much memory will be needed. A
1188 bit of new cunning has reduced the workspace needed for groups with
1189 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1190 workspace instead of running out of the 4096 that are available.
1192 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1194 11. Applied patch from Google to remove an optimization that didn't quite work.
1195 The report of the bug said:
1197 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1198 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1199 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1201 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1202 it matched the wrong number of bytes.
1205 Version 7.1 24-Apr-07
1206 ---------------------
1208 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1209 that is more "standard", making use of automake and other Autotools. There
1210 is some re-arrangement of the files and adjustment of comments consequent
1211 on this.
1213 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1214 for recursive directory scanning broke on some systems because the files
1215 are not scanned in any specific order and on different systems the order
1216 was different. A call to "sort" has been inserted into RunGrepTest for the
1217 approprate test as a short-term fix. In the longer term there may be an
1218 alternative.
1220 3. I had an email from Eric Raymond about problems translating some of PCRE's
1221 man pages to HTML (despite the fact that I distribute HTML pages, some
1222 people do their own conversions for various reasons). The problems
1223 concerned the use of low-level troff macros .br and .in. I have therefore
1224 removed all such uses from the man pages (some were redundant, some could
1225 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1226 HTML has been updated to handle .nf/.fi and to complain if it encounters
1227 .br or .in.
1229 4. Updated comments in configure.ac that get placed in config.h.in and also
1230 arranged for config.h to be included in the distribution, with the name
1231 config.h.generic, for the benefit of those who have to compile without
1232 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1234 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1235 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1236 makevp.bat for the current PCRE, using the additional files
1237 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1239 6. A Windows user reported a minor discrepancy with test 2, which turned out
1240 to be caused by a trailing space on an input line that had got lost in his
1241 copy. The trailing space was an accident, so I've just removed it.
1243 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1244 that is needed.
1246 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1247 as "const" (a) because they are and (b) because it helps the PHP
1248 maintainers who have recently made a script to detect big data structures
1249 in the php code that should be moved to the .rodata section. I remembered
1250 to update Builducptable as well, so it won't revert if ucptable.h is ever
1251 re-created.
1253 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1254 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1255 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1256 support is not required. This saves 1.5-2K of code, which is important in
1257 some applications.
1259 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1260 so as not to refer to the tables, even though these functions will never be
1261 called when UTF-8 support is disabled. Otherwise there are problems with a
1262 shared library.
1264 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1266 (a) It was defining its arguments as char * instead of void *.
1268 (b) It was assuming that all moves were upwards in memory; this was true
1269 a long time ago when I wrote it, but is no longer the case.
1271 The emulated memove() is provided for those environments that have neither
1272 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1273 is clearly not the case, as these two bugs were recently reported.
1275 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1276 and Detrail to create the HTML documentation, the .txt form of the man
1277 pages, and it removes trailing spaces from listed files. It also creates
1278 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1279 case, it wraps all the #defines with #ifndefs. This script should be run
1280 before "make dist".
1282 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1283 with Unicode property support.
1285 (a) For a maximizing quantifier, if the two different cases of the
1286 character were of different lengths in their UTF-8 codings (there are
1287 some cases like this - I found 11), and the matching function had to
1288 back up over a mixture of the two cases, it incorrectly assumed they
1289 were both the same length.
1291 (b) When PCRE was configured to use the heap rather than the stack for
1292 recursion during matching, it was not correctly preserving the data for
1293 the other case of a UTF-8 character when checking ahead for a match
1294 while processing a minimizing repeat. If the check also involved
1295 matching a wide character, but failed, corruption could cause an
1296 erroneous result when trying to check for a repeat of the original
1297 character.
1299 13. Some tidying changes to the testing mechanism:
1301 (a) The RunTest script now detects the internal link size and whether there
1302 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1303 values substituted by "configure". (The RunGrepTest script already did
1304 this for UTF-8.) The configure.ac script no longer substitutes the
1305 relevant variables.
1307 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1308 with length and offset values. This means that the output is different
1309 for different internal link sizes. Test 2 is skipped for link sizes
1310 other than 2 because of this, bypassing the problem. Unfortunately,
1311 there was also a test in test 3 (the locale tests) that used /B and
1312 failed for link sizes other than 2. Rather than cut the whole test out,
1313 I have added a new /Z option to pcretest that replaces the length and
1314 offset values with spaces. This is now used to make test 3 independent
1315 of link size. (Test 2 will be tidied up later.)
1317 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1318 segmentation fault instead of returning the appropriate error message.
1320 15. In multiline mode when the newline sequence was set to "any", the pattern
1321 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1322 This doesn't seem right; it now treats the CRLF combination as the line
1323 ending, and so does not match in that case. It's only a pattern such as ^$
1324 that would hit this one: something like ^ABC$ would have failed after \r
1325 and then tried again after \r\n.
1327 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1328 in an attempt to make files that differ only in their line terminators
1329 compare equal. This works on Linux.
1331 17. Under certain error circumstances pcregrep might try to free random memory
1332 as it exited. This is now fixed, thanks to valgrind.
1334 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1335 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1336 was because its rules for how to advance for /g after matching an empty
1337 string at the end of a line did not allow for this case. They now check for
1338 it specially.
1340 20. pcretest is supposed to handle patterns and data of any length, by
1341 extending its buffers when necessary. It was getting this wrong when the
1342 buffer for a data line had to be extended.
1344 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1345 CRLF as a newline sequence.
1347 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1348 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1349 I have nevertheless tidied it up.
1351 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1353 24. Added a man page for pcre-config.
1356 Version 7.0 19-Dec-06
1357 ---------------------
1359 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1360 moving to gcc 4.1.1.
1362 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1363 sys/time.h, which is documented as needed for this function. It doesn't
1364 seem to matter on Linux, but it showed up on some releases of OS X.
1366 3. It seems that there are systems where bytes whose values are greater than
1367 127 match isprint() in the "C" locale. The "C" locale should be the
1368 default when a C program starts up. In most systems, only ASCII printing
1369 characters match isprint(). This difference caused the output from pcretest
1370 to vary, making some of the tests fail. I have changed pcretest so that:
1372 (a) When it is outputting text in the compiled version of a pattern, bytes
1373 other than 32-126 are always shown as hex escapes.
1375 (b) When it is outputting text that is a matched part of a subject string,
1376 it does the same, unless a different locale has been set for the match
1377 (using the /L modifier). In this case, it uses isprint() to decide.
1379 4. Fixed a major bug that caused incorrect computation of the amount of memory
1380 required for a compiled pattern when options that changed within the
1381 pattern affected the logic of the preliminary scan that determines the
1382 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1383 that the computed length was too small. The symptoms of this bug were
1384 either the PCRE error "internal error: code overflow" from pcre_compile(),
1385 or a glibc crash with a message such as "pcretest: free(): invalid next
1386 size (fast)". Examples of patterns that provoked this bug (shown in
1387 pcretest format) are:
1389 /(?-x: )/x
1390 /(?x)(?-x: \s*#\s*)/
1391 /((?i)[\x{c0}])/8
1392 /(?i:[\x{c0}])/8
1394 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1395 is now done differently.
1397 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1398 wrapper classes; (b) implement a new function in the C++ scanner that is
1399 more efficient than the old way of doing things because it avoids levels of
1400 recursion in the regex matching; (c) add a paragraph to the documentation
1401 for the FullMatch() function.
1403 6. The escape sequence \n was being treated as whatever was defined as
1404 "newline". Not only was this contrary to the documentation, which states
1405 that \n is character 10 (hex 0A), but it also went horribly wrong when
1406 "newline" was defined as CRLF. This has been fixed.
1408 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1409 was being set to -1 for the "end of line" case (supposedly a value that no
1410 character can have). Though this value is never used (the check for end of
1411 line is "zero bytes in current character"), it caused compiler complaints.
1412 I've changed it to 0xffffffff.
1414 8. In pcre_version.c, the version string was being built by a sequence of
1415 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1416 string (as it is for production releases) called a macro with an empty
1417 argument. The C standard says the result of this is undefined. The gcc
1418 compiler treats it as an empty string (which was what was wanted) but it is
1419 reported that Visual C gives an error. The source has been hacked around to
1420 avoid this problem.
1422 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1423 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1424 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1425 of them did).
1427 10. Originally, pcretest opened its input and output without "b"; then I was
1428 told that "b" was needed in some environments, so it was added for release
1429 5.0 to both the input and output. (It makes no difference on Unix-like
1430 systems.) Later I was told that it is wrong for the input on Windows. I've
1431 now abstracted the modes into two macros, to make it easier to fiddle with
1432 them, and removed "b" from the input mode under Windows.
1434 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1436 12. Added -help and --help to pcretest as an official way of being reminded
1437 of the options.
1439 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1440 and pcrecpp.cc because they annoy compilers at high warning levels.
1442 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1444 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1445 scripts are not C programs :-) and which was not noticed because it works
1446 on Linux.
1448 16. pcretest is supposed to handle any length of pattern and data line (as one
1449 line or as a continued sequence of lines) by extending its input buffer if
1450 necessary. This feature was broken for very long pattern lines, leading to
1451 a string of junk being passed to pcre_compile() if the pattern was longer
1452 than about 50K.
1454 17. I have done a major re-factoring of the way pcre_compile() computes the
1455 amount of memory needed for a compiled pattern. Previously, there was code
1456 that made a preliminary scan of the pattern in order to do this. That was
1457 OK when PCRE was new, but as the facilities have expanded, it has become
1458 harder and harder to keep it in step with the real compile phase, and there
1459 have been a number of bugs (see for example, 4 above). I have now found a
1460 cunning way of running the real compile function in a "fake" mode that
1461 enables it to compute how much memory it would need, while actually only
1462 ever using a few hundred bytes of working memory and without too many
1463 tests of the mode. This should make future maintenance and development
1464 easier. A side effect of this work is that the limit of 200 on the nesting
1465 depth of parentheses has been removed (though this was never a serious
1466 limitation, I suspect). However, there is a downside: pcre_compile() now
1467 runs more slowly than before (30% or more, depending on the pattern). I
1468 hope this isn't a big issue. There is no effect on runtime performance.
1470 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1471 newline (only possible for the last line of a file) and it was a
1472 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1474 19. Added additional timing features to pcretest. (1) The -tm option now times
1475 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1476 separate command line item, by a number that specifies the number of
1477 repeats to use when timing. The default is 50000; this gives better
1478 precision, but takes uncomfortably long for very large patterns.
1480 20. Extended pcre_study() to be more clever in cases where a branch of a
1481 subpattern has no definite first character. For example, (a*|b*)[cd] would
1482 previously give no result from pcre_study(). Now it recognizes that the
1483 first character must be a, b, c, or d.
1485 21. There was an incorrect error "recursive call could loop indefinitely" if
1486 a subpattern (or the entire pattern) that was being tested for matching an
1487 empty string contained only one non-empty item after a nested subpattern.
1488 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1489 incorrectly, because the \d was being skipped in the check.
1491 22. The pcretest program now has a new pattern option /B and a command line
1492 option -b, which is equivalent to adding /B to every pattern. This causes
1493 it to show the compiled bytecode, without the additional information that
1494 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1495 is the same as /B/I).
1497 23. A new optimization is now able automatically to treat some sequences such
1498 as a*b as a*+b. More specifically, if something simple (such as a character
1499 or a simple class like \d) has an unlimited quantifier, and is followed by
1500 something that cannot possibly match the quantified thing, the quantifier
1501 is automatically "possessified".
1503 24. A recursive reference to a subpattern whose number was greater than 39
1504 went wrong under certain circumstances in UTF-8 mode. This bug could also
1505 have affected the operation of pcre_study().
1507 25. Realized that a little bit of performance could be had by replacing
1508 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1510 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1512 27. Possessive quantifiers such as a++ were previously implemented by turning
1513 them into atomic groups such as ($>a+). Now they have their own opcodes,
1514 which improves performance. This includes the automatically created ones
1515 from 23 above.
1517 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1518 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1519 the first matched character to be a colon. This applied both to named and
1520 numbered groups.
1522 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1524 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1525 building PCRE on Windows easier, so I have included it in the distribution.
1527 31. There is now a check in pcretest against a ridiculously large number being
1528 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1529 loop, the loop is abandoned.
1531 32. Forward references to subpatterns in conditions such as (?(2)...) where
1532 subpattern 2 is defined later cause pcre_compile() to search forwards in
1533 the pattern for the relevant set of parentheses. This search went wrong
1534 when there were unescaped parentheses in a character class, parentheses
1535 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1537 33. "Subroutine" calls and backreferences were previously restricted to
1538 referencing subpatterns earlier in the regex. This restriction has now
1539 been removed.
1541 34. Added a number of extra features that are going to be in Perl 5.10. On the
1542 whole, these are just syntactic alternatives for features that PCRE had
1543 previously implemented using the Python syntax or my own invention. The
1544 other formats are all retained for compatibility.
1546 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1547 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1548 also .NET compatible.
1550 (b) A recursion or subroutine call to a named group can now be defined as
1551 (?&name) as well as (?P>name).
1553 (c) A backreference to a named group can now be defined as \k<name> or
1554 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1555 5.10, are also .NET compatible.
1557 (d) A conditional reference to a named group can now use the syntax
1558 (?(<name>) or (?('name') as well as (?(name).
1560 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1561 groups (named and numbered) that are never evaluated inline, but can be
1562 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1563 is always false. There may be only one alternative in such a group.
1565 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1566 as the simple (?(R). The condition is true only if the most recent
1567 recursion is that of the given number or name. It does not search out
1568 through the entire recursion stack.
1570 (g) The escape \gN or \g{N} has been added, where N is a positive or
1571 negative number, specifying an absolute or relative reference.
1573 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1574 some "unreachable code" warnings.
1576 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1577 things, this adds five new scripts.
1579 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1580 There were also incompatibilities regarding the handling of \Q..\E inside
1581 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1582 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1584 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1585 matches an empty string, and forcibly breaks the loop. There were bugs in
1586 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1587 against aaaa the result was just "a" rather than "aaaa", for example. Two
1588 separate and independent bugs (that affected different cases) have been
1589 fixed.
1591 39. Refactored the code to abolish the use of different opcodes for small
1592 capturing bracket numbers. This is a tidy that I avoided doing when I
1593 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1594 The new approach is not only tidier, it makes it possible to reduce the
1595 memory needed to fix the previous bug (38).
1597 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1598 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1599 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1600 mode.
1602 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1603 report.
1605 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1606 copy construction and assignment in the C++ wrapper.
1608 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1609 couple of bugs that could have given wrong results in the "--newline=crlf"
1610 case.
1612 44. Added a number of casts and did some reorganization of signed/unsigned int
1613 variables following suggestions from Dair Grant. Also renamed the variable
1614 "this" as "item" because it is a C++ keyword.
1616 45. Arranged for dftables to add
1618 #include "pcre_internal.h"
1620 to pcre_chartables.c because without it, gcc 4.x may remove the array
1621 definition from the final binary if PCRE is built into a static library and
1622 dead code stripping is activated.
1624 46. For an unanchored pattern, if a match attempt fails at the start of a
1625 newline sequence, and the newline setting is CRLF or ANY, and the next two
1626 characters are CRLF, advance by two characters instead of one.
1629 Version 6.7 04-Jul-06
1630 ---------------------
1632 1. In order to handle tests when input lines are enormously long, pcretest has
1633 been re-factored so that it automatically extends its buffers when
1634 necessary. The code is crude, but this _is_ just a test program. The
1635 default size has been increased from 32K to 50K.
1637 2. The code in pcre_study() was using the value of the re argument before
1638 testing it for NULL. (Of course, in any sensible call of the function, it
1639 won't be NULL.)
1641 3. The memmove() emulation function in pcre_internal.h, which is used on
1642 systems that lack both memmove() and bcopy() - that is, hardly ever -
1643 was missing a "static" storage class specifier.
1645 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1646 containing an extended class (one that cannot be represented by a bitmap
1647 because it contains high-valued characters or Unicode property items, e.g.
1648 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1649 pattern, but PCRE should not loop if you do not (it no longer does).
1650 [Detail: two cases were found: (a) a repeated subpattern containing an
1651 extended class; (b) a recursive reference to a subpattern that followed a
1652 previous extended class. It wasn't skipping over the extended class
1653 correctly when UTF-8 mode was not set.]
1655 5. A negated single-character class was not being recognized as fixed-length
1656 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1657 compile error "lookbehind assertion is not fixed length".
1659 6. The RunPerlTest auxiliary script was showing an unexpected difference
1660 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1661 write a Perl script that can interpret lines of an input file either as
1662 byte characters or as UTF-8, which is what "perltest" was being required to
1663 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1664 can't do is switch easily at run time between having the "use utf8;" pragma
1665 or not. In the end, I fudged it by using the RunPerlTest script to insert
1666 "use utf8;" explicitly for the UTF-8 tests.
1668 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1669 the end of the subject string, contrary to the documentation and to what
1670 Perl does. This was true of both matching functions. Now it matches only at
1671 the start of the subject and immediately after *internal* newlines.
1673 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1674 a pointer to an int instead of a pointer to an unsigned long int. This
1675 caused problems on 64-bit systems.
1677 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1678 instance of the 'standard' template library not being so standard".
1680 10. There was no check on the number of named subpatterns nor the maximum
1681 length of a subpattern name. The product of these values is used to compute
1682 the size of the memory block for a compiled pattern. By supplying a very
1683 long subpattern name and a large number of named subpatterns, the size
1684 computation could be caused to overflow. This is now prevented by limiting
1685 the length of names to 32 characters, and the number of named subpatterns
1686 to 10,000.
1688 11. Subpatterns that are repeated with specific counts have to be replicated in
1689 the compiled pattern. The size of memory for this was computed from the
1690 length of the subpattern and the repeat count. The latter is limited to
1691 65535, but there was no limit on the former, meaning that integer overflow
1692 could in principle occur. The compiled length of a repeated subpattern is
1693 now limited to 30,000 bytes in order to prevent this.
1695 12. Added the optional facility to have named substrings with the same name.
1697 13. Added the ability to use a named substring as a condition, using the
1698 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1699 are numbers (not recommended). Forward references are permitted.
1701 14. Added forward references in named backreferences (if you see what I mean).
1703 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1704 pattern could run off the end of the subject. For example, the pattern
1705 "(?s)(.{1,5})"8 did this with the subject "ab".
1707 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1708 PCRE_CASELESS was set when matching characters that were quantified with ?
1709 or *.
1711 17. A character class other than a single negated character that had a minimum
1712 but no maximum quantifier - for example [ab]{6,} - was not handled
1713 correctly by pce_dfa_exec(). It would match only one character.
1715 18. A valid (though odd) pattern that looked like a POSIX character
1716 class but used an invalid character after [ (for example [[,abc,]]) caused
1717 pcre_compile() to give the error "Failed: internal error: code overflow" or
1718 in some cases to crash with a glibc free() error. This could even happen if
1719 the pattern terminated after [[ but there just happened to be a sequence of
1720 letters, a binary zero, and a closing ] in the memory that followed.
1722 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1723 over the years. Originally (before any Unicode support), just the bottom 8
1724 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1725 output from "man perlunicode" includes this:
1727 The regular expression compiler produces polymorphic opcodes. That
1728 is, the pattern adapts to the data and automatically switches to
1729 the Unicode character scheme when presented with Unicode data--or
1730 instead uses a traditional byte scheme when presented with byte
1731 data.
1733 Sadly, a wide octal escape does not cause a switch, and in a string with
1734 no other multibyte characters, these octal escapes are treated as before.
1735 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1736 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1737 Unicode string.
1739 I have not perpetrated such confusion in PCRE. Up till now, it took just
1740 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1741 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1742 translate to the appropriate multibyte character.
1744 29. Applied some refactoring to reduce the number of warnings from Microsoft
1745 and Borland compilers. This has included removing the fudge introduced
1746 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1747 a warning about an unused variable.
1749 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1750 characters since release 4.0, because Perl (from release 5.004) does not.
1751 [Or at least, is documented not to: some releases seem to be in conflict
1752 with the documentation.] However, when a pattern was studied with
1753 pcre_study() and all its branches started with \s, PCRE still included VT
1754 as a possible starting character. Of course, this did no harm; it just
1755 caused an unnecessary match attempt.
1757 22. Removed a now-redundant internal flag bit that recorded the fact that case
1758 dependency changed within the pattern. This was once needed for "required
1759 byte" processing, but is no longer used. This recovers a now-scarce options
1760 bit. Also moved the least significant internal flag bit to the most-
1761 significant bit of the word, which was not previously used (hangover from
1762 the days when it was an int rather than a uint) to free up another bit for
1763 the future.
1765 23. Added support for CRLF line endings as well as CR and LF. As well as the
1766 default being selectable at build time, it can now be changed at runtime
1767 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1768 specify that it is scanning data with non-default line endings.
1770 24. Changed the definition of CXXLINK to make it agree with the definition of
1771 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1773 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1774 recursions. This makes a big different to stack usage for some patterns.
1776 26. If a subpattern containing a named recursion or subroutine reference such
1777 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1778 the space required for the compiled pattern went wrong and gave too small a
1779 value. Depending on the environment, this could lead to "Failed: internal
1780 error: code overflow at offset 49" or "glibc detected double free or
1781 corruption" errors.
1783 27. Applied patches from Google (a) to support the new newline modes and (b) to
1784 advance over multibyte UTF-8 characters in GlobalReplace.
1786 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1787 difference for some implementation of PCRE in some Windows version.
1789 29. Added some extra testing facilities to pcretest:
1791 \q<number> in a data line sets the "match limit" value
1792 \Q<number> in a data line sets the "match recursion limt" value
1793 -S <number> sets the stack size, where <number> is in megabytes
1795 The -S option isn't available for Windows.
1798 Version 6.6 06-Feb-06
1799 ---------------------
1801 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1802 in pcreposix.h. I have copied the definition from pcre.h.
1804 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1805 because pcre.h is no longer a built file.
1807 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1808 not normally included in the compiled code.
1811 Version 6.5 01-Feb-06
1812 ---------------------
1814 1. When using the partial match feature with pcre_dfa_exec(), it was not
1815 anchoring the second and subsequent partial matches at the new starting
1816 point. This could lead to incorrect results. For example, with the pattern
1817 /1234/, partially matching against "123" and then "a4" gave a match.
1819 2. Changes to pcregrep:
1821 (a) All non-match returns from pcre_exec() were being treated as failures
1822 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1823 error message is output. Some extra information is given for the
1825 probably the only errors that are likely to be caused by users (by
1826 specifying a regex that has nested indefinite repeats, for instance).
1827 If there are more than 20 of these errors, pcregrep is abandoned.
1829 (b) A binary zero was treated as data while matching, but terminated the
1830 output line if it was written out. This has been fixed: binary zeroes
1831 are now no different to any other data bytes.
1833 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1834 used to set a locale for matching. The --locale=xxxx long option has
1835 been added (no short equivalent) to specify a locale explicitly on the
1836 pcregrep command, overriding the environment variables.
1838 (d) When -B was used with -n, some line numbers in the output were one less
1839 than they should have been.
1841 (e) Added the -o (--only-matching) option.
1843 (f) If -A or -C was used with -c (count only), some lines of context were
1844 accidentally printed for the final match.
1846 (g) Added the -H (--with-filename) option.
1848 (h) The combination of options -rh failed to suppress file names for files
1849 that were found from directory arguments.
1851 (i) Added the -D (--devices) and -d (--directories) options.
1853 (j) Added the -F (--fixed-strings) option.
1855 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1857 (l) Added the --colo(u)r option.
1859 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1860 is not present by default.
1862 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1863 items such as (?R) or (?1), when the recursion could match a number of
1864 alternatives. If it matched one of the alternatives, but subsequently,
1865 outside the recursion, there was a failure, the code tried to back up into
1866 the recursion. However, because of the way PCRE is implemented, this is not
1867 possible, and the result was an incorrect result from the match.
1869 In order to prevent this happening, the specification of recursion has
1870 been changed so that all such subpatterns are automatically treated as
1871 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1873 4. I had overlooked the fact that, in some locales, there are characters for
1874 which isalpha() is true but neither isupper() nor islower() are true. In
1875 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1876 and ordfeminine) are like this. This affected the treatment of \w and \W
1877 when they appeared in character classes, but not when they appeared outside
1878 a character class. The bit map for "word" characters is now created
1879 separately from the results of isalnum() instead of just taking it from the
1880 upper, lower, and digit maps. (Plus the underscore character, of course.)
1882 5. The above bug also affected the handling of POSIX character classes such as
1883 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1884 permanent tables. Instead, the bit maps for such a class were previously
1885 created as the appropriate unions of the upper, lower, and digit bitmaps.
1886 Now they are created by subtraction from the [[:word:]] class, which has
1887 its own bitmap.
1889 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1890 It is created by subtracting the vertical space characters (\x09, \x0a,
1891 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1892 subtraction was done in the overall bitmap for a character class, meaning
1893 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1894 be recognized. This bug has been fixed.
1896 7. Patches from the folks at Google:
1898 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1899 real life, but is still worth protecting against".
1901 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1902 regular expressions".
1904 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1905 have it.
1907 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1908 "configure" and the latter not, in order to fix a problem somebody had
1909 with compiling the Arg class on HP-UX.
1911 (e) Improve the error-handling of the C++ wrapper a little bit.
1913 (f) New tests for checking recursion limiting.
1915 8. The pcre_memmove() function, which is used only if the environment does not
1916 have a standard memmove() function (and is therefore rarely compiled),
1917 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1918 returning a result (though PCRE never actually uses the result).
1920 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1921 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1922 returned instead of calling malloc() with an overflowing number that would
1923 most likely cause subsequent chaos.
1925 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1927 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1928 with this option is matched, the nmatch and pmatch options of regexec() are
1929 ignored.
1931 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1932 provided in case anyone wants to the the POSIX interface with UTF-8
1933 strings.
1935 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1936 C++ linking (needed for some HP-UX environments).
1938 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1939 (unused parameter) and in the pcre_printint() function (omitted "default"
1940 switch label when the default is to do nothing).
1942 15. Added some code to make it possible, when PCRE is compiled as a C++
1943 library, to replace subject pointers for pcre_exec() with a smart pointer
1944 class, thus making it possible to process discontinuous strings.
1946 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1947 much the same function. They were added by different people who were trying
1948 to make PCRE easy to compile on non-Unix systems. It has been suggested
1949 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1950 for compiling on Windows systems. I have therefore replaced it with
1951 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1952 defaults to "extern" for C or "extern C" for C++, which works fine on
1953 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1954 SCOPE with something explicit in config.h. In addition:
1956 (a) pcreposix.h still had just "extern" instead of either of these macros;
1957 I have replaced it with PCRE_DATA_SCOPE.
1959 (b) Functions such as _pcre_xclass(), which are internal to the library,
1960 but external in the C sense, all had PCRE_EXPORT in their definitions.
1961 This is apparently wrong for the Windows case, so I have removed it.
1962 (It makes no difference on Unix-like systems.)
1964 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1965 of recursive calls to match(). This is different to MATCH_LIMIT because
1966 that limits the total number of calls to match(), not all of which increase
1967 the depth of recursion. Limiting the recursion depth limits the amount of
1968 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1969 when PCRE is compiled, and changed at run time. A patch from Google adds
1970 this functionality to the C++ interface.
1972 18. Changes to the handling of Unicode character properties:
1974 (a) Updated the table to Unicode 4.1.0.
1976 (b) Recognize characters that are not in the table as "Cn" (undefined).
1978 (c) I revised the way the table is implemented to a much improved format
1979 which includes recognition of ranges. It now supports the ranges that
1980 are defined in UnicodeData.txt, and it also amalgamates other
1981 characters into ranges. This has reduced the number of entries in the
1982 table from around 16,000 to around 3,000, thus reducing its size
1983 considerably. I realized I did not need to use a tree structure after
1984 all - a binary chop search is just as efficient. Having reduced the
1985 number of entries, I extended their size from 6 bytes to 8 bytes to
1986 allow for more data.
1988 (d) Added support for Unicode script names via properties such as \p{Han}.
1990 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1991 matching that character.
1993 20. When matching a repeated Unicode property with a minimum greater than zero,
1994 (for example \pL{2,}), PCRE could look past the end of the subject if it
1995 reached it while seeking the minimum number of characters. This could
1996 happen only if some of the characters were more than one byte long, because
1997 there is a check for at least the minimum number of bytes.
1999 21. Refactored the implementation of \p and \P so as to be more general, to
2000 allow for more different types of property in future. This has changed the
2001 compiled form incompatibly. Anybody with saved compiled patterns that use
2002 \p or \P will have to recompile them.
2004 22. Added "Any" and "L&" to the supported property types.
2006 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2007 but give a compile time error if the value is greater than 0xff.
2009 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2010 accidentally not being installed or uninstalled.
2012 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2013 made were to insert the current release number. This seemed silly, because
2014 it made things harder for people building PCRE on systems that don't run
2015 "configure". I have turned pcre.h into a distributed file, no longer built
2016 by "configure", with the version identification directly included. There is
2017 no longer a pcre.h.in file.
2019 However, this change necessitated a change to the pcre-config script as
2020 well. It is built from pcre-config.in, and one of the substitutions was the
2021 release number. I have updated configure.ac so that ./configure now finds
2022 the release number by grepping pcre.h.
2024 26. Added the ability to run the tests under valgrind.
2027 Version 6.4 05-Sep-05
2028 ---------------------
2030 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2031 "--" to be printed when multiple files were scanned, even when none of the
2032 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2033 consider it to be a bug, and have restored the previous behaviour.
2035 2. A couple of code tidies to get rid of compiler warnings.
2037 3. The pcretest program used to cheat by referring to symbols in the library
2038 whose names begin with _pcre_. These are internal symbols that are not
2039 really supposed to be visible externally, and in some environments it is
2040 possible to suppress them. The cheating is now confined to including
2041 certain files from the library's source, which is a bit cleaner.
2043 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2044 file's purpose clearer.
2046 5. Reorganized pcre_ucp_findchar().
2049 Version 6.3 15-Aug-05
2050 ---------------------
2052 1. The file libpcre.pc.in did not have general read permission in the tarball.
2054 2. There were some problems when building without C++ support:
2056 (a) If C++ support was not built, "make install" and "make test" still
2057 tried to test it.
2059 (b) There were problems when the value of CXX was explicitly set. Some
2060 changes have been made to try to fix these, and ...
2062 (c) --disable-cpp can now be used to explicitly disable C++ support.
2064 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2065 backslash in a target when C++ was disabled. This confuses some
2066 versions of "make", apparently. Using an intermediate variable solves
2067 this. (Same for CPP_LOBJ.)
2069 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2070 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2071 necessary on certain architectures.
2073 4. Added a setting of -export-symbols-regex to the link command to remove
2074 those symbols that are exported in the C sense, but actually are local
2075 within the library, and not documented. Their names all begin with
2076 "_pcre_". This is not a perfect job, because (a) we have to except some
2077 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2078 available (and never for static libraries). I have made a note to try to
2079 find a way round (a) in the future.
2082 Version 6.2 01-Aug-05
2083 ---------------------
2085 1. There was no test for integer overflow of quantifier values. A construction
2086 such as {1111111111111111} would give undefined results. What is worse, if
2087 a minimum quantifier for a parenthesized subpattern overflowed and became
2088 negative, the calculation of the memory size went wrong. This could have
2089 led to memory overwriting.
2091 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2093 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2094 operating environments where this matters.
2096 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2097 PCRE options from within the C++ wrapper.
2099 5. Named capturing subpatterns were not being correctly counted when a pattern
2100 was compiled. This caused two problems: (a) If there were more than 100
2101 such subpatterns, the calculation of the memory needed for the whole
2102 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2103 back references of the form \12, where the number was greater than 9, were
2104 not recognized as back references, even though there were sufficient
2105 previous subpatterns.
2107 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2108 versions of gcc, e.g. 2.95.4.
2111 Version 6.1 21-Jun-05
2112 ---------------------
2114 1. There was one reference to the variable "posix" in pcretest.c that was not
2115 surrounded by "#if !defined NOPOSIX".
2117 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2118 the cross-check on the old pcre_info() function, for the benefit of the
2119 cut-down version of PCRE that is currently imported into Exim.
2121 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2122 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2123 patterns that start (?i) but I don't think that matters. The use of (?i) is
2124 just an example; this all applies to the other options as well.
2126 4. Since libtool seems to echo the compile commands it is issuing, the output
2127 from "make" can be reduced a bit by putting "@" in front of each libtool
2128 compile command.
2130 5. Patch from the folks at Google for configure.in to be a bit more thorough
2131 in checking for a suitable C++ installation before trying to compile the
2132 C++ stuff. This should fix a reported problem when a compiler was present,
2133 but no suitable headers.
2135 6. The man pages all had just "PCRE" as their title. I have changed them to
2136 be the relevant file name. I have also arranged that these names are
2137 retained in the file doc/pcre.txt, which is a concatenation in text format
2138 of all the man pages except the little individual ones for each function.
2140 7. The NON-UNIX-USE file had not been updated for the different set of source
2141 files that come with release 6. I also added a few comments about the C++
2142 wrapper.
2145 Version 6.0 07-Jun-05
2146 ---------------------
2148 1. Some minor internal re-organization to help with my DFA experiments.
2150 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2151 didn't matter for the library itself when fully configured, but did matter
2152 when compiling without UCP support, or within Exim, where the ucp files are
2153 not imported.
2155 3. Refactoring of the library code to split up the various functions into
2156 different source modules. The addition of the new DFA matching code (see
2157 below) to a single monolithic source would have made it really too
2158 unwieldy, quite apart from causing all the code to be include in a
2159 statically linked application, when only some functions are used. This is
2160 relevant even without the DFA addition now that patterns can be compiled in
2161 one application and matched in another.
2163 The downside of splitting up is that there have to be some external
2164 functions and data tables that are used internally in different modules of
2165 the library but which are not part of the API. These have all had their
2166 names changed to start with "_pcre_" so that they are unlikely to clash
2167 with other external names.
2169 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2170 a different (DFA) algorithm. Although it is slower than the original
2171 function, it does have some advantages for certain types of matching
2172 problem.
2174 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2175 including restarting after a partial match.
2177 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2178 defined when compiling for Windows was sent to me. I have put it into the
2179 code, though I have no means of testing or verifying it.
2181 7. Added the pcre_refcount() auxiliary function.
2183 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2184 match before or at the first newline in the subject string. In pcretest,
2185 the /f option on a pattern can be used to set this.
2187 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2188 would behave wrongly. This has been present in PCRE since release 4.0.
2190 10. A number of changes to the pcregrep command:
2192 (a) Refactored how -x works; insert ^(...)$ instead of setting
2193 PCRE_ANCHORED and checking the length, in preparation for adding
2194 something similar for -w.
2196 (b) Added the -w (match as a word) option.
2198 (c) Refactored the way lines are read and buffered so as to have more
2199 than one at a time available.
2201 (d) Implemented a pcregrep test script.
2203 (e) Added the -M (multiline match) option. This allows patterns to match
2204 over several lines of the subject. The buffering ensures that at least
2205 8K, or the rest of the document (whichever is the shorter) is available
2206 for matching (and similarly the previous 8K for lookbehind assertions).
2208 (f) Changed the --help output so that it now says
2210 -w, --word-regex(p)
2212 instead of two lines, one with "regex" and the other with "regexp"
2213 because that confused at least one person since the short forms are the
2214 same. (This required a bit of code, as the output is generated
2215 automatically from a table. It wasn't just a text change.)
2217 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2218 option but starts with a hyphen. Could be a pattern or a path name
2219 starting with a hyphen, for instance.
2221 (h) "-" can be given as a file name to represent stdin.
2223 (i) When file names are being printed, "(standard input)" is used for
2224 the standard input, for compatibility with GNU grep. Previously
2225 "<stdin>" was used.
2227 (j) The option --label=xxx can be used to supply a name to be used for
2228 stdin when file names are being printed. There is no short form.
2230 (k) Re-factored the options decoding logic because we are going to add
2231 two more options that take data. Such options can now be given in four
2232 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2234 (l) Added the -A, -B, and -C options for requesting that lines of context
2235 around matches be printed.
2237 (m) Added the -L option to print the names of files that do not contain
2238 any matching lines, that is, the complement of -l.
2240 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2241 continue to scan other files.
2243 (o) The -s option was incorrectly implemented. For compatibility with other
2244 greps, it now suppresses the error message for a non-existent or non-
2245 accessible file (but not the return code). There is a new option called
2246 -q that suppresses the output of matching lines, which was what -s was
2247 previously doing.
2249 (p) Added --include and --exclude options to specify files for inclusion
2250 and exclusion when recursing.
2252 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2253 Hopefully, it now does.
2255 12. Missing cast in pcre_study().
2257 13. Added an "uninstall" target to the makefile.
2259 14. Replaced "extern" in the function prototypes in Makefile.in with
2260 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2261 world, but is set differently for Windows.
2263 15. Added a second compiling function called pcre_compile2(). The only
2264 difference is that it has an extra argument, which is a pointer to an
2265 integer error code. When there is a compile-time failure, this is set
2266 non-zero, in addition to the error test pointer being set to point to an
2267 error message. The new argument may be NULL if no error number is required
2268 (but then you may as well call pcre_compile(), which is now just a
2269 wrapper). This facility is provided because some applications need a
2270 numeric error indication, but it has also enabled me to tidy up the way
2271 compile-time errors are handled in the POSIX wrapper.
2273 16. Added VPATH=.libs to the makefile; this should help when building with one
2274 prefix path and installing with another. (Or so I'm told by someone who
2275 knows more about this stuff than I do.)
2277 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2278 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2279 match everything, including newlines. This is not POSIX-compatible, but
2280 somebody wanted the feature. From pcretest it can be activated by using
2281 both the P and the s flags.
2283 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2285 19. libpcre.pc was being incorrectly installed as executable.
2287 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2288 it now also looks for '\r' so that it will work unmodified on Windows.
2290 21. Added Google's contributed C++ wrapper to the distribution.
2292 22. Added some untidy missing memory free() calls in pcretest, to keep
2293 Electric Fence happy when testing.
2297 Version 5.0 13-Sep-04
2298 ---------------------
2300 1. Internal change: literal characters are no longer packed up into items
2301 containing multiple characters in a single byte-string. Each character
2302 is now matched using a separate opcode. However, there may be more than one
2303 byte in the character in UTF-8 mode.
2305 2. The pcre_callout_block structure has two new fields: pattern_position and
2306 next_item_length. These contain the offset in the pattern to the next match
2307 item, and its length, respectively.
2309 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2310 insertion of callouts before each pattern item. Added the /C option to
2311 pcretest to make use of this.
2313 4. On the advice of a Windows user, the lines
2315 #if defined(_WIN32) || defined(WIN32)
2316 _setmode( _fileno( stdout ), 0x8000 );
2317 #endif /* defined(_WIN32) || defined(WIN32) */
2319 have been added to the source of pcretest. This apparently does useful
2320 magic in relation to line terminators.
2322 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2323 for the benefit of those environments where the "b" makes a difference.
2325 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2326 to know about it. I have put a hack into configure.in that adds in code
2327 to set GCC=yes if CC=icc. This seems to end up at a point in the
2328 generated configure script that is early enough to affect the setting of
2329 compiler options, which is what is needed, but I have no means of testing
2330 whether it really works. (The user who reported this had patched the
2331 generated configure script, which of course I cannot do.)
2333 LATER: After change 22 below (new libtool files), the configure script
2334 seems to know about icc (and also ecc). Therefore, I have commented out
2335 this hack in configure.in.
2337 7. Added support for pkg-config (2 patches were sent in).
2339 8. Negated POSIX character classes that used a combination of internal tables
2340 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2341 [[:^ascii]]. Typically, they would match almost any characters. The other
2342 POSIX classes were not broken in this way.
2344 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2345 to find the match, as PCRE was deluded into thinking that the match had to
2346 start at the start point or following a newline. The same bug applied to
2347 patterns with negative forward assertions or any backward assertions
2348 preceding ".*" at the start, unless the pattern required a fixed first
2349 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2351 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2352 starting at the last subject character, bytes beyond the end of the subject
2353 string were read.
2355 11. Renamed the variable "class" as "classbits" to make life easier for C++
2356 users. (Previously there was a macro definition, but it apparently wasn't
2357 enough.)
2359 12. Added the new field "tables" to the extra data so that tables can be passed
2360 in at exec time, or the internal tables can be re-selected. This allows
2361 a compiled regex to be saved and re-used at a later time by a different
2362 program that might have everything at different addresses.
2364 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2365 -R library as well as a -L library.
2367 14. The debugging options of pcretest (-d on the command line or D on a
2368 pattern) showed incorrect output for anything following an extended class
2369 that contained multibyte characters and which was followed by a quantifier.
2371 15. Added optional support for general category Unicode character properties
2372 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2373 support. It adds about 90K to the size of the library. The meanings of the
2374 inbuilt class escapes such as \d and \s have NOT been changed.
2376 16. Updated pcredemo.c to include calls to free() to release the memory for the
2377 compiled pattern.
2379 17. The generated file chartables.c was being created in the source directory
2380 instead of in the building directory. This caused the build to fail if the
2381 source directory was different from the building directory, and was
2382 read-only.
2384 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2385 file. No doubt somebody will tell me if they don't make sense... Also added
2386 Dan Mooney's comments about building on OpenVMS.
2388 19. Added support for partial matching via the PCRE_PARTIAL option for
2389 pcre_exec() and the \P data escape in pcretest.
2391 20. Extended pcretest with 3 new pattern features:
2393 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2394 write the compiled pattern to the file whose name is "rest-of-line".
2395 This is a straight binary dump of the data, with the saved pointer to
2396 the character tables forced to be NULL. The study data, if any, is
2397 written too. After writing, pcretest reads a new pattern.
2399 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2400 compiled pattern from the given file. There must not be any
2401 occurrences of "<" in the file name (pretty unlikely); if there are,
2402 pcretest will instead treat the initial "<" as a pattern delimiter.
2403 After reading in the pattern, pcretest goes on to read data lines as
2404 usual.
2406 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2407 and 16-bit fields in a compiled pattern, to simulate a pattern that
2408 was compiled on a host of opposite endianness.
2410 21. The pcre-exec() function can now cope with patterns that were compiled on
2411 hosts of opposite endianness, with this restriction:
2413 As for any compiled expression that is saved and used later, the tables
2414 pointer field cannot be preserved; the extra_data field in the arguments
2415 to pcre_exec() should be used to pass in a tables address if a value
2416 other than the default internal tables were used at compile time.
2418 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2419 now diagnosed as an error. Previously, most of the time, a negative number
2420 would have been treated as zero, but if in addition "ovector" was passed as
2421 NULL, a crash could occur.
2423 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2424 new versions from the libtool 1.5 distribution (the last one is a copy of
2425 a file called libtool.m4). This seems to have fixed the need to patch
2426 "configure" to support Darwin 1.3 (which I used to do). However, I still
2427 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2428 workstation).
2430 24. Changed the PCRE licence to be the more standard "BSD" licence.
2433 Version 4.5 01-Dec-03
2434 ---------------------
2436 1. There has been some re-arrangement of the code for the match() function so
2437 that it can be compiled in a version that does not call itself recursively.
2438 Instead, it keeps those local variables that need separate instances for
2439 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2440 needs to "recurse". Keeping track of where control must go is done by means
2441 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2442 hide most of the details from the main code, and operates only if
2443 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2444 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2445 operating.
2447 To make it easier for callers to provide specially tailored get/free
2448 functions for this usage, two new functions, pcre_stack_malloc, and
2449 pcre_stack_free, are used. They are always called in strict stacking order,
2450 and the size of block requested is always the same.
2452 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2453 PCRE has been compiled to use the stack or the heap for recursion. The
2454 -C option of pcretest uses this to show which version is compiled.
2456 A new data escape \S, is added to pcretest; it causes the amounts of store
2457 obtained and freed by both kinds of malloc/free at match time to be added
2458 to the output.
2460 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2461 what's available on my current Linux desktop machine.
2463 3. When matching a UTF-8 string, the test for a valid string at the start has
2464 been extended. If start_offset is not zero, PCRE now checks that it points
2465 to a byte that is the start of a UTF-8 character. If not, it returns
2466 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2467 this is necessary because there may be backward assertions in the pattern.
2468 When matching the same subject several times, it may save resources to use
2469 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2471 4. The code for checking the validity of UTF-8 strings has been tightened so
2472 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2473 containing "overlong sequences".
2475 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2476 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2477 should have been "&", but it just so happened that all the cases this let
2478 through by mistake were picked up later in the function.
2480 6. I had used a variable called "isblank" - this is a C99 function, causing
2481 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2483 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2484 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2485 so the version is shown; (c) stop "make test" echoing "./RunTest".
2487 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2489 9. The prototype for memmove() for systems that don't have it was using
2490 size_t, but the inclusion of the header that defines size_t was later. I've
2491 moved the #includes for the C headers earlier to avoid this.
2493 10. Added some adjustments to the code to make it easier to compiler on certain
2494 special systems:
2496 (a) Some "const" qualifiers were missing.
2497 (b) Added the macro EXPORT before all exported functions; by default this
2498 is defined to be empty.
2499 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2500 that it reads its output file name as an argument instead of writing
2501 to the standard output and assuming this can be redirected.
2503 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2504 class containing characters with values greater than 255, PCRE compilation
2505 went into a loop.
2507 12. A recursive reference to a subpattern that was within another subpattern
2508 that had a minimum quantifier of zero caused PCRE to crash. For example,
2509 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2510 recursion. If the recursively-called subpattern itself had a zero repeat,
2511 that was OK.
2513 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2514 buffer into which it was copied (for escape processing) was still set at
2515 1024, so long lines caused crashes.
2517 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2518 "internal error: code overflow...". This applied to any character class
2519 that was followed by a possessive quantifier.
2521 15. Modified the Makefile to add libpcre.la as a prerequisite for
2522 libpcreposix.la because I was told this is needed for a parallel build to
2523 work.
2525 16. If a pattern that contained .* following optional items at the start was
2526 studied, the wrong optimizing data was generated, leading to matching
2527 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2528 matching string must start with a or b or c. The correct conclusion for
2529 this pattern is that a match can start with any character.
2532 Version 4.4 13-Aug-03
2533 ---------------------
2535 1. In UTF-8 mode, a character class containing characters with values between
2536 127 and 255 was not handled correctly if the compiled pattern was studied.
2537 In fixing this, I have also improved the studying algorithm for such
2538 classes (slightly).
2540 2. Three internal functions had redundant arguments passed to them. Removal
2541 might give a very teeny performance improvement.
2543 3. Documentation bug: the value of the capture_top field in a callout is *one
2544 more than* the number of the hightest numbered captured substring.
2546 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2547 in incorrectly linking with a previously installed version. They now link
2548 explicitly with libpcre.la.
2550 5. configure.in no longer needs to recognize Cygwin specially.
2552 6. A problem in pcre.in for Windows platforms is fixed.
2554 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2555 pcretest, it used to include the size of the study block as part of its
2556 output. Unfortunately, the structure contains a field that has a different
2557 size on different hardware architectures. This meant that the tests that
2558 showed this size failed. As the block is currently always of a fixed size,
2559 this information isn't actually particularly useful in pcretest output, so
2560 I have just removed it.
2562 8. Three pre-processor statements accidentally did not start in column 1.
2563 Sadly, there are *still* compilers around that complain, even though
2564 standard C has not required this for well over a decade. Sigh.
2566 9. In pcretest, the code for checking callouts passed small integers in the
2567 callout_data field, which is a void * field. However, some picky compilers
2568 complained about the casts involved for this on 64-bit systems. Now
2569 pcretest passes the address of the small integer instead, which should get
2570 rid of the warnings.
2572 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2573 both compile and run time, and gives an error if an invalid UTF-8 sequence
2574 is found. There is a option for disabling this check in cases where the
2575 string is known to be correct and/or the maximum performance is wanted.
2577 11. In response to a bug report, I changed one line in Makefile.in from
2579 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2580 to
2581 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2583 to look similar to other lines, but I have no way of telling whether this
2584 is the right thing to do, as I do not use Windows. No doubt I'll get told
2585 if it's wrong...
2588 Version 4.3 21-May-03
2589 ---------------------
2591 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2592 Makefile.
2594 2. Some refactoring to improve the quality of the code:
2596 (i) The utf8_table... variables are now declared "const".
2598 (ii) The code for \cx, which used the "case flipping" table to upper case
2599 lower case letters, now just substracts 32. This is ASCII-specific,
2600 but the whole concept of \cx is ASCII-specific, so it seems
2601 reasonable.
2603 (iii) PCRE was using its character types table to recognize decimal and
2604 hexadecimal digits in the pattern. This is silly, because it handles
2605 only 0-9, a-f, and A-F, but the character types table is locale-
2606 specific, which means strange things might happen. A private
2607 table is now used for this - though it costs 256 bytes, a table is
2608 much faster than multiple explicit tests. Of course, the standard
2609 character types table is still used for matching digits in subject
2610 strings against \d.
2612 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2613 ending in _t are). So I've renamed it as ESC_tee.
2615 3. The first argument for regexec() in the POSIX wrapper should have been
2616 defined as "const".
2618 4. Changed pcretest to use malloc() for its buffers so that they can be
2619 Electric Fenced for debugging.
2621 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2622 to read one or more bytes before the start of the subject string. Often this
2623 had no effect on PCRE's behaviour, but in some circumstances it could
2624 provoke a segmentation fault.
2626 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2627 to try to read one or more bytes before the start of the subject string.
2629 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2630 UTF-8 support could misbehave in various ways if the subject string
2631 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2632 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2633 back over UTF-8 characters.)
2636 Version 4.2 14-Apr-03
2637 ---------------------
2639 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2641 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2642 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2643 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2644 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2645 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2647 Note: automatic setting of the BUILD variables is not yet working
2648 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2649 compile-time but not at link-time
2650 [LINK]: use for linking executables only
2651 make different versions for Windows and non-Windows
2652 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2653 libraries
2654 [LINK_FOR_BUILD]: new variable
2655 [OBJEXT]: use throughout
2656 [EXEEXT]: use throughout
2657 <winshared>: new target
2658 <wininstall>: new target
2659 <dftables.o>: use native compiler
2660 <dftables>: use native linker
2661 <install>: handle Windows platform correctly
2662 <clean>: ditto
2663 <check>: ditto
2664 copy DLL to top builddir before testing
2666 As part of these changes, -no-undefined was removed again. This was reported
2667 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2668 in any case.
2670 3. Some tidies to get rid of compiler warnings:
2672 . In the match_data structure, match_limit was an unsigned long int, whereas
2673 match_call_count was an int. I've made them both unsigned long ints.
2675 . In pcretest the fact that a const uschar * doesn't automatically cast to
2676 a void * provoked a warning.
2678 . Turning on some more compiler warnings threw up some "shadow" variables
2679 and a few more missing casts.
2681 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2682 option, a class that contained a single character with a value between 128
2683 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2685 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2686 option, a class that contained several characters, but with at least one
2687 whose value was between 128 and 255 caused PCRE to crash.
2690 Version 4.1 12-Mar-03
2691 ---------------------
2693 1. Compiling with gcc -pedantic found a couple of places where casts were
2694 needed, and a string in dftables.c that was longer than standard compilers are
2695 required to support.
2697 2. Compiling with Sun's compiler found a few more places where the code could
2698 be tidied up in order to avoid warnings.
2700 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2701 first of these names is deprecated in the latest Autoconf in favour of the name
2702 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2703 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2704 analogy I have changed it to CFLAGS_FOR_BUILD.
2706 4. Added -no-undefined to the linking command in the Makefile, because this is
2707 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2708 linking step for the pcreposix library.
2710 5. PCRE was failing to diagnose the case of two named groups with the same
2711 name.
2713 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2714 literal character that is needed in the subject for a match, and scans along to
2715 ensure that it is present before embarking on the full matching process. This
2716 saves time in cases of nested unlimited repeats that are never going to match.
2717 Problem: the scan can take a lot of time if the subject is very long (e.g.
2718 megabytes), thus penalizing straightforward matches. It is now done only if the
2719 amount of subject to be scanned is less than 1000 bytes.
2721 7. A lesser problem with the same optimization is that it was recording the
2722 first character of an anchored pattern as "needed", thus provoking a search
2723 right along the subject, even when the first match of the pattern was going to
2724 fail. The "needed" character is now not set for anchored patterns, unless it
2725 follows something in the pattern that is of non-fixed length. Thus, it still
2726 fulfils its original purpose of finding quick non-matches in cases of nested
2727 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2730 Version 4.0 17-Feb-03
2731 ---------------------
2733 1. If a comment in an extended regex that started immediately after a meta-item
2734 extended to the end of string, PCRE compiled incorrect data. This could lead to
2735 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2737 2. Moved to autoconf 2.53 and libtool 1.4.2.
2739 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2740 the special perltest8 script is no longer needed - all the tests can be run
2741 from a single perltest script.
2743 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2744 by \s. It has now been removed in PCRE. This means it isn't recognized as
2745 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2746 class [:space:] *does* include VT, thereby creating a mess.
2748 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2749 space and tab.
2751 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2752 its new features into the main test script, reducing the number of scripts.
2754 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2755 were backward compatible, and made the (?i) apply to the whole pattern, as if
2756 /i were given. Now it behaves more logically, and applies the option setting
2757 only to what follows. PCRE has been changed to follow suit. However, if it
2758 finds options settings right at the start of the pattern, it extracts them into
2759 the global options, as before. Thus, they show up in the info data.
2761 8. Added support for the \Q...\E escape sequence. Characters in between are
2762 treated as literals. This is slightly different from Perl in that $ and @ are
2763 also handled as literals inside the quotes. In Perl, they will cause variable
2764 interpolation. Note the following examples:
2766 Pattern PCRE matches Perl matches
2768 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2769 \Qabc\$xyz\E abc\$xyz abc\$xyz
2770 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2772 For compatibility with Perl, \Q...\E sequences are recognized inside character
2773 classes as well as outside them.
2775 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2776 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2777 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2778 signed/unsigned warnings.
2780 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2781 option for pcretest, so I've replaced it by a simple function that does just
2782 that job.
2784 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2785 "pcregrep -".
2787 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2788 Java package. This provides some syntactic sugar for simple cases of what my
2789 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2790 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2791 item, you can use this simplified notation. Note that only makes sense with
2792 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2793 greediness, whatever the setting of the PCRE_UNGREEDY option.
2795 13. A change of greediness default within a pattern was not taking effect at
2796 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2797 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2798 was abstracted outside.
2800 14. PCRE now supports the \G assertion. It is true when the current matching
2801 position is at the start point of the match. This differs from \A when the
2802 starting offset is non-zero. Used with the /g option of pcretest (or similar
2803 code), it works in the same way as it does for Perl's /g option. If all
2804 alternatives of a regex begin with \G, the expression is anchored to the start
2805 match position, and the "anchored" flag is set in the compiled expression.
2807 15. Some bugs concerning the handling of certain option changes within patterns
2808 have been fixed. These applied to options other than (?ims). For example,
2809 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2810 been the other way round. Some of this was related to change 7 above.
2812 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2813 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2814 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2815 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2817 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2818 mode. Unlike ".", it always matches newline, whatever the setting of
2819 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2820 assertions. Perl allows it, but it doesn't (in general) work because it can't
2821 calculate the length of the lookbehind. At least, that's the case for Perl
2822 5.8.0 - I've been told they are going to document that it doesn't work in
2823 future.
2825 18. Added an error diagnosis for escapes that PCRE does not support: these are
2826 \L, \l, \N, \P, \p, \U, \u, and \X.
2828 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2829 reading past the end of the pattern in cases such as /[abcd/.
2831 20. PCRE was getting more memory than necessary for patterns with classes that
2832 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2834 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2835 compiling PCRE for use with Virtual Pascal.
2837 22. Small fix to the Makefile to make it work properly if the build is done
2838 outside the source tree.
2840 23. Added a new extension: a condition to go with recursion. If a conditional
2841 subpattern starts with (?(R) the "true" branch is used if recursion has
2842 happened, whereas the "false" branch is used only at the top level.
2844 24. When there was a very long string of literal characters (over 255 bytes
2845 without UTF support, over 250 bytes with UTF support), the computation of how
2846 much memory was required could be incorrect, leading to segfaults or other
2847 strange effects.
2849 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2850 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2851 there was a subsequent back reference to those brackets. This meant that, for
2852 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2853 possible to check for precisely this case. All we can do is abandon the
2854 optimization if .* occurs inside capturing brackets when there are any back
2855 references whatsoever. (See below for a better fix that came later.)
2857 26. The handling of the optimization for finding the first character of a
2858 non-anchored pattern, and for finding a character that is required later in the
2859 match were failing in some cases. This didn't break the matching; it just
2860 failed to optimize when it could. The way this is done has been re-implemented.
2862 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2864 28. Added a new feature that provides some of the functionality that Perl
2865 provides with (?{...}). The facility is termed a "callout". The way it is done
2866 in PCRE is for the caller to provide an optional function, by setting
2867 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2868 global variable. By default it is unset, which disables all calling out. To get
2869 the function called, the regex must include (?C) at appropriate points. This
2870 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2871 This provides a means of identifying different callout points. When PCRE
2872 reaches such a point in the regex, if pcre_callout has been set, the external
2873 function is called. It is provided with data in a structure called
2874 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2875 matching continues; if it returns a non-zero value, the match at the current
2876 point fails. However, backtracking will occur if possible. [This was changed
2877 later and other features added - see item 49 below.]
2879 29. pcretest is upgraded to test the callout functionality. It provides a
2880 callout function that displays information. By default, it shows the start of
2881 the match and the current position in the text. There are some new data escapes
2882 to vary what happens:
2884 \C+ in addition, show current contents of captured substrings
2885 \C- do not supply a callout function
2886 \C!n return 1 when callout number n is reached
2887 \C!n!m return 1 when callout number n is reached for the mth time
2889 30. If pcregrep was called with the -l option and just a single file name, it
2890 output "<stdin>" if a match was found, instead of the file name.
2892 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2893 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2894 pcre_exec(). This saves a malloc/free per call. The default value of
2895 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2896 when configuring.
2898 32. The default maximum size of a compiled pattern is 64K. There have been a
2899 few cases of people hitting this limit. The code now uses macros to handle the
2900 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2901 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2902 configuring. Tests 2 and 5 work only with 2-byte links because they output
2903 debugging information about compiled patterns.
2905 33. Internal code re-arrangements:
2907 (a) Moved the debugging function for printing out a compiled regex into
2908 its own source file (printint.c) and used #include to pull it into
2909 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2910 separate copies.
2912 (b) Defined the list of op-code names for debugging as a macro in
2913 internal.h so that it is next to the definition of the opcodes.
2915 (c) Defined a table of op-code lengths for simpler skipping along compiled
2916 code. This is again a macro in internal.h so that it is next to the
2917 definition of the opcodes.
2919 34. Added support for recursive calls to individual subpatterns, along the
2920 lines of Robin Houston's patch (but implemented somewhat differently).
2922 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2923 allow it to read and process whole directories in Win32. This code was
2924 contributed by Lionel Fourquaux; it has not been tested by me.
2926 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2927 used to name a group. Names consist of alphanumerics and underscores, and must
2928 be unique. Back references use the syntax (?P=name) and recursive calls use
2929 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2930 numbers. The function pcre_fullinfo() can be used after compilation to extract
2931 a name/number map. There are three relevant calls:
2933 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2934 PCRE_INFO_NAMECOUNT yields the number of entries
2935 PCRE_INFO_NAMETABLE yields a pointer to the map.
2937 The map is a vector of fixed-size entries. The size of each entry depends on
2938 the length of the longest name used. The first two bytes of each entry are the
2939 group number, most significant byte first. There follows the corresponding
2940 name, zero terminated. The names are in alphabetical order.
2942 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2943 case instead of 255. Making it the same both with and without UTF-8 support
2944 means that the same test output works with both.
2946 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2947 calling malloc() with a zero argument.
2949 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2950 optimization. I've improved things by keeping a bitmap of backreferences with
2951 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2952 fact referenced, the optimization can be applied. It is unlikely that a
2953 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2954 the match to follow \n) will appear inside brackets with a number greater than
2955 31, but if it does, any back reference > 31 suppresses the optimization.
2957 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2958 of disabling numbered capturing parentheses. Any opening parenthesis that is
2959 not followed by ? behaves as if it were followed by ?: but named parentheses
2960 can still be used for capturing (and they will acquire numbers in the usual
2961 way).
2963 41. Redesigned the return codes from the match() function into yes/no/error so
2964 that errors can be passed back from deep inside the nested calls. A malloc
2965 failure while inside a recursive subpattern call now causes the
2966 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2968 42. It is now possible to set a limit on the number of times the match()
2969 function is called in a call to pcre_exec(). This facility makes it possible to
2970 limit the amount of recursion and backtracking, though not in a directly
2971 obvious way, because the match() function is used in a number of different
2972 circumstances. The count starts from zero for each position in the subject
2973 string (for non-anchored patterns). The default limit is, for compatibility, a
2974 large number, namely 10 000 000. You can change this in two ways:
2976 (a) When configuring PCRE before making, you can use --with-match-limit=n
2977 to set a default value for the compiled library.
2979 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2980 a different value is set. See 45 below.
2982 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2984 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2985 of things that can be changed at compile time. The first argument specifies
2986 what is wanted and the second points to where the information is to be placed.
2987 The current list of available information is:
2991 The output is an integer that is set to one if UTF-8 support is available;
2992 otherwise it is set to zero.
2996 The output is an integer that it set to the value of the code that is used for
2997 newline. It is either LF (10) or CR (13).
3001 The output is an integer that contains the number of bytes used for internal
3002 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3006 The output is an integer that contains the threshold above which the POSIX
3007 interface uses malloc() for output vectors. See item 31 above.
3011 The output is an unsigned integer that contains the default limit of the number
3012 of match() calls in a pcre_exec() execution. See 42 above.
3014 44. pcretest has been upgraded by the addition of the -C option. This causes it
3015 to extract all the available output from the new pcre_config() function, and to
3016 output it. The program then exits immediately.
3018 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3019 order to support additional features. One way would have been to define
3020 pcre_exec2() (for example) with extra arguments, but this would not have been
3021 extensible, and would also have required all calls to the original function to
3022 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3023 is used for passing in "extra" data from pcre_study().
3025 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3026 contains the following fields:
3028 flags a bitmap indicating which of the following fields are set
3029 study_data opaque data from pcre_study()
3030 match_limit a way of specifying a limit on match() calls for a specific
3031 call to pcre_exec()
3032 callout_data data for callouts (see 49 below)
3034 The flag bits are also defined in pcre.h, and are
3040 The pcre_study() function now returns one of these new pcre_extra blocks, with
3041 the actual study data pointed to by the study_data field, and the
3042 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3043 before. That is, this change is entirely upwards-compatible and requires no
3044 change to existing code.
3046 If you want to pass in additional data to pcre_exec(), you can either place it
3047 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3048 block.
3050 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3051 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3052 times with different match limits, until it finds the minimum value needed for
3053 pcre_exec() to complete. The value is then output. This can be instructive; for
3054 most simple matches the number is quite small, but for pathological cases it
3055 gets very large very quickly.
3057 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3058 returns the size of the data block pointed to by the study_data field in a
3059 pcre_extra block, that is, the value that was passed as the argument to
3060 pcre_malloc() when PCRE was getting memory in which to place the information
3061 created by pcre_study(). The fourth argument should point to a size_t variable.
3062 pcretest has been extended so that this information is shown after a successful
3063 pcre_study() call when information about the compiled regex is being displayed.
3065 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3066 because what follows is always an absolute path. (Later: it turns out that this
3067 is more than cosmetic for MinGW, because it doesn't like empty path
3068 components.)
3070 49. Some changes have been made to the callout feature (see 28 above):
3072 (i) A callout function now has three choices for what it returns:
3074 0 => success, carry on matching
3075 > 0 => failure at this point, but backtrack if possible
3076 < 0 => serious error, return this value from pcre_exec()
3078 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3079 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3080 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3081 use by callout functions. It will never be used by PCRE itself.
3083 (ii) The pcre_extra structure (see 45 above) has a void * field called
3084 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3085 pcre_callout_block structure has a field of the same name. The contents of
3086 the field passed in the pcre_extra structure are passed to the callout
3087 function in the corresponding field in the callout block. This makes it
3088 easier to use the same callout-containing regex from multiple threads. For
3089 testing, the pcretest program has a new data escape
3091 \C*n pass the number n (may be negative) as callout_data
3093 If the callout function in pcretest receives a non-zero value as
3094 callout_data, it returns that value.
3096 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3097 there were some redundant $(CFLAGS) in commands that are now specified as
3098 $(LINK), which already includes $(CFLAGS).
3100 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3101 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3102 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3103 one-byte characters throughout. Note that case-insensitive matching applies
3104 only to characters whose values are less than 256. PCRE doesn't support the
3105 notion of cases for higher-valued characters.
3107 (i) A character class whose characters are all within 0-255 is handled as
3108 a bit map, and the map is inverted for negative classes. Previously, a
3109 character > 255 always failed to match such a class; however it should
3110 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3112 (ii) A negated character class with a single character < 255 is coded as
3113 "not this character" (OP_NOT). This wasn't working properly when the test
3114 character was multibyte, either singly or repeated.
3116 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3117 mode, for example: \x{100}{2,3}.
3119 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3120 singly or repeated) now correctly test multibyte characters. However,
3121 PCRE doesn't recognize any characters with values greater than 255 as
3122 digits, spaces, or word characters. Such characters always match \D, \S,
3123 and \W, and never match \d, \s, or \w.
3125 (v) Classes may now contain characters and character ranges with values
3126 greater than 255. For example: [ab\x{100}-\x{400}].
3128 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3129 PCRE in UTF-8 mode.
3131 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3132 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3133 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3134 value.)
3136 53. The single man page has become too large. I have therefore split it up into
3137 a number of separate man pages. These also give rise to individual HTML pages;
3138 these are now put in a separate directory, and there is an index.html page that
3139 lists them all. Some hyperlinking between the pages has been installed.
3141 54. Added convenience functions for handling named capturing parentheses.
3143 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3144 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3145 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3146 are faulted.
3148 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3149 calling configure. These values are used when compiling the dftables.c program
3150 which is run to generate the source of the default character tables. They
3151 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3152 you will need to set these values.
3154 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3157 Version 3.9 02-Jan-02
3158 ---------------------
3160 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3162 2. If --disable-static was given, the building process failed when trying to
3163 build pcretest and pcregrep. (For some reason it was using libtool to compile
3164 them, which is not right, as they aren't part of the library.)
3167 Version 3.8 18-Dec-01
3168 ---------------------
3170 1. The experimental UTF-8 code was completely screwed up. It was packing the
3171 bytes in the wrong order. How dumb can you get?
3174 Version 3.7 29-Oct-01
3175 ---------------------
3177 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3178 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3179 this didn't happen under Solaris 8, where I normally test things.
3181 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3182 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3183 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3184 this fix an hour or so after the initial 3.7 release.)
3187 Version 3.6 23-Oct-01
3188 ---------------------
3190 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3191 offsets passed as NULL with zero offset count.
3193 2. The config.guess and config.sub files had not been updated when I moved to
3194 the latest autoconf.
3197 Version 3.5 15-Aug-01
3198 ---------------------
3200 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3201 had been forgotten.
3203 2. By using declared but undefined structures, we can avoid using "void"
3204 definitions in pcre.h while keeping the internal definitions of the structures
3205 private.
3207 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3208 user point of view, this means that both static and shared libraries are built
3209 by default, but this can be individually controlled. More of the work of
3210 handling this static/shared cases is now inside libtool instead of PCRE's make
3211 file.
3213 4. The pcretest utility is now installed along with pcregrep because it is
3214 useful for users (to test regexs) and by doing this, it automatically gets
3215 relinked by libtool. The documentation has been turned into a man page, so
3216 there are now .1, .txt, and .html versions in /doc.
3218 5. Upgrades to pcregrep:
3219 (i) Added long-form option names like gnu grep.
3220 (ii) Added --help to list all options with an explanatory phrase.
3221 (iii) Added -r, --recursive to recurse into sub-directories.
3222 (iv) Added -f, --file to read patterns from a file.
3224 6. pcre_exec() was referring to its "code" argument before testing that
3225 argument for NULL (and giving an error if it was NULL).
3227 7. Upgraded Makefile.in to allow for compiling in a different directory from
3228 the source directory.
3230 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3231 options bits, the pointer it was passed was to an int instead of to an unsigned
3232 long int. This mattered only on 64-bit systems.
3234 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3235 generated) instead of pcre.in, which it its source. Also made the same change
3236 in several of the .c files.
3238 10. A new release of gcc defines printf() as a macro, which broke pcretest
3239 because it had an ifdef in the middle of a string argument for printf(). Fixed
3240 by using separate calls to printf().
3242 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3243 script, to force use of CR or LF instead of \n in the source. On non-Unix
3244 systems, the value can be set in config.h.
3246 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3247 absolute limit. Changed the text of the error message to make this clear, and
3248 likewise updated the man page.
3250 13. The limit of 99 on the number of capturing subpatterns has been removed.
3251 The new limit is 65535, which I hope will not be a "real" limit.
3254 Version 3.4 22-Aug-00
3255 ---------------------
3257 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3259 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3262 Version 3.3 01-Aug-00
3263 ---------------------
3265 1. If an octal character was given, but the value was greater than \377, it
3266 was not getting masked to the least significant bits, as documented. This could
3267 lead to crashes in some systems.
3269 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3270 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3272 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3273 These just pass their arguments on to (pcre_free)(), but they are provided
3274 because some uses of PCRE bind it to non-C systems that can call its functions,
3275 but cannot call free() or pcre_free() directly.
3277 4. Add "make test" as a synonym for "make check". Corrected some comments in
3278 the Makefile.
3280 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3281 Makefile.
3283 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3284 command called pgrep for grepping around the active processes.
3286 7. Added the beginnings of support for UTF-8 character strings.
3288 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3289 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3290 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3291 out for the ar command.)
3294 Version 3.2 12-May-00
3295 ---------------------
3297 This is purely a bug fixing release.
3299 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3300 of ZA. This was just one example of several cases that could provoke this bug,
3301 which was introduced by change 9 of version 2.00. The code for breaking
3302 infinite loops after an iteration that matches an empty string was't working
3303 correctly.
3305 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3306 when matched against abbab (for example). After matching an empty string, it
3307 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3308 caused it to match further down the string than it should.
3310 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3311 was there because it doesn't seem to be needed, and it causes trouble on some
3312 systems, as it is not a Standard C header. It has been removed.
3314 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3315 were reported on the Macintosh. The changes were from
3317 while ((c = *(++ptr)) != 0 && c != '\n');
3318 to
3319 while ((c = *(++ptr)) != 0 && c != '\n') ;
3321 Totally extraordinary, but if that's what it takes...
3323 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3324 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3325 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3326 assumes the way PCRE uses memmove() (always moving upwards).
3328 6. PCRE is being used in one environment where strchr() is not available. There
3329 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3330 faster code anyway.
3333 Version 3.1 09-Feb-00
3334 ---------------------
3336 The only change in this release is the fixing of some bugs in Makefile.in for
3337 the "install" target:
3339 (1) It was failing to install pcreposix.h.
3341 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3344 Version 3.0 01-Feb-00
3345 ---------------------
3347 1. Add support for the /+ modifier to perltest (to output $` like it does in
3348 pcretest).
3350 2. Add support for the /g modifier to perltest.
3352 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3353 matches null strings.
3355 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3356 pattern. Perl treats empty patterns specially - it reuses the most recent
3357 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3358 effect.
3360 5. The POSIX interface was broken in that it was just handing over the POSIX
3361 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3362 required a bigger vector, with some working space on the end. This means that
3363 the POSIX wrapper now has to get and free some memory, and copy the results.
3365 6. Added some simple autoconf support, placing the test data and the
3366 documentation in separate directories, re-organizing some of the
3367 information files, and making it build pcre-config (a GNU standard). Also added
3368 libtool support for building PCRE as a shared library, which is now the
3369 default.
3371 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3372 09 are not valid octal constants. Single digits will be used for minor values
3373 less than 10.
3375 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3376 existing programs that set these in the POSIX interface can use PCRE without
3377 modification.
3379 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3380 return all that pcre_info() returns, plus additional data. The pcre_info()
3381 function is retained for compatibility, but is considered to be obsolete.
3383 10. Added experimental recursion feature (?R) to handle one common case that
3384 Perl 5.6 will be able to do with (?p{...}).
3386 11. Added support for POSIX character classes like [:alpha:], which Perl is
3387 adopting.
3390 Version 2.08 31-Aug-99
3391 ----------------------
3393 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3394 trying to match at the startoffset position, but instead was moving forward to
3395 the next newline as if a previous match had failed.
3397 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3398 and could get into a loop if a null string was matched other than at the start
3399 of the subject.
3401 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3402 be distinguished at compile time, and for completeness also added PCRE_DATE.
3404 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3405 in GnuWin32 environments.
3408 Version 2.07 29-Jul-99
3409 ----------------------
3411 1. The documentation is now supplied in plain text form and HTML as well as in
3412 the form of man page sources.
3414 2. C++ compilers don't like assigning (void *) values to other pointer types.
3415 In particular this affects malloc(). Although there is no problem in Standard
3416 C, I've put in casts to keep C++ compilers happy.
3418 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3419 should be (const char *).
3421 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3422 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3423 However, I haven't made this a standard facility. The documentation doesn't
3424 mention it, and the Makefile doesn't support it.
3426 5. The Makefile now contains an "install" target, with editable destinations at
3427 the top of the file. The pcretest program is not installed.
3429 6. pgrep -V now gives the PCRE version number and date.
3431 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3432 causing the entire string to be ignored, instead of just the last character.
3434 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3435 non-matching string, it can take a very, very long time, even for strings of
3436 quite modest length, because of the nested recursion. PCRE now does better in
3437 some of these cases. It does this by remembering the last required literal
3438 character in the pattern, and pre-searching the subject to ensure it is present
3439 before running the real match. In other words, it applies a heuristic to detect
3440 some types of certain failure quickly, and in the above example, if presented
3441 with a string that has no trailing " it gives "no match" very quickly.
3443 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3444 other alternatives are tried instead.
3447 Version 2.06 09-Jun-99
3448 ----------------------
3450 1. Change pcretest's output for amount of store used to show just the code
3451 space, because the remainder (the data block) varies in size between 32-bit and
3452 64-bit systems.
3454 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3455 start matching at. This allows lookbehinds to work when searching for multiple
3456 occurrences in a string.
3458 3. Added additional options to pcretest for testing multiple occurrences:
3460 /+ outputs the rest of the string that follows a match
3461 /g loops for multiple occurrences, using the new startoffset argument
3462 /G loops for multiple occurrences by passing an incremented pointer
3464 4. PCRE wasn't doing the "first character" optimization for patterns starting
3465 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3466 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3467 the letter 'x'. On long subject strings, this gives a significant speed-up.
3470 Version 2.05 21-Apr-99
3471 ----------------------
3473 1. Changed the type of magic_number from int to long int so that it works
3474 properly on 16-bit systems.
3476 2. Fixed a bug which caused patterns starting with .* not to work correctly
3477 when the subject string contained newline characters. PCRE was assuming
3478 anchoring for such patterns in all cases, which is not correct because .* will
3479 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3480 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3481 must be retried after every newline in the subject.
3484 Version 2.04 18-Feb-99
3485 ----------------------
3487 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3488 computation of the store needed to hold the pattern was incorrect (too large).
3489 If such patterns were nested a few deep, this could multiply and become a real
3490 problem.
3492 2. Added /M option to pcretest to show the memory requirement of a specific
3493 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3495 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3496 compiled in such a way that the backtracking after subsequent failure was
3497 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3498 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3501 Version 2.03 02-Feb-99
3502 ----------------------
3504 1. Fixed typo and small mistake in man page.
3506 2. Added 4th condition (GPL supersedes if conflict) and created separate
3507 LICENCE file containing the conditions.
3509 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3510 Perl, that is the internal \ allows the delimiter to be included in the
3511 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3512 the final delimiter, add \ to the end of the pattern (to test the error).
3514 4. Added the convenience functions for extracting substrings after a successful
3515 match. Updated pcretest to make it able to test these functions.
3518 Version 2.02 14-Jan-99
3519 ----------------------
3521 1. Initialized the working variables associated with each extraction so that
3522 their saving and restoring doesn't refer to uninitialized store.
3524 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3525 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3526 fix the problem.
3528 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3529 calls, and wasn't printing the correct value for compiling calls. Increased the
3530 default value of LOOPREPEAT, and the number of significant figures in the
3531 times.
3533 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3535 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3536 a building problem on Windows NT with a FAT file system.
3539 Version 2.01 21-Oct-98
3540 ----------------------
3542 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3543 to character tables built by pcre_maketables() in the current locale. If NULL
3544 is passed, the default tables are used.
3547 Version 2.00 24-Sep-98
3548 ----------------------
3550 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3551 it any more.
3553 2. Allow quantification of (?>) groups, and make it work correctly.
3555 3. The first character computation wasn't working for (?>) groups.
3557 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3558 end of the subject) and add 5.005's \z, which really does match only at the
3559 very end of the subject.
3561 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3563 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3564 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3565 localized options. All options to pcre_study() were also removed.
3567 7. Add other new features from 5.005:
3569 $(?<= positive lookbehind
3570 $(?<! negative lookbehind
3571 (?imsx-imsx) added the unsetting capability
3572 such a setting is global if at outer level; local otherwise
3573 (?imsx-imsx:) non-capturing groups with option setting
3574 (?(cond)re|re) conditional pattern matching
3576 A backreference to itself in a repeated group matches the previous
3577 captured string.
3579 8. General tidying up of studying (both automatic and via "study")
3580 consequential on the addition of new assertions.
3582 9. As in 5.005, unlimited repeated groups that could match an empty substring
3583 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3584 runtime if any iteration does actually match an empty substring.
3586 10. Include the RunTest script in the distribution.
3588 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3589 discrepancies, some of which were old and were also with respect to 5.004. They
3590 have now been fixed.
3593 Version 1.09 28-Apr-98
3594 ----------------------
3596 1. A negated single character class followed by a quantifier with a minimum
3597 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3598 program crashes, or just wrong answers. This did not apply to negated classes
3599 containing more than one character, or to minima other than one.
3602 Version 1.08 27-Mar-98
3603 ----------------------
3605 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3607 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3608 latter must appear before anything that relies on it in the pattern.
3611 Version 1.07 16-Feb-98
3612 ----------------------
3614 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3615 repeat of a potentially empty string).
3618 Version 1.06 23-Jan-98
3619 ----------------------
3621 1. Added Markus Oberhumer's little patches for C++.
3623 2. Literal strings longer than 255 characters were broken.
3626 Version 1.05 23-Dec-97
3627 ----------------------
3629 1. Negated character classes containing more than one character were failing if
3630 PCRE_CASELESS was set at run time.
3633 Version 1.04 19-Dec-97
3634 ----------------------
3636 1. Corrected the man page, where some "const" qualifiers had been omitted.
3638 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3639 input syntax.
3641 3. Fixed memory leak which occurred when a regex with back references was
3642 matched with an offsets vector that wasn't big enough. The temporary memory
3643 that is used in this case wasn't being freed if the match failed.
3645 4. Tidied pcretest to ensure it frees memory that it gets.
3647 5. Temporary memory was being obtained in the case where the passed offsets
3648 vector was exactly big enough.
3650 6. Corrected definition of offsetof() from change 5 below.
3652 7. I had screwed up change 6 below and broken the rules for the use of
3653 setjmp(). Now fixed.
3656 Version 1.03 18-Dec-97
3657 ----------------------
3659 1. A erroneous regex with a missing opening parenthesis was correctly
3660 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3661 on some systems.
3663 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3664 it was reported that one broken compiler failed on the former because "code" is
3665 also an independent variable.
3667 3. The erroneous regex a[]b caused an array overrun reference.
3669 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3670 fail on data ending with that character. (It was going on too far, and checking
3671 the next character, typically a binary zero.) This was specific to the
3672 optimized code for single-character negative classes.
3674 5. Added a contributed patch from the TIN world which does the following:
3676 + Add an undef for memmove, in case the the system defines a macro for it.
3678 + Add a definition of offsetof(), in case there isn't one. (I don't know
3679 the reason behind this - offsetof() is part of the ANSI standard - but
3680 it does no harm).
3682 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3683 most of the places where whitespace preceded '#'. I have given up and
3684 allowed the remaining 2 cases to be at the margin.
3686 + Rename some variables in pcre to eliminate shadowing. This seems very
3687 pedantic, but does no harm, of course.
3689 6. Moved the call to setjmp() into its own function, to get rid of warnings
3690 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3692 7. Constructs such as \d{8,} were compiling into the equivalent of
3693 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3694 outcome, but in this particular case used more store than had been allocated,
3695 which caused the bug to be discovered because it threw up an internal error.
3697 8. The debugging code in both pcre and pcretest for outputting the compiled
3698 form of a regex was going wrong in the case of back references followed by
3699 curly-bracketed repeats.
3702 Version 1.02 12-Dec-97
3703 ----------------------
3705 1. Typos in pcre.3 and comments in the source fixed.
3707 2. Applied a contributed patch to get rid of places where it used to remove
3708 'const' from variables, and fixed some signed/unsigned and uninitialized
3709 variable warnings.
3711 3. Added the "runtest" target to Makefile.
3713 4. Set default compiler flag to -O2 rather than just -O.
3716 Version 1.01 19-Nov-97
3717 ----------------------
3719 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3720 like /([ab]*)*/, that is, for classes with more than one character in them.
3722 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3723 as /((?>a*))*/ (a PCRE_EXTRA facility).
3726 Version 1.00 18-Nov-97
3727 ----------------------
3729 1. Added compile-time macros to support systems such as SunOS4 which don't have
3730 memmove() or strerror() but have other things that can be used instead.
3732 2. Arranged that "make clean" removes the executables.
3735 Version 0.99 27-Oct-97
3736 ----------------------
3738 1. Fixed bug in code for optimizing classes with only one character. It was
3739 initializing a 32-byte map regardless, which could cause it to run off the end
3740 of the memory it had got.
3742 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3745 Version 0.98 22-Oct-97
3746 ----------------------
3748 1. Fixed bug in code for handling temporary memory usage when there are more
3749 back references than supplied space in the ovector. This could cause segfaults.
3752 Version 0.97 21-Oct-97
3753 ----------------------
3755 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3757 2. Optimized negated single characters not to use a bit map.
3759 3. Brought error texts together as macro definitions; clarified some of them;
3760 fixed one that was wrong - it said "range out of order" when it meant "invalid
3761 escape sequence".
3763 4. Changed some char * arguments to const char *.
3765 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3767 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3768 pcretest.
3771 Version 0.96 16-Oct-97
3772 ----------------------
3774 1. Added a simple "pgrep" utility to the distribution.
3776 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3777 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3778 where "ddd" means "one or more decimal digits".
3780 3. Fixed serious bug. If a pattern had a back reference, but the call to
3781 pcre_exec() didn't supply a large enough ovector to record the related
3782 identifying subpattern, the match always failed. PCRE now remembers the number
3783 of the largest back reference, and gets some temporary memory in which to save
3784 the offsets during matching if necessary, in order to ensure that
3785 backreferences always work.
3787 4. Increased the compatibility with Perl in a number of ways:
3789 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3790 to request this handling. The option can be set at compile or exec time.
3792 (b) $ matches before a terminating newline by default; an option
3793 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3794 mode). The option can be set at compile or exec time.
3796 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3797 the same as Perl's. If the decimal number it represents is less than 10
3798 or there aren't that many previous left capturing parentheses, an octal
3799 escape is read. Inside a character class, it's always an octal escape,
3800 even if it is a single digit.
3802 (d) An escaped but undefined alphabetic character is taken as a literal,
3803 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3804 escapes.
3806 (e) {0} is now permitted. (The previous item is removed from the compiled
3807 pattern).
3809 5. Changed all the names of code files so that the basic parts are no longer
3810 than 10 characters, and abolished the teeny "globals.c" file.
3812 6. Changed the handling of character classes; they are now done with a 32-byte
3813 bit map always.
3815 7. Added the -d and /D options to pcretest to make it possible to look at the
3816 internals of compilation without having to recompile pcre.
3819 Version 0.95 23-Sep-97
3820 ----------------------
3822 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3823 \x20 at the start of a run of normal characters. These were being treated as
3824 real characters, instead of the source characters being re-checked.
3827 Version 0.94 18-Sep-97
3828 ----------------------
3830 1. The functions are now thread-safe, with the caveat that the global variables
3831 containing pointers to malloc() and free() or alternative functions are the
3832 same for all threads.
3834 2. Get pcre_study() to generate a bitmap of initial characters for non-
3835 anchored patterns when this is possible, and use it if passed to pcre_exec().
3838 Version 0.93 15-Sep-97
3839 ----------------------
3841 1. /(b)|(:+)/ was computing an incorrect first character.
3843 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3844 but not actually doing anything yet.
3846 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3847 as Perl does (e.g. [-az] or [az-]).
3849 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3850 all possible positions.
3852 5. Split up into different modules to avoid including unneeded functions in a
3853 compiled binary. However, compile and exec are still in one module. The "study"
3854 function is split off.
3856 6. The character tables are now in a separate module whose source is generated
3857 by an auxiliary program - but can then be edited by hand if required. There are
3858 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3859 toupper() in the code.
3861 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3862 make them global. Abolish the function for setting them, as the caller can now
3863 set them directly.
3866 Version 0.92 11-Sep-97
3867 ----------------------
3869 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3870 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3872 2. Caseless matching was not working in character classes if the characters in
3873 the pattern were in upper case.
3875 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3877 4. Make PCRE_ANCHORED public and accept as a compile option.
3879 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3880 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3881 pass them.
3883 6. Give an error if bad option bits passed at compile or run time.
3885 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3886 pcretest to cause it to pass that flag.
3888 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3889 options, and the first character, if set.
3891 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3894 Version 0.91 10-Sep-97
3895 ----------------------
3897 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3898 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3900 2. PCRE was looping on encountering an indefinitely repeated back reference to
3901 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3902 Perl does - treats the match as successful.
3904 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5