ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 483 - (show annotations)
Mon Jan 4 16:19:17 2010 UTC (10 years, 9 months ago) by ph10
File size: 179959 byte(s)
Fix some discrepancies between "USPTR" and "unsigned char *".
1 ChangeLog for PCRE
2 ------------------
4 Version 8.01 11-Dec-09
5 ----------------------
7 1. If a pattern contained a conditional subpattern with only one branch (in
8 particular, this includes all (DEFINE) patterns), a call to pcre_study()
9 computed the wrong minimum data length (which is of course zero for such
10 subpatterns).
12 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
13 the pattern is reset in the first branch, pcre_compile() failed with
14 "internal error: code overflow at offset...". This happened only when
15 the reset was to the original external option setting. (An optimization
16 abstracts leading options settings into an external setting, which was the
17 cause of this.)
19 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
20 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
21 assertion pattern did not match (meaning that the assertion was true), it
22 was incorrectly treated as false if the SKIP had been reached during the
23 matching. This also applied to assertions used as conditions.
25 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
26 assertion subpattern, including such a pattern used as a condition,
27 unpredictable results occurred, instead of the error return
30 5. The C++ GlobalReplace function was not working like Perl for the special
31 situation when an empty string is matched. It now does the fancy magic
32 stuff that is necessary.
34 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
35 removed. (These were left over from very, very early versions of PCRE.)
37 7. Some cosmetic changes to the code to make life easier when compiling it
38 as part of something else:
40 (a) Change DEBUG to PCRE_DEBUG.
42 (b) In pcre_compile(), rename the member of the "branch_chain" structure
43 called "current" as "current_branch", to prevent a collision with the
44 Linux macro when compiled as a kernel module.
46 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
47 prevent a collision with the Linux macro when compiled as a kernel
48 module.
50 8. In pcre_compile() there are some checks for integer overflows that used to
51 cast potentially large values to (double). This has been changed to that
52 when building, a check for int64_t is made, and if it is found, it is used
53 instead, thus avoiding the use of floating point arithmetic. (There is no
54 other use of FP in PCRE.) If int64_t is not found, the fallback is to
55 double.
57 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
58 2005 (difference between two addresses compared to an unsigned value).
60 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
61 custom one, because of the following reported problem in Windows:
63 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
64 under Win32.
65 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
66 therefore missing the function definition.
67 - The compiler thus generates a "C" signature for the test function.
68 - The linker fails to find the "C" function.
69 - PCRE fails to configure if asked to do so against libbz2.
71 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
72 messages were output:
74 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
75 rerunning libtoolize, to keep the correct libtool macros in-tree.
76 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
78 I have done both of these things.
80 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
81 most of the time, it *can* run out if it is given a pattern that contains a
82 runaway infinite recursion. I updated the discussion in the pcrestack man
83 page.
85 13. Now that we have gone to the x.xx style of version numbers, the minor
86 version may start with zero. Using 08 or 09 is a bad idea because users
87 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
88 interpreted as invalid octal numbers. I've updated the previous comment in
89 configure.ac, and also added a check that gives an error if 08 or 09 are
90 used.
92 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
93 causing partial matching to fail when the end of the subject matched \W
94 in a UTF-8 pattern where \W was quantified with a minimum of 3.
96 15. There were some discrepancies between the declarations in pcre_internal.h
97 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
98 their definitions. The declarations used "const uschar *" and the
99 definitions used USPTR. Even though USPTR is normally defined as "const
100 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
101 reported that: "This difference in casting confuses some C++ compilers, for
102 example, SunCC recognizes above declarations as different functions and
103 generates broken code for hbpcre." I have changed the declarations to use
104 USPTR.
109 Version 8.00 19-Oct-09
110 ----------------------
112 1. The table for translating pcre_compile() error codes into POSIX error codes
113 was out-of-date, and there was no check on the pcre_compile() error code
114 being within the table. This could lead to an OK return being given in
115 error.
117 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
118 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
119 in a Windows environment.
121 3. The pcregrep --count option prints the count for each file even when it is
122 zero, as does GNU grep. However, pcregrep was also printing all files when
123 --files-with-matches was added. Now, when both options are given, it prints
124 counts only for those files that have at least one match. (GNU grep just
125 prints the file name in this circumstance, but including the count seems
126 more useful - otherwise, why use --count?) Also ensured that the
127 combination -clh just lists non-zero counts, with no names.
129 4. The long form of the pcregrep -F option was incorrectly implemented as
130 --fixed_strings instead of --fixed-strings. This is an incompatible change,
131 but it seems right to fix it, and I didn't think it was worth preserving
132 the old behaviour.
134 5. The command line items --regex=pattern and --regexp=pattern were not
135 recognized by pcregrep, which required --regex pattern or --regexp pattern
136 (with a space rather than an '='). The man page documented the '=' forms,
137 which are compatible with GNU grep; these now work.
139 6. No libpcreposix.pc file was created for pkg-config; there was just
140 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
142 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
143 when UCP support is not needed, by modifying the Python script that
144 generates it from Unicode data files. This should not matter if the module
145 is correctly used as a library, but I received one complaint about 50K of
146 unwanted data. My guess is that the person linked everything into his
147 program rather than using a library. Anyway, it does no harm.
149 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
150 was a minimum greater than 1 for a wide character in a possessive
151 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
152 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
153 character. Chaos in the form of incorrect output or a compiling loop could
154 result.
156 9. The restrictions on what a pattern can contain when partial matching is
157 requested for pcre_exec() have been removed. All patterns can now be
158 partially matched by this function. In addition, if there are at least two
159 slots in the offset vector, the offset of the earliest inspected character
160 for the match and the offset of the end of the subject are set in them when
161 PCRE_ERROR_PARTIAL is returned.
163 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
164 synonymous with PCRE_PARTIAL, for backwards compatibility, and
165 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
166 and may be more useful for multi-segment matching.
168 11. Partial matching with pcre_exec() is now more intuitive. A partial match
169 used to be given if ever the end of the subject was reached; now it is
170 given only if matching could not proceed because another character was
171 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
172 string "Z", which now yields "no match" instead of "partial match". In the
173 case of pcre_dfa_exec(), "no match" is given if every matching path for the
174 final character ended with (*FAIL).
176 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
177 if the pattern had a "must contain" character that was already found in the
178 earlier partial match, unless partial matching was again requested. For
179 example, with the pattern /dog.(body)?/, the "must contain" character is
180 "g". If the first part-match was for the string "dog", restarting with
181 "sbody" failed. This bug has been fixed.
183 13. The string returned by pcre_dfa_exec() after a partial match has been
184 changed so that it starts at the first inspected character rather than the
185 first character of the match. This makes a difference only if the pattern
186 starts with a lookbehind assertion or \b or \B (\K is not supported by
187 pcre_dfa_exec()). It's an incompatible change, but it makes the two
188 matching functions compatible, and I think it's the right thing to do.
190 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
191 so that the demonstration program is easily available in environments where
192 PCRE has not been installed from source.
194 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
195 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
196 library.
198 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
199 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
200 is not the first non-POSIX option to be added. Clearly some people find
201 these options useful.
203 17. If a caller to the POSIX matching function regexec() passes a non-zero
204 value for nmatch with a NULL value for pmatch, the value of
205 nmatch is forced to zero.
207 18. RunGrepTest did not have a test for the availability of the -u option of
208 the diff command, as RunTest does. It now checks in the same way as
209 RunTest, and also checks for the -b option.
211 19. If an odd number of negated classes containing just a single character
212 interposed, within parentheses, between a forward reference to a named
213 subpattern and the definition of the subpattern, compilation crashed with
214 an internal error, complaining that it could not find the referenced
215 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
216 [The bug was that it was starting one character too far in when skipping
217 over the character class, thus treating the ] as data rather than
218 terminating the class. This meant it could skip too much.]
220 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
221 /g option in pcretest when the pattern contains \K, which makes it possible
222 to have an empty string match not at the start, even when the pattern is
223 anchored. Updated pcretest and pcredemo to use this option.
225 21. If the maximum number of capturing subpatterns in a recursion was greater
226 than the maximum at the outer level, the higher number was returned, but
227 with unset values at the outer level. The correct (outer level) value is
228 now given.
230 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
231 PCRE did not set those parentheses (unlike Perl). I have now found a way to
232 make it do so. The string so far is captured, making this feature
233 compatible with Perl.
235 23. The tests have been re-organized, adding tests 11 and 12, to make it
236 possible to check the Perl 5.10 features against Perl 5.10.
238 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
239 pattern matches a fixed length string. PCRE did not allow this; now it
240 does. Neither allows recursion.
242 25. I finally figured out how to implement a request to provide the minimum
243 length of subject string that was needed in order to match a given pattern.
244 (It was back references and recursion that I had previously got hung up
245 on.) This code has now been added to pcre_study(); it finds a lower bound
246 to the length of subject needed. It is not necessarily the greatest lower
247 bound, but using it to avoid searching strings that are too short does give
248 some useful speed-ups. The value is available to calling programs via
249 pcre_fullinfo().
251 26. While implementing 25, I discovered to my embarrassment that pcretest had
252 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
253 study optimizations had never been tested with that matching function.
254 Oops. What is worse, even when it was passed study data, there was a bug in
255 pcre_dfa_exec() that meant it never actually used it. Double oops. There
256 were also very few tests of studied patterns with pcre_dfa_exec().
258 27. If (?| is used to create subpatterns with duplicate numbers, they are now
259 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
260 on the other side of the coin, they are no longer allowed to have different
261 names, because these cannot be distinguished in PCRE, and this has caused
262 confusion. (This is a difference from Perl.)
264 28. When duplicate subpattern names are present (necessarily with different
265 numbers, as required by 27 above), and a test is made by name in a
266 conditional pattern, either for a subpattern having been matched, or for
267 recursion in such a pattern, all the associated numbered subpatterns are
268 tested, and the overall condition is true if the condition is true for any
269 one of them. This is the way Perl works, and is also more like the way
270 testing by number works.
273 Version 7.9 11-Apr-09
274 ---------------------
276 1. When building with support for bzlib/zlib (pcregrep) and/or readline
277 (pcretest), all targets were linked against these libraries. This included
278 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
279 libraries. This caused unwanted dependencies to be created. This problem
280 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
281 pcretest is linked with readline.
283 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
284 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
285 moved outside it again, because FALSE and TRUE are already defined in AIX,
286 but BOOL is not.
288 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
289 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
291 4. The pcregrep documentation said spaces were inserted as well as colons (or
292 hyphens) following file names and line numbers when outputting matching
293 lines. This is not true; no spaces are inserted. I have also clarified the
294 wording for the --colour (or --color) option.
296 5. In pcregrep, when --colour was used with -o, the list of matching strings
297 was not coloured; this is different to GNU grep, so I have changed it to be
298 the same.
300 6. When --colo(u)r was used in pcregrep, only the first matching substring in
301 each matching line was coloured. Now it goes on to look for further matches
302 of any of the test patterns, which is the same behaviour as GNU grep.
304 7. A pattern that could match an empty string could cause pcregrep to loop; it
305 doesn't make sense to accept an empty string match in pcregrep, so I have
306 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
307 seems to be how GNU grep behaves.
309 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
310 start or after a newline", because the conditional assertion was not being
311 correctly handled. The rule now is that both the assertion and what follows
312 in the first alternative must satisfy the test.
314 9. If auto-callout was enabled in a pattern with a conditional group whose
315 condition was an assertion, PCRE could crash during matching, both with
316 pcre_exec() and pcre_dfa_exec().
318 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
319 used for matching.
321 11. Unicode property support in character classes was not working for
322 characters (bytes) greater than 127 when not in UTF-8 mode.
324 12. Added the -M command line option to pcretest.
326 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
328 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
330 16. Added comments and documentation about mis-use of no_arg in the C++
331 wrapper.
333 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
334 from Martin Jerabek that uses macro names for all relevant character and
335 string constants.
337 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
338 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
339 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
340 these, but not everybody uses configure.
342 19. A conditional group that had only one branch was not being correctly
343 recognized as an item that could match an empty string. This meant that an
344 enclosing group might also not be so recognized, causing infinite looping
345 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
346 with the subject "ab", where knowledge that the repeated group can match
347 nothing is needed in order to break the loop.
349 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
350 exec(), but without supplying a callout function, matching went wrong.
352 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
353 leak if the size of the offset vector was greater than 30. When the vector
354 is smaller, the saved offsets during recursion go onto a local stack
355 vector, but for larger vectors malloc() is used. It was failing to free
356 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
357 error, in fact).
359 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
360 heapframe that is used only when UTF-8 support is enabled. This caused no
361 problem, but was untidy.
363 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
364 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
365 included within another project.
367 24. Steven Van Ingelgem's patches to add more options to the CMake support,
368 slightly modified by me:
370 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
371 not building pcregrep.
373 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
374 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
376 25. Forward references, both numeric and by name, in patterns that made use of
377 duplicate group numbers, could behave incorrectly or give incorrect errors,
378 because when scanning forward to find the reference group, PCRE was not
379 taking into account the duplicate group numbers. A pattern such as
380 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
382 26. Changed a few more instances of "const unsigned char *" to USPTR, making
383 the feature of a custom pointer more persuasive (as requested by a user).
385 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
386 pcretest.c, inside #ifndefs, because it seems they are sometimes already
387 pre-defined.
389 28. Added support for (*UTF8) at the start of a pattern.
391 29. Arrange for flags added by the "release type" setting in CMake to be shown
392 in the configuration summary.
395 Version 7.8 05-Sep-08
396 ---------------------
398 1. Replaced UCP searching code with optimized version as implemented for Ad
399 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
400 stage table and inline lookup instead of a function, giving speed ups of 2
401 to 5 times on some simple patterns that I tested. Permission was given to
402 distribute the MultiStage2.py script that generates the tables (it's not in
403 the tarball, but is in the Subversion repository).
405 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
406 scripts.
408 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
409 a group with a zero qualifier. The result of the study could be incorrect,
410 or the function might crash, depending on the pattern.
412 4. Caseless matching was not working for non-ASCII characters in back
413 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
414 It now works when Unicode Property Support is available.
416 5. In pcretest, an escape such as \x{de} in the data was always generating
417 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
418 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
419 truncation.
421 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
423 7. Added two (int) casts to pcregrep when printing the difference of two
424 pointers, in case they are 64-bit values.
426 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
427 test 2 if it fails.
429 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
430 and a #define of that name to empty if it is not externally set. This is to
431 allow users of MSVC to set it if necessary.
433 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
434 the convenience functions in the pcre_get.c source file.
436 11. An option change at the start of a pattern that had top-level alternatives
437 could cause overwriting and/or a crash. This command provoked a crash in
438 some environments:
440 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
442 This potential security problem was recorded as CVE-2008-2371.
444 12. For a pattern where the match had to start at the beginning or immediately
445 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
446 pcre_dfa_exec() could read past the end of the passed subject if there was
447 no match. To help with detecting such bugs (e.g. with valgrind), I modified
448 pcretest so that it places the subject at the end of its malloc-ed buffer.
450 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
451 exec() might read past the end of the data buffer in UTF-8 mode.
453 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
454 the data contained the byte 0x85 as part of a UTF-8 character within its
455 first line. This applied both to normal and DFA matching.
457 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
458 /^[^d]*?$/8 failed to match "abc".
460 16. Added a missing copyright notice to pcrecpp_internal.h.
462 17. Make it more clear in the documentation that values returned from
463 pcre_exec() in ovector are byte offsets, not character counts.
465 18. Tidied a few places to stop certain compilers from issuing warnings.
467 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
468 supplied by Stefan Weber. I made a further small update for 7.8 because
469 there is a change of source arrangements: the pcre_searchfuncs.c module is
470 replaced by pcre_ucd.c.
473 Version 7.7 07-May-08
474 ---------------------
476 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
477 a string to a long long, pretend we don't even have a long long." This is
478 done by checking for the strtoq, strtoll, and _strtoi64 functions.
480 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
481 pre-7.6 versions, which defined a global no_arg variable instead of putting
482 it in the RE class. (See also #8 below.)
484 3. Remove a line of dead code, identified by coverity and reported by Nuno
485 Lopes.
487 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
489 (1) The include/exclude patterns were being applied to the whole pathnames
490 of files, instead of just to the final components.
492 (2) If there was more than one level of directory, the subdirectories were
493 skipped unless they satisfied the include/exclude conditions. This is
494 inconsistent with GNU grep (and could even be seen as contrary to the
495 pcregrep specification - which I improved to make it absolutely clear).
496 The action now is always to scan all levels of directory, and just
497 apply the include/exclude patterns to regular files.
499 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
500 --exclude_dir in the tests to avoid scanning .svn directories.
502 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
503 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
504 doesn't support NULs in patterns.
506 7. Added some missing "const"s to declarations of static tables in
507 pcre_compile.c and pcre_dfa_exec.c.
509 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
510 caused by fix #2 above. (Subsequently also a second patch to fix the
511 first patch. And a third patch - this was a messy problem.)
513 9. Applied Craig's patch to remove the use of push_back().
515 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
516 matching function regexec().
518 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
519 which, however, unlike Perl's \g{...}, are subroutine calls, not back
520 references. PCRE supports relative numbers with this syntax (I don't think
521 Oniguruma does).
523 12. Previously, a group with a zero repeat such as (...){0} was completely
524 omitted from the compiled regex. However, this means that if the group
525 was called as a subroutine from elsewhere in the pattern, things went wrong
526 (an internal error was given). Such groups are now left in the compiled
527 pattern, with a new opcode that causes them to be skipped at execution
528 time.
530 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
531 to the way PCRE behaves:
533 (a) A lone ] character is dis-allowed (Perl treats it as data).
535 (b) A back reference to an unmatched subpattern matches an empty string
536 (Perl fails the current match path).
538 (c) A data ] in a character class must be notated as \] because if the
539 first data character in a class is ], it defines an empty class. (In
540 Perl it is not possible to have an empty class.) The empty class []
541 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
542 The negative empty class [^] matches any one character, independently
543 of the DOTALL setting.
545 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
546 non-existent subpattern following a character class starting with ']' and
547 containing () gave an internal compiling error instead of "reference to
548 non-existent subpattern". Fortunately, when the pattern did exist, the
549 compiled code was correct. (When scanning forwards to check for the
550 existencd of the subpattern, it was treating the data ']' as terminating
551 the class, so got the count wrong. When actually compiling, the reference
552 was subsequently set up correctly.)
554 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
555 it was being rejected as not supported by pcre_dfa_exec(), even though
556 other assertions are supported. I have made pcre_dfa_exec() support
557 (*FAIL).
559 16. The implementation of 13c above involved the invention of a new opcode,
560 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
561 cannot be changed at match time, I realized I could make a small
562 improvement to matching performance by compiling OP_ALLANY instead of
563 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
564 on the OP_ANY path.
566 17. Compiling pcretest on Windows with readline support failed without the
567 following two fixes: (1) Make the unistd.h include conditional on
568 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
570 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
571 ncurses library to be included for pcretest when ReadLine support is
572 requested, but also to allow for it to be overridden. This patch came from
573 Daniel Bergström.
575 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
576 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
577 any errors with the current Unicode tables. Thanks to Peter Kankowski for
578 spotting this.
581 Version 7.6 28-Jan-08
582 ---------------------
584 1. A character class containing a very large number of characters with
585 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
586 overflow.
588 2. Patch to cut out the "long long" test in pcrecpp_unittest when
589 HAVE_LONG_LONG is not defined.
591 3. Applied Christian Ehrlicher's patch to update the CMake build files to
592 bring them up to date and include new features. This patch includes:
594 - Fixed PH's badly added libz and libbz2 support.
595 - Fixed a problem with static linking.
596 - Added pcredemo. [But later removed - see 7 below.]
597 - Fixed dftables problem and added an option.
598 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
600 - Added readline support for pcretest.
601 - Added an listing of the option settings after cmake has run.
603 4. A user submitted a patch to Makefile that makes it easy to create
604 "pcre.dll" under mingw when using Configure/Make. I added stuff to
605 Makefile.am that cause it to include this special target, without
606 affecting anything else. Note that the same mingw target plus all
607 the other distribution libraries and programs are now supported
608 when configuring with CMake (see 6 below) instead of with
609 Configure/Make.
611 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
612 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
613 exported in the Windows port". It has not yet been confirmed that the patch
614 solves the problem, but it does no harm.
616 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
617 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
618 with CMake, and also correct the comment about stack recursion.
620 7. Remove the automatic building of pcredemo from the ./configure system and
621 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
622 of a program that users should build themselves after PCRE is installed, so
623 building it automatically is not really right. What is more, it gave
624 trouble in some build environments.
626 8. Further tidies to CMakeLists.txt from Sheri and Christian.
629 Version 7.5 10-Jan-08
630 ---------------------
632 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
633 values in parens when parsing an RE using the C++ wrapper."
635 2. Negative specials like \S did not work in character classes in UTF-8 mode.
636 Characters greater than 255 were excluded from the class instead of being
637 included.
639 3. The same bug as (2) above applied to negated POSIX classes such as
640 [:^space:].
642 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
643 defined or documented. It seems to have been a typo for PCRE_STATIC, so
644 I have changed it.
646 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
647 first named subpattern) and a construct such as (?&a) would reference the
648 first named subpattern whose name started with "a" (in other words, the
649 length check was missing). Both these problems are fixed. "Subpattern name
650 expected" is now given for (?&) (a zero-length name), and this patch also
651 makes it give the same error for \k'' (previously it complained that that
652 was a reference to a non-existent subpattern).
654 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
655 this is right because (?- can be followed by option settings as well as by
656 digits. I have, however, made the messages clearer.
658 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
659 than the number used in the conditional) now cause a compile-time error.
660 This is actually not compatible with Perl, which accepts such patterns, but
661 treats the conditional as always being FALSE (as PCRE used to), but it
662 seems to me that giving a diagnostic is better.
664 8. Change "alphameric" to the more common word "alphanumeric" in comments
665 and messages.
667 9. Fix two occurrences of "backslash" in comments that should have been
668 "backspace".
670 10. Remove two redundant lines of code that can never be obeyed (their function
671 was moved elsewhere).
673 11. The program that makes PCRE's Unicode character property table had a bug
674 which caused it to generate incorrect table entries for sequences of
675 characters that have the same character type, but are in different scripts.
676 It amalgamated them into a single range, with the script of the first of
677 them. In other words, some characters were in the wrong script. There were
678 thirteen such cases, affecting characters in the following ranges:
680 U+002b0 - U+002c1
681 U+0060c - U+0060d
682 U+0061e - U+00612
683 U+0064b - U+0065e
684 U+0074d - U+0076d
685 U+01800 - U+01805
686 U+01d00 - U+01d77
687 U+01d9b - U+01dbf
688 U+0200b - U+0200f
689 U+030fc - U+030fe
690 U+03260 - U+0327f
691 U+0fb46 - U+0fbb1
692 U+10450 - U+1049d
694 12. The -o option (show only the matching part of a line) for pcregrep was not
695 compatible with GNU grep in that, if there was more than one match in a
696 line, it showed only the first of them. It now behaves in the same way as
697 GNU grep.
699 13. If the -o and -v options were combined for pcregrep, it printed a blank
700 line for every non-matching line. GNU grep prints nothing, and pcregrep now
701 does the same. The return code can be used to tell if there were any
702 non-matching lines.
704 14. Added --file-offsets and --line-offsets to pcregrep.
706 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
707 infinitely looping recursion. The bug was that positive lookaheads were not
708 being skipped when checking for a possible empty match (negative lookaheads
709 and both kinds of lookbehind were skipped).
711 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
712 inclusion of <windows.h> to before rather than after the definition of
713 INVALID_FILE_ATTRIBUTES (patch from David Byron).
715 17. Specifying a possessive quantifier with a specific limit for a Unicode
716 character property caused pcre_compile() to compile bad code, which led at
717 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
718 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
719 caused the error; without that there was no problem.
721 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
723 19. Added --enable-pcretest-libreadline.
725 20. In pcrecpp.cc, the variable 'count' was incremented twice in
726 RE::GlobalReplace(). As a result, the number of replacements returned was
727 double what it should be. I removed one of the increments, but Craig sent a
728 later patch that removed the other one (the right fix) and added unit tests
729 that check the return values (which was not done before).
731 21. Several CMake things:
733 (1) Arranged that, when cmake is used on Unix, the libraries end up with
734 the names libpcre and libpcreposix, not just pcre and pcreposix.
736 (2) The above change means that pcretest and pcregrep are now correctly
737 linked with the newly-built libraries, not previously installed ones.
741 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
742 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
743 UTF-8 newline character). The key issue is that the pattern starts .*;
744 this means that the match must be either at the beginning, or after a
745 newline. The bug was in the code for advancing after a failed match and
746 checking that the new position followed a newline. It was not taking
747 account of UTF-8 characters correctly.
749 23. PCRE was behaving differently from Perl in the way it recognized POSIX
750 character classes. PCRE was not treating the sequence [:...:] as a
751 character class unless the ... were all letters. Perl, however, seems to
752 allow any characters between [: and :], though of course it rejects as
753 unknown any "names" that contain non-letters, because all the known class
754 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
755 for example, whereas PCRE did not - it did not recognize a POSIX character
756 class. This seemed a bit dangerous, so the code has been changed to be
757 closer to Perl. The behaviour is not identical to Perl, because PCRE will
758 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
759 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
760 Perl does, and where it didn't before.
762 24. Rewrite so as to remove the single use of %n from pcregrep because in some
763 Windows environments %n is disabled by default.
766 Version 7.4 21-Sep-07
767 ---------------------
769 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
770 means that a class such as [\s] counted as "explicit reference to CR or
771 LF". That isn't really right - the whole point of the change was to try to
772 help when there was an actual mention of one of the two characters. So now
773 the change happens only if \r or \n (or a literal CR or LF) character is
774 encountered.
776 2. The 32-bit options word was also used for 6 internal flags, but the numbers
777 of both had grown to the point where there were only 3 bits left.
778 Fortunately, there was spare space in the data structure, and so I have
779 moved the internal flags into a new 16-bit field to free up more option
780 bits.
782 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
783 but did not set the internal JCHANGED flag - either of these is enough to
784 control the way the "get" function works - but the PCRE_INFO_JCHANGED
785 facility is supposed to tell if (?J) was ever used, so now (?J) at the
786 start sets both bits.
788 4. Added options (at build time, compile time, exec time) to change \R from
789 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
791 5. doc/pcresyntax.html was missing from the distribution.
793 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
794 compatibility, even though it is no longer used.
796 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
797 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
798 windows.h file is present (where different names are used). [This was
799 reversed later after testing - see 16 below.]
801 8. Changed all #include <config.h> to #include "config.h". There were also
802 some further <pcre.h> cases that I changed to "pcre.h".
804 9. When pcregrep was used with the --colour option, it missed the line ending
805 sequence off the lines that it output.
807 10. It was pointed out to me that arrays of string pointers cause lots of
808 relocations when a shared library is dynamically loaded. A technique of
809 using a single long string with a table of offsets can drastically reduce
810 these. I have refactored PCRE in four places to do this. The result is
811 dramatic:
813 Originally: 290
814 After changing UCP table: 187
815 After changing error message table: 43
816 After changing table of "verbs" 36
817 After changing table of Posix names 22
819 Thanks to the folks working on Gregex for glib for this insight.
821 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
822 unicode-properties was also set.
824 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
826 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
827 checked only for CRLF.
829 14. Added casts to pcretest.c to avoid compiler warnings.
831 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
833 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
834 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
835 entirely. This removes changes made in 7 above.
837 17. The CMake files have been updated, and there is now more information about
838 building with CMake in the NON-UNIX-USE document.
841 Version 7.3 28-Aug-07
842 ---------------------
844 1. In the rejigging of the build system that eventually resulted in 7.1, the
845 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
846 brackets there is not right, since it causes compilers to look for an
847 installed pcre.h, not the version that is in the source that is being
848 compiled (which of course may be different). I have changed it back to:
850 #include "pcre.h"
852 I have a vague recollection that the change was concerned with compiling in
853 different directories, but in the new build system, that is taken care of
854 by the VPATH setting the Makefile.
856 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
857 when the subject happened to end in the byte 0x85 (e.g. if the last
858 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
859 characters but of course it shouldn't be taken as a newline when it is part
860 of another character. The bug was that, for an unlimited repeat of . in
861 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
862 characters when looking for a newline.
864 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
866 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
867 in debug output.
869 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
870 long printing in the pcrecpp unittest when running under MinGW.
872 6. ESC_K was left out of the EBCDIC table.
874 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
875 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
876 limit also applies to "virtual nesting" when a pattern is recursive, and in
877 this case 1000 isn't so big. I have been able to remove this limit at the
878 expense of backing off one optimization in certain circumstances. Normally,
879 when pcre_exec() would call its internal match() function recursively and
880 immediately return the result unconditionally, it uses a "tail recursion"
881 feature to save stack. However, when a subpattern that can match an empty
882 string has an unlimited repetition quantifier, it no longer makes this
883 optimization. That gives it a stack frame in which to save the data for
884 checking that an empty string has been matched. Previously this was taken
885 from the 1000-entry workspace that had been reserved. So now there is no
886 explicit limit, but more stack is used.
888 8. Applied Daniel's patches to solve problems with the import/export magic
889 syntax that is required for Windows, and which was going wrong for the
890 pcreposix and pcrecpp parts of the library. These were overlooked when this
891 problem was solved for the main library.
893 9. There were some crude static tests to avoid integer overflow when computing
894 the size of patterns that contain repeated groups with explicit upper
895 limits. As the maximum quantifier is 65535, the maximum group length was
896 set at 30,000 so that the product of these two numbers did not overflow a
897 32-bit integer. However, it turns out that people want to use groups that
898 are longer than 30,000 bytes (though not repeat them that many times).
899 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
900 made it possible to implement the integer overflow checks in a much more
901 dynamic way, which I have now done. The artificial limitation on group
902 length has been removed - we now have only the limit on the total length of
903 the compiled pattern, which depends on the LINK_SIZE setting.
905 10. Fixed a bug in the documentation for get/copy named substring when
906 duplicate names are permitted. If none of the named substrings are set, the
907 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
908 empty string.
910 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
911 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
912 because the ] is interpreted as the first data character and the
913 terminating ] is not found. PCRE has been made compatible with Perl in this
914 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
915 cause memory overwriting.
917 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
918 string has been matched (to stop an infinite loop). It was not recognizing
919 a conditional subpattern that could match an empty string if that
920 subpattern was within another subpattern. For example, it looped when
921 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
922 condition was not nested. This bug has been fixed.
924 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
925 past the start of the subject in the presence of bytes with the top bit
926 set, for example "\x8aBCD".
928 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
929 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
931 14. Optimized (?!) to (*FAIL).
933 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
934 This restricts code points to be within the range 0 to 0x10FFFF, excluding
935 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
936 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
937 does: it's just the validity check that is more restrictive.
939 16. Inserted checks for integer overflows during escape sequence (backslash)
940 processing, and also fixed erroneous offset values for syntax errors during
941 backslash processing.
943 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
944 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
946 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
947 caused an overrun.
949 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
950 something other than just ASCII characters) inside a group that had an
951 unlimited repeat caused a loop at compile time (while checking to see
952 whether the group could match an empty string).
954 20. Debugging a pattern containing \p or \P could cause a crash. For example,
955 [\P{Any}] did so. (Error in the code for printing property names.)
957 21. An orphan \E inside a character class could cause a crash.
959 22. A repeated capturing bracket such as (A)? could cause a wild memory
960 reference during compilation.
962 23. There are several functions in pcre_compile() that scan along a compiled
963 expression for various reasons (e.g. to see if it's fixed length for look
964 behind). There were bugs in these functions when a repeated \p or \P was
965 present in the pattern. These operators have additional parameters compared
966 with \d, etc, and these were not being taken into account when moving along
967 the compiled data. Specifically:
969 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
970 length.
972 (b) An item such as \pL+ within a repeated group could cause crashes or
973 loops.
975 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
976 "reference to non-existent subpattern" error.
978 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
980 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
981 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
983 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
985 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
986 character were causing crashes (broken optimization).
988 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
989 \p or \P) caused a compile-time loop.
991 28. More problems have arisen in unanchored patterns when CRLF is a valid line
992 break. For example, the unstudied pattern [\r\n]A does not match the string
993 "\r\nA" because change 7.0/46 below moves the current point on by two
994 characters after failing to match at the start. However, the pattern \nA
995 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
996 the same is true. There doesn't seem any very clean way out of this, but
997 what I have chosen to do makes the common cases work: PCRE now takes note
998 of whether there can be an explicit match for \r or \n anywhere in the
999 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1000 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1001 pattern has explicit CR or LF references.
1003 29. Added (*CR) etc for changing newline setting at start of pattern.
1006 Version 7.2 19-Jun-07
1007 ---------------------
1009 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1010 which is apparently normally available under Windows.
1012 2. Re-jig the pcregrep tests with different newline settings in an attempt
1013 to make them independent of the local environment's newline setting.
1015 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1017 4. Some of the "internals" tests were previously cut out when the link size
1018 was not 2, because the output contained actual offsets. The recent new
1019 "Z" feature of pcretest means that these can be cut out, making the tests
1020 usable with all link sizes.
1022 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1023 stack recursion. This gives a massive performance boost under BSD, but just
1024 a small improvement under Linux. However, it saves one field in the frame
1025 in all cases.
1027 6. Added more features from the forthcoming Perl 5.10:
1029 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1030 recursion call. It refers to the nth most recently opened parentheses.
1032 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1033 to be opened parentheses.
1035 (c) Conditions that refer to capturing parentheses can be specified
1036 relatively, for example, (?(-2)... or (?(+3)...
1038 (d) \K resets the start of the current match so that everything before
1039 is not part of it.
1041 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1043 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1044 reference syntax.
1046 (g) (?| introduces a group in which the numbering of parentheses in each
1047 alternative starts with the same number.
1049 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1051 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1054 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1055 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1056 for detecting groups that can match an empty string.
1058 9. A pattern with a very large number of alternatives (more than several
1059 hundred) was running out of internal workspace during the pre-compile
1060 phase, where pcre_compile() figures out how much memory will be needed. A
1061 bit of new cunning has reduced the workspace needed for groups with
1062 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1063 workspace instead of running out of the 4096 that are available.
1065 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1067 11. Applied patch from Google to remove an optimization that didn't quite work.
1068 The report of the bug said:
1070 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1071 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1072 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1074 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1075 it matched the wrong number of bytes.
1078 Version 7.1 24-Apr-07
1079 ---------------------
1081 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1082 that is more "standard", making use of automake and other Autotools. There
1083 is some re-arrangement of the files and adjustment of comments consequent
1084 on this.
1086 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1087 for recursive directory scanning broke on some systems because the files
1088 are not scanned in any specific order and on different systems the order
1089 was different. A call to "sort" has been inserted into RunGrepTest for the
1090 approprate test as a short-term fix. In the longer term there may be an
1091 alternative.
1093 3. I had an email from Eric Raymond about problems translating some of PCRE's
1094 man pages to HTML (despite the fact that I distribute HTML pages, some
1095 people do their own conversions for various reasons). The problems
1096 concerned the use of low-level troff macros .br and .in. I have therefore
1097 removed all such uses from the man pages (some were redundant, some could
1098 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1099 HTML has been updated to handle .nf/.fi and to complain if it encounters
1100 .br or .in.
1102 4. Updated comments in configure.ac that get placed in config.h.in and also
1103 arranged for config.h to be included in the distribution, with the name
1104 config.h.generic, for the benefit of those who have to compile without
1105 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1107 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1108 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1109 makevp.bat for the current PCRE, using the additional files
1110 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1112 6. A Windows user reported a minor discrepancy with test 2, which turned out
1113 to be caused by a trailing space on an input line that had got lost in his
1114 copy. The trailing space was an accident, so I've just removed it.
1116 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1117 that is needed.
1119 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1120 as "const" (a) because they are and (b) because it helps the PHP
1121 maintainers who have recently made a script to detect big data structures
1122 in the php code that should be moved to the .rodata section. I remembered
1123 to update Builducptable as well, so it won't revert if ucptable.h is ever
1124 re-created.
1126 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1127 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1128 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1129 support is not required. This saves 1.5-2K of code, which is important in
1130 some applications.
1132 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1133 so as not to refer to the tables, even though these functions will never be
1134 called when UTF-8 support is disabled. Otherwise there are problems with a
1135 shared library.
1137 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1139 (a) It was defining its arguments as char * instead of void *.
1141 (b) It was assuming that all moves were upwards in memory; this was true
1142 a long time ago when I wrote it, but is no longer the case.
1144 The emulated memove() is provided for those environments that have neither
1145 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1146 is clearly not the case, as these two bugs were recently reported.
1148 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1149 and Detrail to create the HTML documentation, the .txt form of the man
1150 pages, and it removes trailing spaces from listed files. It also creates
1151 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1152 case, it wraps all the #defines with #ifndefs. This script should be run
1153 before "make dist".
1155 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1156 with Unicode property support.
1158 (a) For a maximizing quantifier, if the two different cases of the
1159 character were of different lengths in their UTF-8 codings (there are
1160 some cases like this - I found 11), and the matching function had to
1161 back up over a mixture of the two cases, it incorrectly assumed they
1162 were both the same length.
1164 (b) When PCRE was configured to use the heap rather than the stack for
1165 recursion during matching, it was not correctly preserving the data for
1166 the other case of a UTF-8 character when checking ahead for a match
1167 while processing a minimizing repeat. If the check also involved
1168 matching a wide character, but failed, corruption could cause an
1169 erroneous result when trying to check for a repeat of the original
1170 character.
1172 13. Some tidying changes to the testing mechanism:
1174 (a) The RunTest script now detects the internal link size and whether there
1175 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1176 values substituted by "configure". (The RunGrepTest script already did
1177 this for UTF-8.) The configure.ac script no longer substitutes the
1178 relevant variables.
1180 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1181 with length and offset values. This means that the output is different
1182 for different internal link sizes. Test 2 is skipped for link sizes
1183 other than 2 because of this, bypassing the problem. Unfortunately,
1184 there was also a test in test 3 (the locale tests) that used /B and
1185 failed for link sizes other than 2. Rather than cut the whole test out,
1186 I have added a new /Z option to pcretest that replaces the length and
1187 offset values with spaces. This is now used to make test 3 independent
1188 of link size. (Test 2 will be tidied up later.)
1190 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1191 segmentation fault instead of returning the appropriate error message.
1193 15. In multiline mode when the newline sequence was set to "any", the pattern
1194 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1195 This doesn't seem right; it now treats the CRLF combination as the line
1196 ending, and so does not match in that case. It's only a pattern such as ^$
1197 that would hit this one: something like ^ABC$ would have failed after \r
1198 and then tried again after \r\n.
1200 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1201 in an attempt to make files that differ only in their line terminators
1202 compare equal. This works on Linux.
1204 17. Under certain error circumstances pcregrep might try to free random memory
1205 as it exited. This is now fixed, thanks to valgrind.
1207 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1208 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1209 was because its rules for how to advance for /g after matching an empty
1210 string at the end of a line did not allow for this case. They now check for
1211 it specially.
1213 20. pcretest is supposed to handle patterns and data of any length, by
1214 extending its buffers when necessary. It was getting this wrong when the
1215 buffer for a data line had to be extended.
1217 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1218 CRLF as a newline sequence.
1220 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1221 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1222 I have nevertheless tidied it up.
1224 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1226 24. Added a man page for pcre-config.
1229 Version 7.0 19-Dec-06
1230 ---------------------
1232 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1233 moving to gcc 4.1.1.
1235 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1236 sys/time.h, which is documented as needed for this function. It doesn't
1237 seem to matter on Linux, but it showed up on some releases of OS X.
1239 3. It seems that there are systems where bytes whose values are greater than
1240 127 match isprint() in the "C" locale. The "C" locale should be the
1241 default when a C program starts up. In most systems, only ASCII printing
1242 characters match isprint(). This difference caused the output from pcretest
1243 to vary, making some of the tests fail. I have changed pcretest so that:
1245 (a) When it is outputting text in the compiled version of a pattern, bytes
1246 other than 32-126 are always shown as hex escapes.
1248 (b) When it is outputting text that is a matched part of a subject string,
1249 it does the same, unless a different locale has been set for the match
1250 (using the /L modifier). In this case, it uses isprint() to decide.
1252 4. Fixed a major bug that caused incorrect computation of the amount of memory
1253 required for a compiled pattern when options that changed within the
1254 pattern affected the logic of the preliminary scan that determines the
1255 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1256 that the computed length was too small. The symptoms of this bug were
1257 either the PCRE error "internal error: code overflow" from pcre_compile(),
1258 or a glibc crash with a message such as "pcretest: free(): invalid next
1259 size (fast)". Examples of patterns that provoked this bug (shown in
1260 pcretest format) are:
1262 /(?-x: )/x
1263 /(?x)(?-x: \s*#\s*)/
1264 /((?i)[\x{c0}])/8
1265 /(?i:[\x{c0}])/8
1267 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1268 is now done differently.
1270 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1271 wrapper classes; (b) implement a new function in the C++ scanner that is
1272 more efficient than the old way of doing things because it avoids levels of
1273 recursion in the regex matching; (c) add a paragraph to the documentation
1274 for the FullMatch() function.
1276 6. The escape sequence \n was being treated as whatever was defined as
1277 "newline". Not only was this contrary to the documentation, which states
1278 that \n is character 10 (hex 0A), but it also went horribly wrong when
1279 "newline" was defined as CRLF. This has been fixed.
1281 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1282 was being set to -1 for the "end of line" case (supposedly a value that no
1283 character can have). Though this value is never used (the check for end of
1284 line is "zero bytes in current character"), it caused compiler complaints.
1285 I've changed it to 0xffffffff.
1287 8. In pcre_version.c, the version string was being built by a sequence of
1288 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1289 string (as it is for production releases) called a macro with an empty
1290 argument. The C standard says the result of this is undefined. The gcc
1291 compiler treats it as an empty string (which was what was wanted) but it is
1292 reported that Visual C gives an error. The source has been hacked around to
1293 avoid this problem.
1295 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1296 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1297 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1298 of them did).
1300 10. Originally, pcretest opened its input and output without "b"; then I was
1301 told that "b" was needed in some environments, so it was added for release
1302 5.0 to both the input and output. (It makes no difference on Unix-like
1303 systems.) Later I was told that it is wrong for the input on Windows. I've
1304 now abstracted the modes into two macros, to make it easier to fiddle with
1305 them, and removed "b" from the input mode under Windows.
1307 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1309 12. Added -help and --help to pcretest as an official way of being reminded
1310 of the options.
1312 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1313 and pcrecpp.cc because they annoy compilers at high warning levels.
1315 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1317 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1318 scripts are not C programs :-) and which was not noticed because it works
1319 on Linux.
1321 16. pcretest is supposed to handle any length of pattern and data line (as one
1322 line or as a continued sequence of lines) by extending its input buffer if
1323 necessary. This feature was broken for very long pattern lines, leading to
1324 a string of junk being passed to pcre_compile() if the pattern was longer
1325 than about 50K.
1327 17. I have done a major re-factoring of the way pcre_compile() computes the
1328 amount of memory needed for a compiled pattern. Previously, there was code
1329 that made a preliminary scan of the pattern in order to do this. That was
1330 OK when PCRE was new, but as the facilities have expanded, it has become
1331 harder and harder to keep it in step with the real compile phase, and there
1332 have been a number of bugs (see for example, 4 above). I have now found a
1333 cunning way of running the real compile function in a "fake" mode that
1334 enables it to compute how much memory it would need, while actually only
1335 ever using a few hundred bytes of working memory and without too many
1336 tests of the mode. This should make future maintenance and development
1337 easier. A side effect of this work is that the limit of 200 on the nesting
1338 depth of parentheses has been removed (though this was never a serious
1339 limitation, I suspect). However, there is a downside: pcre_compile() now
1340 runs more slowly than before (30% or more, depending on the pattern). I
1341 hope this isn't a big issue. There is no effect on runtime performance.
1343 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1344 newline (only possible for the last line of a file) and it was a
1345 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1347 19. Added additional timing features to pcretest. (1) The -tm option now times
1348 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1349 separate command line item, by a number that specifies the number of
1350 repeats to use when timing. The default is 50000; this gives better
1351 precision, but takes uncomfortably long for very large patterns.
1353 20. Extended pcre_study() to be more clever in cases where a branch of a
1354 subpattern has no definite first character. For example, (a*|b*)[cd] would
1355 previously give no result from pcre_study(). Now it recognizes that the
1356 first character must be a, b, c, or d.
1358 21. There was an incorrect error "recursive call could loop indefinitely" if
1359 a subpattern (or the entire pattern) that was being tested for matching an
1360 empty string contained only one non-empty item after a nested subpattern.
1361 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1362 incorrectly, because the \d was being skipped in the check.
1364 22. The pcretest program now has a new pattern option /B and a command line
1365 option -b, which is equivalent to adding /B to every pattern. This causes
1366 it to show the compiled bytecode, without the additional information that
1367 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1368 is the same as /B/I).
1370 23. A new optimization is now able automatically to treat some sequences such
1371 as a*b as a*+b. More specifically, if something simple (such as a character
1372 or a simple class like \d) has an unlimited quantifier, and is followed by
1373 something that cannot possibly match the quantified thing, the quantifier
1374 is automatically "possessified".
1376 24. A recursive reference to a subpattern whose number was greater than 39
1377 went wrong under certain circumstances in UTF-8 mode. This bug could also
1378 have affected the operation of pcre_study().
1380 25. Realized that a little bit of performance could be had by replacing
1381 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1383 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1385 27. Possessive quantifiers such as a++ were previously implemented by turning
1386 them into atomic groups such as ($>a+). Now they have their own opcodes,
1387 which improves performance. This includes the automatically created ones
1388 from 23 above.
1390 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1391 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1392 the first matched character to be a colon. This applied both to named and
1393 numbered groups.
1395 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1397 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1398 building PCRE on Windows easier, so I have included it in the distribution.
1400 31. There is now a check in pcretest against a ridiculously large number being
1401 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1402 loop, the loop is abandoned.
1404 32. Forward references to subpatterns in conditions such as (?(2)...) where
1405 subpattern 2 is defined later cause pcre_compile() to search forwards in
1406 the pattern for the relevant set of parentheses. This search went wrong
1407 when there were unescaped parentheses in a character class, parentheses
1408 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1410 33. "Subroutine" calls and backreferences were previously restricted to
1411 referencing subpatterns earlier in the regex. This restriction has now
1412 been removed.
1414 34. Added a number of extra features that are going to be in Perl 5.10. On the
1415 whole, these are just syntactic alternatives for features that PCRE had
1416 previously implemented using the Python syntax or my own invention. The
1417 other formats are all retained for compatibility.
1419 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1420 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1421 also .NET compatible.
1423 (b) A recursion or subroutine call to a named group can now be defined as
1424 (?&name) as well as (?P>name).
1426 (c) A backreference to a named group can now be defined as \k<name> or
1427 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1428 5.10, are also .NET compatible.
1430 (d) A conditional reference to a named group can now use the syntax
1431 (?(<name>) or (?('name') as well as (?(name).
1433 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1434 groups (named and numbered) that are never evaluated inline, but can be
1435 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1436 is always false. There may be only one alternative in such a group.
1438 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1439 as the simple (?(R). The condition is true only if the most recent
1440 recursion is that of the given number or name. It does not search out
1441 through the entire recursion stack.
1443 (g) The escape \gN or \g{N} has been added, where N is a positive or
1444 negative number, specifying an absolute or relative reference.
1446 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1447 some "unreachable code" warnings.
1449 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1450 things, this adds five new scripts.
1452 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1453 There were also incompatibilities regarding the handling of \Q..\E inside
1454 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1455 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1457 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1458 matches an empty string, and forcibly breaks the loop. There were bugs in
1459 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1460 against aaaa the result was just "a" rather than "aaaa", for example. Two
1461 separate and independent bugs (that affected different cases) have been
1462 fixed.
1464 39. Refactored the code to abolish the use of different opcodes for small
1465 capturing bracket numbers. This is a tidy that I avoided doing when I
1466 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1467 The new approach is not only tidier, it makes it possible to reduce the
1468 memory needed to fix the previous bug (38).
1470 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1471 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1472 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1473 mode.
1475 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1476 report.
1478 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1479 copy construction and assignment in the C++ wrapper.
1481 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1482 couple of bugs that could have given wrong results in the "--newline=crlf"
1483 case.
1485 44. Added a number of casts and did some reorganization of signed/unsigned int
1486 variables following suggestions from Dair Grant. Also renamed the variable
1487 "this" as "item" because it is a C++ keyword.
1489 45. Arranged for dftables to add
1491 #include "pcre_internal.h"
1493 to pcre_chartables.c because without it, gcc 4.x may remove the array
1494 definition from the final binary if PCRE is built into a static library and
1495 dead code stripping is activated.
1497 46. For an unanchored pattern, if a match attempt fails at the start of a
1498 newline sequence, and the newline setting is CRLF or ANY, and the next two
1499 characters are CRLF, advance by two characters instead of one.
1502 Version 6.7 04-Jul-06
1503 ---------------------
1505 1. In order to handle tests when input lines are enormously long, pcretest has
1506 been re-factored so that it automatically extends its buffers when
1507 necessary. The code is crude, but this _is_ just a test program. The
1508 default size has been increased from 32K to 50K.
1510 2. The code in pcre_study() was using the value of the re argument before
1511 testing it for NULL. (Of course, in any sensible call of the function, it
1512 won't be NULL.)
1514 3. The memmove() emulation function in pcre_internal.h, which is used on
1515 systems that lack both memmove() and bcopy() - that is, hardly ever -
1516 was missing a "static" storage class specifier.
1518 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1519 containing an extended class (one that cannot be represented by a bitmap
1520 because it contains high-valued characters or Unicode property items, e.g.
1521 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1522 pattern, but PCRE should not loop if you do not (it no longer does).
1523 [Detail: two cases were found: (a) a repeated subpattern containing an
1524 extended class; (b) a recursive reference to a subpattern that followed a
1525 previous extended class. It wasn't skipping over the extended class
1526 correctly when UTF-8 mode was not set.]
1528 5. A negated single-character class was not being recognized as fixed-length
1529 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1530 compile error "lookbehind assertion is not fixed length".
1532 6. The RunPerlTest auxiliary script was showing an unexpected difference
1533 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1534 write a Perl script that can interpret lines of an input file either as
1535 byte characters or as UTF-8, which is what "perltest" was being required to
1536 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1537 can't do is switch easily at run time between having the "use utf8;" pragma
1538 or not. In the end, I fudged it by using the RunPerlTest script to insert
1539 "use utf8;" explicitly for the UTF-8 tests.
1541 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1542 the end of the subject string, contrary to the documentation and to what
1543 Perl does. This was true of both matching functions. Now it matches only at
1544 the start of the subject and immediately after *internal* newlines.
1546 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1547 a pointer to an int instead of a pointer to an unsigned long int. This
1548 caused problems on 64-bit systems.
1550 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1551 instance of the 'standard' template library not being so standard".
1553 10. There was no check on the number of named subpatterns nor the maximum
1554 length of a subpattern name. The product of these values is used to compute
1555 the size of the memory block for a compiled pattern. By supplying a very
1556 long subpattern name and a large number of named subpatterns, the size
1557 computation could be caused to overflow. This is now prevented by limiting
1558 the length of names to 32 characters, and the number of named subpatterns
1559 to 10,000.
1561 11. Subpatterns that are repeated with specific counts have to be replicated in
1562 the compiled pattern. The size of memory for this was computed from the
1563 length of the subpattern and the repeat count. The latter is limited to
1564 65535, but there was no limit on the former, meaning that integer overflow
1565 could in principle occur. The compiled length of a repeated subpattern is
1566 now limited to 30,000 bytes in order to prevent this.
1568 12. Added the optional facility to have named substrings with the same name.
1570 13. Added the ability to use a named substring as a condition, using the
1571 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1572 are numbers (not recommended). Forward references are permitted.
1574 14. Added forward references in named backreferences (if you see what I mean).
1576 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1577 pattern could run off the end of the subject. For example, the pattern
1578 "(?s)(.{1,5})"8 did this with the subject "ab".
1580 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1581 PCRE_CASELESS was set when matching characters that were quantified with ?
1582 or *.
1584 17. A character class other than a single negated character that had a minimum
1585 but no maximum quantifier - for example [ab]{6,} - was not handled
1586 correctly by pce_dfa_exec(). It would match only one character.
1588 18. A valid (though odd) pattern that looked like a POSIX character
1589 class but used an invalid character after [ (for example [[,abc,]]) caused
1590 pcre_compile() to give the error "Failed: internal error: code overflow" or
1591 in some cases to crash with a glibc free() error. This could even happen if
1592 the pattern terminated after [[ but there just happened to be a sequence of
1593 letters, a binary zero, and a closing ] in the memory that followed.
1595 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1596 over the years. Originally (before any Unicode support), just the bottom 8
1597 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1598 output from "man perlunicode" includes this:
1600 The regular expression compiler produces polymorphic opcodes. That
1601 is, the pattern adapts to the data and automatically switches to
1602 the Unicode character scheme when presented with Unicode data--or
1603 instead uses a traditional byte scheme when presented with byte
1604 data.
1606 Sadly, a wide octal escape does not cause a switch, and in a string with
1607 no other multibyte characters, these octal escapes are treated as before.
1608 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1609 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1610 Unicode string.
1612 I have not perpetrated such confusion in PCRE. Up till now, it took just
1613 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1614 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1615 translate to the appropriate multibyte character.
1617 29. Applied some refactoring to reduce the number of warnings from Microsoft
1618 and Borland compilers. This has included removing the fudge introduced
1619 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1620 a warning about an unused variable.
1622 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1623 characters since release 4.0, because Perl (from release 5.004) does not.
1624 [Or at least, is documented not to: some releases seem to be in conflict
1625 with the documentation.] However, when a pattern was studied with
1626 pcre_study() and all its branches started with \s, PCRE still included VT
1627 as a possible starting character. Of course, this did no harm; it just
1628 caused an unnecessary match attempt.
1630 22. Removed a now-redundant internal flag bit that recorded the fact that case
1631 dependency changed within the pattern. This was once needed for "required
1632 byte" processing, but is no longer used. This recovers a now-scarce options
1633 bit. Also moved the least significant internal flag bit to the most-
1634 significant bit of the word, which was not previously used (hangover from
1635 the days when it was an int rather than a uint) to free up another bit for
1636 the future.
1638 23. Added support for CRLF line endings as well as CR and LF. As well as the
1639 default being selectable at build time, it can now be changed at runtime
1640 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1641 specify that it is scanning data with non-default line endings.
1643 24. Changed the definition of CXXLINK to make it agree with the definition of
1644 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1646 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1647 recursions. This makes a big different to stack usage for some patterns.
1649 26. If a subpattern containing a named recursion or subroutine reference such
1650 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1651 the space required for the compiled pattern went wrong and gave too small a
1652 value. Depending on the environment, this could lead to "Failed: internal
1653 error: code overflow at offset 49" or "glibc detected double free or
1654 corruption" errors.
1656 27. Applied patches from Google (a) to support the new newline modes and (b) to
1657 advance over multibyte UTF-8 characters in GlobalReplace.
1659 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1660 difference for some implementation of PCRE in some Windows version.
1662 29. Added some extra testing facilities to pcretest:
1664 \q<number> in a data line sets the "match limit" value
1665 \Q<number> in a data line sets the "match recursion limt" value
1666 -S <number> sets the stack size, where <number> is in megabytes
1668 The -S option isn't available for Windows.
1671 Version 6.6 06-Feb-06
1672 ---------------------
1674 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1675 in pcreposix.h. I have copied the definition from pcre.h.
1677 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1678 because pcre.h is no longer a built file.
1680 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1681 not normally included in the compiled code.
1684 Version 6.5 01-Feb-06
1685 ---------------------
1687 1. When using the partial match feature with pcre_dfa_exec(), it was not
1688 anchoring the second and subsequent partial matches at the new starting
1689 point. This could lead to incorrect results. For example, with the pattern
1690 /1234/, partially matching against "123" and then "a4" gave a match.
1692 2. Changes to pcregrep:
1694 (a) All non-match returns from pcre_exec() were being treated as failures
1695 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1696 error message is output. Some extra information is given for the
1698 probably the only errors that are likely to be caused by users (by
1699 specifying a regex that has nested indefinite repeats, for instance).
1700 If there are more than 20 of these errors, pcregrep is abandoned.
1702 (b) A binary zero was treated as data while matching, but terminated the
1703 output line if it was written out. This has been fixed: binary zeroes
1704 are now no different to any other data bytes.
1706 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1707 used to set a locale for matching. The --locale=xxxx long option has
1708 been added (no short equivalent) to specify a locale explicitly on the
1709 pcregrep command, overriding the environment variables.
1711 (d) When -B was used with -n, some line numbers in the output were one less
1712 than they should have been.
1714 (e) Added the -o (--only-matching) option.
1716 (f) If -A or -C was used with -c (count only), some lines of context were
1717 accidentally printed for the final match.
1719 (g) Added the -H (--with-filename) option.
1721 (h) The combination of options -rh failed to suppress file names for files
1722 that were found from directory arguments.
1724 (i) Added the -D (--devices) and -d (--directories) options.
1726 (j) Added the -F (--fixed-strings) option.
1728 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1730 (l) Added the --colo(u)r option.
1732 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1733 is not present by default.
1735 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1736 items such as (?R) or (?1), when the recursion could match a number of
1737 alternatives. If it matched one of the alternatives, but subsequently,
1738 outside the recursion, there was a failure, the code tried to back up into
1739 the recursion. However, because of the way PCRE is implemented, this is not
1740 possible, and the result was an incorrect result from the match.
1742 In order to prevent this happening, the specification of recursion has
1743 been changed so that all such subpatterns are automatically treated as
1744 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1746 4. I had overlooked the fact that, in some locales, there are characters for
1747 which isalpha() is true but neither isupper() nor islower() are true. In
1748 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1749 and ordfeminine) are like this. This affected the treatment of \w and \W
1750 when they appeared in character classes, but not when they appeared outside
1751 a character class. The bit map for "word" characters is now created
1752 separately from the results of isalnum() instead of just taking it from the
1753 upper, lower, and digit maps. (Plus the underscore character, of course.)
1755 5. The above bug also affected the handling of POSIX character classes such as
1756 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1757 permanent tables. Instead, the bit maps for such a class were previously
1758 created as the appropriate unions of the upper, lower, and digit bitmaps.
1759 Now they are created by subtraction from the [[:word:]] class, which has
1760 its own bitmap.
1762 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1763 It is created by subtracting the vertical space characters (\x09, \x0a,
1764 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1765 subtraction was done in the overall bitmap for a character class, meaning
1766 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1767 be recognized. This bug has been fixed.
1769 7. Patches from the folks at Google:
1771 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1772 real life, but is still worth protecting against".
1774 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1775 regular expressions".
1777 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1778 have it.
1780 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1781 "configure" and the latter not, in order to fix a problem somebody had
1782 with compiling the Arg class on HP-UX.
1784 (e) Improve the error-handling of the C++ wrapper a little bit.
1786 (f) New tests for checking recursion limiting.
1788 8. The pcre_memmove() function, which is used only if the environment does not
1789 have a standard memmove() function (and is therefore rarely compiled),
1790 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1791 returning a result (though PCRE never actually uses the result).
1793 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1794 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1795 returned instead of calling malloc() with an overflowing number that would
1796 most likely cause subsequent chaos.
1798 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1800 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1801 with this option is matched, the nmatch and pmatch options of regexec() are
1802 ignored.
1804 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1805 provided in case anyone wants to the the POSIX interface with UTF-8
1806 strings.
1808 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1809 C++ linking (needed for some HP-UX environments).
1811 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1812 (unused parameter) and in the pcre_printint() function (omitted "default"
1813 switch label when the default is to do nothing).
1815 15. Added some code to make it possible, when PCRE is compiled as a C++
1816 library, to replace subject pointers for pcre_exec() with a smart pointer
1817 class, thus making it possible to process discontinuous strings.
1819 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1820 much the same function. They were added by different people who were trying
1821 to make PCRE easy to compile on non-Unix systems. It has been suggested
1822 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1823 for compiling on Windows systems. I have therefore replaced it with
1824 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1825 defaults to "extern" for C or "extern C" for C++, which works fine on
1826 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1827 SCOPE with something explicit in config.h. In addition:
1829 (a) pcreposix.h still had just "extern" instead of either of these macros;
1830 I have replaced it with PCRE_DATA_SCOPE.
1832 (b) Functions such as _pcre_xclass(), which are internal to the library,
1833 but external in the C sense, all had PCRE_EXPORT in their definitions.
1834 This is apparently wrong for the Windows case, so I have removed it.
1835 (It makes no difference on Unix-like systems.)
1837 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1838 of recursive calls to match(). This is different to MATCH_LIMIT because
1839 that limits the total number of calls to match(), not all of which increase
1840 the depth of recursion. Limiting the recursion depth limits the amount of
1841 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1842 when PCRE is compiled, and changed at run time. A patch from Google adds
1843 this functionality to the C++ interface.
1845 18. Changes to the handling of Unicode character properties:
1847 (a) Updated the table to Unicode 4.1.0.
1849 (b) Recognize characters that are not in the table as "Cn" (undefined).
1851 (c) I revised the way the table is implemented to a much improved format
1852 which includes recognition of ranges. It now supports the ranges that
1853 are defined in UnicodeData.txt, and it also amalgamates other
1854 characters into ranges. This has reduced the number of entries in the
1855 table from around 16,000 to around 3,000, thus reducing its size
1856 considerably. I realized I did not need to use a tree structure after
1857 all - a binary chop search is just as efficient. Having reduced the
1858 number of entries, I extended their size from 6 bytes to 8 bytes to
1859 allow for more data.
1861 (d) Added support for Unicode script names via properties such as \p{Han}.
1863 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1864 matching that character.
1866 20. When matching a repeated Unicode property with a minimum greater than zero,
1867 (for example \pL{2,}), PCRE could look past the end of the subject if it
1868 reached it while seeking the minimum number of characters. This could
1869 happen only if some of the characters were more than one byte long, because
1870 there is a check for at least the minimum number of bytes.
1872 21. Refactored the implementation of \p and \P so as to be more general, to
1873 allow for more different types of property in future. This has changed the
1874 compiled form incompatibly. Anybody with saved compiled patterns that use
1875 \p or \P will have to recompile them.
1877 22. Added "Any" and "L&" to the supported property types.
1879 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1880 but give a compile time error if the value is greater than 0xff.
1882 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1883 accidentally not being installed or uninstalled.
1885 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1886 made were to insert the current release number. This seemed silly, because
1887 it made things harder for people building PCRE on systems that don't run
1888 "configure". I have turned pcre.h into a distributed file, no longer built
1889 by "configure", with the version identification directly included. There is
1890 no longer a pcre.h.in file.
1892 However, this change necessitated a change to the pcre-config script as
1893 well. It is built from pcre-config.in, and one of the substitutions was the
1894 release number. I have updated configure.ac so that ./configure now finds
1895 the release number by grepping pcre.h.
1897 26. Added the ability to run the tests under valgrind.
1900 Version 6.4 05-Sep-05
1901 ---------------------
1903 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1904 "--" to be printed when multiple files were scanned, even when none of the
1905 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1906 consider it to be a bug, and have restored the previous behaviour.
1908 2. A couple of code tidies to get rid of compiler warnings.
1910 3. The pcretest program used to cheat by referring to symbols in the library
1911 whose names begin with _pcre_. These are internal symbols that are not
1912 really supposed to be visible externally, and in some environments it is
1913 possible to suppress them. The cheating is now confined to including
1914 certain files from the library's source, which is a bit cleaner.
1916 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1917 file's purpose clearer.
1919 5. Reorganized pcre_ucp_findchar().
1922 Version 6.3 15-Aug-05
1923 ---------------------
1925 1. The file libpcre.pc.in did not have general read permission in the tarball.
1927 2. There were some problems when building without C++ support:
1929 (a) If C++ support was not built, "make install" and "make test" still
1930 tried to test it.
1932 (b) There were problems when the value of CXX was explicitly set. Some
1933 changes have been made to try to fix these, and ...
1935 (c) --disable-cpp can now be used to explicitly disable C++ support.
1937 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1938 backslash in a target when C++ was disabled. This confuses some
1939 versions of "make", apparently. Using an intermediate variable solves
1940 this. (Same for CPP_LOBJ.)
1942 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1943 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1944 necessary on certain architectures.
1946 4. Added a setting of -export-symbols-regex to the link command to remove
1947 those symbols that are exported in the C sense, but actually are local
1948 within the library, and not documented. Their names all begin with
1949 "_pcre_". This is not a perfect job, because (a) we have to except some
1950 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1951 available (and never for static libraries). I have made a note to try to
1952 find a way round (a) in the future.
1955 Version 6.2 01-Aug-05
1956 ---------------------
1958 1. There was no test for integer overflow of quantifier values. A construction
1959 such as {1111111111111111} would give undefined results. What is worse, if
1960 a minimum quantifier for a parenthesized subpattern overflowed and became
1961 negative, the calculation of the memory size went wrong. This could have
1962 led to memory overwriting.
1964 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1966 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1967 operating environments where this matters.
1969 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1970 PCRE options from within the C++ wrapper.
1972 5. Named capturing subpatterns were not being correctly counted when a pattern
1973 was compiled. This caused two problems: (a) If there were more than 100
1974 such subpatterns, the calculation of the memory needed for the whole
1975 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1976 back references of the form \12, where the number was greater than 9, were
1977 not recognized as back references, even though there were sufficient
1978 previous subpatterns.
1980 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1981 versions of gcc, e.g. 2.95.4.
1984 Version 6.1 21-Jun-05
1985 ---------------------
1987 1. There was one reference to the variable "posix" in pcretest.c that was not
1988 surrounded by "#if !defined NOPOSIX".
1990 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1991 the cross-check on the old pcre_info() function, for the benefit of the
1992 cut-down version of PCRE that is currently imported into Exim.
1994 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1995 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1996 patterns that start (?i) but I don't think that matters. The use of (?i) is
1997 just an example; this all applies to the other options as well.
1999 4. Since libtool seems to echo the compile commands it is issuing, the output
2000 from "make" can be reduced a bit by putting "@" in front of each libtool
2001 compile command.
2003 5. Patch from the folks at Google for configure.in to be a bit more thorough
2004 in checking for a suitable C++ installation before trying to compile the
2005 C++ stuff. This should fix a reported problem when a compiler was present,
2006 but no suitable headers.
2008 6. The man pages all had just "PCRE" as their title. I have changed them to
2009 be the relevant file name. I have also arranged that these names are
2010 retained in the file doc/pcre.txt, which is a concatenation in text format
2011 of all the man pages except the little individual ones for each function.
2013 7. The NON-UNIX-USE file had not been updated for the different set of source
2014 files that come with release 6. I also added a few comments about the C++
2015 wrapper.
2018 Version 6.0 07-Jun-05
2019 ---------------------
2021 1. Some minor internal re-organization to help with my DFA experiments.
2023 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2024 didn't matter for the library itself when fully configured, but did matter
2025 when compiling without UCP support, or within Exim, where the ucp files are
2026 not imported.
2028 3. Refactoring of the library code to split up the various functions into
2029 different source modules. The addition of the new DFA matching code (see
2030 below) to a single monolithic source would have made it really too
2031 unwieldy, quite apart from causing all the code to be include in a
2032 statically linked application, when only some functions are used. This is
2033 relevant even without the DFA addition now that patterns can be compiled in
2034 one application and matched in another.
2036 The downside of splitting up is that there have to be some external
2037 functions and data tables that are used internally in different modules of
2038 the library but which are not part of the API. These have all had their
2039 names changed to start with "_pcre_" so that they are unlikely to clash
2040 with other external names.
2042 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2043 a different (DFA) algorithm. Although it is slower than the original
2044 function, it does have some advantages for certain types of matching
2045 problem.
2047 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2048 including restarting after a partial match.
2050 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2051 defined when compiling for Windows was sent to me. I have put it into the
2052 code, though I have no means of testing or verifying it.
2054 7. Added the pcre_refcount() auxiliary function.
2056 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2057 match before or at the first newline in the subject string. In pcretest,
2058 the /f option on a pattern can be used to set this.
2060 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2061 would behave wrongly. This has been present in PCRE since release 4.0.
2063 10. A number of changes to the pcregrep command:
2065 (a) Refactored how -x works; insert ^(...)$ instead of setting
2066 PCRE_ANCHORED and checking the length, in preparation for adding
2067 something similar for -w.
2069 (b) Added the -w (match as a word) option.
2071 (c) Refactored the way lines are read and buffered so as to have more
2072 than one at a time available.
2074 (d) Implemented a pcregrep test script.
2076 (e) Added the -M (multiline match) option. This allows patterns to match
2077 over several lines of the subject. The buffering ensures that at least
2078 8K, or the rest of the document (whichever is the shorter) is available
2079 for matching (and similarly the previous 8K for lookbehind assertions).
2081 (f) Changed the --help output so that it now says
2083 -w, --word-regex(p)
2085 instead of two lines, one with "regex" and the other with "regexp"
2086 because that confused at least one person since the short forms are the
2087 same. (This required a bit of code, as the output is generated
2088 automatically from a table. It wasn't just a text change.)
2090 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2091 option but starts with a hyphen. Could be a pattern or a path name
2092 starting with a hyphen, for instance.
2094 (h) "-" can be given as a file name to represent stdin.
2096 (i) When file names are being printed, "(standard input)" is used for
2097 the standard input, for compatibility with GNU grep. Previously
2098 "<stdin>" was used.
2100 (j) The option --label=xxx can be used to supply a name to be used for
2101 stdin when file names are being printed. There is no short form.
2103 (k) Re-factored the options decoding logic because we are going to add
2104 two more options that take data. Such options can now be given in four
2105 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2107 (l) Added the -A, -B, and -C options for requesting that lines of context
2108 around matches be printed.
2110 (m) Added the -L option to print the names of files that do not contain
2111 any matching lines, that is, the complement of -l.
2113 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2114 continue to scan other files.
2116 (o) The -s option was incorrectly implemented. For compatibility with other
2117 greps, it now suppresses the error message for a non-existent or non-
2118 accessible file (but not the return code). There is a new option called
2119 -q that suppresses the output of matching lines, which was what -s was
2120 previously doing.
2122 (p) Added --include and --exclude options to specify files for inclusion
2123 and exclusion when recursing.
2125 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2126 Hopefully, it now does.
2128 12. Missing cast in pcre_study().
2130 13. Added an "uninstall" target to the makefile.
2132 14. Replaced "extern" in the function prototypes in Makefile.in with
2133 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2134 world, but is set differently for Windows.
2136 15. Added a second compiling function called pcre_compile2(). The only
2137 difference is that it has an extra argument, which is a pointer to an
2138 integer error code. When there is a compile-time failure, this is set
2139 non-zero, in addition to the error test pointer being set to point to an
2140 error message. The new argument may be NULL if no error number is required
2141 (but then you may as well call pcre_compile(), which is now just a
2142 wrapper). This facility is provided because some applications need a
2143 numeric error indication, but it has also enabled me to tidy up the way
2144 compile-time errors are handled in the POSIX wrapper.
2146 16. Added VPATH=.libs to the makefile; this should help when building with one
2147 prefix path and installing with another. (Or so I'm told by someone who
2148 knows more about this stuff than I do.)
2150 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2151 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2152 match everything, including newlines. This is not POSIX-compatible, but
2153 somebody wanted the feature. From pcretest it can be activated by using
2154 both the P and the s flags.
2156 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2158 19. libpcre.pc was being incorrectly installed as executable.
2160 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2161 it now also looks for '\r' so that it will work unmodified on Windows.
2163 21. Added Google's contributed C++ wrapper to the distribution.
2165 22. Added some untidy missing memory free() calls in pcretest, to keep
2166 Electric Fence happy when testing.
2170 Version 5.0 13-Sep-04
2171 ---------------------
2173 1. Internal change: literal characters are no longer packed up into items
2174 containing multiple characters in a single byte-string. Each character
2175 is now matched using a separate opcode. However, there may be more than one
2176 byte in the character in UTF-8 mode.
2178 2. The pcre_callout_block structure has two new fields: pattern_position and
2179 next_item_length. These contain the offset in the pattern to the next match
2180 item, and its length, respectively.
2182 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2183 insertion of callouts before each pattern item. Added the /C option to
2184 pcretest to make use of this.
2186 4. On the advice of a Windows user, the lines
2188 #if defined(_WIN32) || defined(WIN32)
2189 _setmode( _fileno( stdout ), 0x8000 );
2190 #endif /* defined(_WIN32) || defined(WIN32) */
2192 have been added to the source of pcretest. This apparently does useful
2193 magic in relation to line terminators.
2195 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2196 for the benefit of those environments where the "b" makes a difference.
2198 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2199 to know about it. I have put a hack into configure.in that adds in code
2200 to set GCC=yes if CC=icc. This seems to end up at a point in the
2201 generated configure script that is early enough to affect the setting of
2202 compiler options, which is what is needed, but I have no means of testing
2203 whether it really works. (The user who reported this had patched the
2204 generated configure script, which of course I cannot do.)
2206 LATER: After change 22 below (new libtool files), the configure script
2207 seems to know about icc (and also ecc). Therefore, I have commented out
2208 this hack in configure.in.
2210 7. Added support for pkg-config (2 patches were sent in).
2212 8. Negated POSIX character classes that used a combination of internal tables
2213 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2214 [[:^ascii]]. Typically, they would match almost any characters. The other
2215 POSIX classes were not broken in this way.
2217 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2218 to find the match, as PCRE was deluded into thinking that the match had to
2219 start at the start point or following a newline. The same bug applied to
2220 patterns with negative forward assertions or any backward assertions
2221 preceding ".*" at the start, unless the pattern required a fixed first
2222 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2224 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2225 starting at the last subject character, bytes beyond the end of the subject
2226 string were read.
2228 11. Renamed the variable "class" as "classbits" to make life easier for C++
2229 users. (Previously there was a macro definition, but it apparently wasn't
2230 enough.)
2232 12. Added the new field "tables" to the extra data so that tables can be passed
2233 in at exec time, or the internal tables can be re-selected. This allows
2234 a compiled regex to be saved and re-used at a later time by a different
2235 program that might have everything at different addresses.
2237 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2238 -R library as well as a -L library.
2240 14. The debugging options of pcretest (-d on the command line or D on a
2241 pattern) showed incorrect output for anything following an extended class
2242 that contained multibyte characters and which was followed by a quantifier.
2244 15. Added optional support for general category Unicode character properties
2245 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2246 support. It adds about 90K to the size of the library. The meanings of the
2247 inbuilt class escapes such as \d and \s have NOT been changed.
2249 16. Updated pcredemo.c to include calls to free() to release the memory for the
2250 compiled pattern.
2252 17. The generated file chartables.c was being created in the source directory
2253 instead of in the building directory. This caused the build to fail if the
2254 source directory was different from the building directory, and was
2255 read-only.
2257 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2258 file. No doubt somebody will tell me if they don't make sense... Also added
2259 Dan Mooney's comments about building on OpenVMS.
2261 19. Added support for partial matching via the PCRE_PARTIAL option for
2262 pcre_exec() and the \P data escape in pcretest.
2264 20. Extended pcretest with 3 new pattern features:
2266 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2267 write the compiled pattern to the file whose name is "rest-of-line".
2268 This is a straight binary dump of the data, with the saved pointer to
2269 the character tables forced to be NULL. The study data, if any, is
2270 written too. After writing, pcretest reads a new pattern.
2272 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2273 compiled pattern from the given file. There must not be any
2274 occurrences of "<" in the file name (pretty unlikely); if there are,
2275 pcretest will instead treat the initial "<" as a pattern delimiter.
2276 After reading in the pattern, pcretest goes on to read data lines as
2277 usual.
2279 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2280 and 16-bit fields in a compiled pattern, to simulate a pattern that
2281 was compiled on a host of opposite endianness.
2283 21. The pcre-exec() function can now cope with patterns that were compiled on
2284 hosts of opposite endianness, with this restriction:
2286 As for any compiled expression that is saved and used later, the tables
2287 pointer field cannot be preserved; the extra_data field in the arguments
2288 to pcre_exec() should be used to pass in a tables address if a value
2289 other than the default internal tables were used at compile time.
2291 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2292 now diagnosed as an error. Previously, most of the time, a negative number
2293 would have been treated as zero, but if in addition "ovector" was passed as
2294 NULL, a crash could occur.
2296 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2297 new versions from the libtool 1.5 distribution (the last one is a copy of
2298 a file called libtool.m4). This seems to have fixed the need to patch
2299 "configure" to support Darwin 1.3 (which I used to do). However, I still
2300 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2301 workstation).
2303 24. Changed the PCRE licence to be the more standard "BSD" licence.
2306 Version 4.5 01-Dec-03
2307 ---------------------
2309 1. There has been some re-arrangement of the code for the match() function so
2310 that it can be compiled in a version that does not call itself recursively.
2311 Instead, it keeps those local variables that need separate instances for
2312 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2313 needs to "recurse". Keeping track of where control must go is done by means
2314 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2315 hide most of the details from the main code, and operates only if
2316 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2317 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2318 operating.
2320 To make it easier for callers to provide specially tailored get/free
2321 functions for this usage, two new functions, pcre_stack_malloc, and
2322 pcre_stack_free, are used. They are always called in strict stacking order,
2323 and the size of block requested is always the same.
2325 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2326 PCRE has been compiled to use the stack or the heap for recursion. The
2327 -C option of pcretest uses this to show which version is compiled.
2329 A new data escape \S, is added to pcretest; it causes the amounts of store
2330 obtained and freed by both kinds of malloc/free at match time to be added
2331 to the output.
2333 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2334 what's available on my current Linux desktop machine.
2336 3. When matching a UTF-8 string, the test for a valid string at the start has
2337 been extended. If start_offset is not zero, PCRE now checks that it points
2338 to a byte that is the start of a UTF-8 character. If not, it returns
2339 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2340 this is necessary because there may be backward assertions in the pattern.
2341 When matching the same subject several times, it may save resources to use
2342 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2344 4. The code for checking the validity of UTF-8 strings has been tightened so
2345 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2346 containing "overlong sequences".
2348 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2349 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2350 should have been "&", but it just so happened that all the cases this let
2351 through by mistake were picked up later in the function.
2353 6. I had used a variable called "isblank" - this is a C99 function, causing
2354 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2356 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2357 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2358 so the version is shown; (c) stop "make test" echoing "./RunTest".
2360 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2362 9. The prototype for memmove() for systems that don't have it was using
2363 size_t, but the inclusion of the header that defines size_t was later. I've
2364 moved the #includes for the C headers earlier to avoid this.
2366 10. Added some adjustments to the code to make it easier to compiler on certain
2367 special systems:
2369 (a) Some "const" qualifiers were missing.
2370 (b) Added the macro EXPORT before all exported functions; by default this
2371 is defined to be empty.
2372 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2373 that it reads its output file name as an argument instead of writing
2374 to the standard output and assuming this can be redirected.
2376 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2377 class containing characters with values greater than 255, PCRE compilation
2378 went into a loop.
2380 12. A recursive reference to a subpattern that was within another subpattern
2381 that had a minimum quantifier of zero caused PCRE to crash. For example,
2382 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2383 recursion. If the recursively-called subpattern itself had a zero repeat,
2384 that was OK.
2386 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2387 buffer into which it was copied (for escape processing) was still set at
2388 1024, so long lines caused crashes.
2390 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2391 "internal error: code overflow...". This applied to any character class
2392 that was followed by a possessive quantifier.
2394 15. Modified the Makefile to add libpcre.la as a prerequisite for
2395 libpcreposix.la because I was told this is needed for a parallel build to
2396 work.
2398 16. If a pattern that contained .* following optional items at the start was
2399 studied, the wrong optimizing data was generated, leading to matching
2400 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2401 matching string must start with a or b or c. The correct conclusion for
2402 this pattern is that a match can start with any character.
2405 Version 4.4 13-Aug-03
2406 ---------------------
2408 1. In UTF-8 mode, a character class containing characters with values between
2409 127 and 255 was not handled correctly if the compiled pattern was studied.
2410 In fixing this, I have also improved the studying algorithm for such
2411 classes (slightly).
2413 2. Three internal functions had redundant arguments passed to them. Removal
2414 might give a very teeny performance improvement.
2416 3. Documentation bug: the value of the capture_top field in a callout is *one
2417 more than* the number of the hightest numbered captured substring.
2419 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2420 in incorrectly linking with a previously installed version. They now link
2421 explicitly with libpcre.la.
2423 5. configure.in no longer needs to recognize Cygwin specially.
2425 6. A problem in pcre.in for Windows platforms is fixed.
2427 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2428 pcretest, it used to include the size of the study block as part of its
2429 output. Unfortunately, the structure contains a field that has a different
2430 size on different hardware architectures. This meant that the tests that
2431 showed this size failed. As the block is currently always of a fixed size,
2432 this information isn't actually particularly useful in pcretest output, so
2433 I have just removed it.
2435 8. Three pre-processor statements accidentally did not start in column 1.
2436 Sadly, there are *still* compilers around that complain, even though
2437 standard C has not required this for well over a decade. Sigh.
2439 9. In pcretest, the code for checking callouts passed small integers in the
2440 callout_data field, which is a void * field. However, some picky compilers
2441 complained about the casts involved for this on 64-bit systems. Now
2442 pcretest passes the address of the small integer instead, which should get
2443 rid of the warnings.
2445 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2446 both compile and run time, and gives an error if an invalid UTF-8 sequence
2447 is found. There is a option for disabling this check in cases where the
2448 string is known to be correct and/or the maximum performance is wanted.
2450 11. In response to a bug report, I changed one line in Makefile.in from
2452 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2453 to
2454 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2456 to look similar to other lines, but I have no way of telling whether this
2457 is the right thing to do, as I do not use Windows. No doubt I'll get told
2458 if it's wrong...
2461 Version 4.3 21-May-03
2462 ---------------------
2464 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2465 Makefile.
2467 2. Some refactoring to improve the quality of the code:
2469 (i) The utf8_table... variables are now declared "const".
2471 (ii) The code for \cx, which used the "case flipping" table to upper case
2472 lower case letters, now just substracts 32. This is ASCII-specific,
2473 but the whole concept of \cx is ASCII-specific, so it seems
2474 reasonable.
2476 (iii) PCRE was using its character types table to recognize decimal and
2477 hexadecimal digits in the pattern. This is silly, because it handles
2478 only 0-9, a-f, and A-F, but the character types table is locale-
2479 specific, which means strange things might happen. A private
2480 table is now used for this - though it costs 256 bytes, a table is
2481 much faster than multiple explicit tests. Of course, the standard
2482 character types table is still used for matching digits in subject
2483 strings against \d.
2485 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2486 ending in _t are). So I've renamed it as ESC_tee.
2488 3. The first argument for regexec() in the POSIX wrapper should have been
2489 defined as "const".
2491 4. Changed pcretest to use malloc() for its buffers so that they can be
2492 Electric Fenced for debugging.
2494 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2495 to read one or more bytes before the start of the subject string. Often this
2496 had no effect on PCRE's behaviour, but in some circumstances it could
2497 provoke a segmentation fault.
2499 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2500 to try to read one or more bytes before the start of the subject string.
2502 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2503 UTF-8 support could misbehave in various ways if the subject string
2504 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2505 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2506 back over UTF-8 characters.)
2509 Version 4.2 14-Apr-03
2510 ---------------------
2512 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2514 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2515 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2516 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2517 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2518 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2520 Note: automatic setting of the BUILD variables is not yet working
2521 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2522 compile-time but not at link-time
2523 [LINK]: use for linking executables only
2524 make different versions for Windows and non-Windows
2525 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2526 libraries
2527 [LINK_FOR_BUILD]: new variable
2528 [OBJEXT]: use throughout
2529 [EXEEXT]: use throughout
2530 <winshared>: new target
2531 <wininstall>: new target
2532 <dftables.o>: use native compiler
2533 <dftables>: use native linker
2534 <install>: handle Windows platform correctly
2535 <clean>: ditto
2536 <check>: ditto
2537 copy DLL to top builddir before testing
2539 As part of these changes, -no-undefined was removed again. This was reported
2540 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2541 in any case.
2543 3. Some tidies to get rid of compiler warnings:
2545 . In the match_data structure, match_limit was an unsigned long int, whereas
2546 match_call_count was an int. I've made them both unsigned long ints.
2548 . In pcretest the fact that a const uschar * doesn't automatically cast to
2549 a void * provoked a warning.
2551 . Turning on some more compiler warnings threw up some "shadow" variables
2552 and a few more missing casts.
2554 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2555 option, a class that contained a single character with a value between 128
2556 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2558 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2559 option, a class that contained several characters, but with at least one
2560 whose value was between 128 and 255 caused PCRE to crash.
2563 Version 4.1 12-Mar-03
2564 ---------------------
2566 1. Compiling with gcc -pedantic found a couple of places where casts were
2567 needed, and a string in dftables.c that was longer than standard compilers are
2568 required to support.
2570 2. Compiling with Sun's compiler found a few more places where the code could
2571 be tidied up in order to avoid warnings.
2573 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2574 first of these names is deprecated in the latest Autoconf in favour of the name
2575 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2576 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2577 analogy I have changed it to CFLAGS_FOR_BUILD.
2579 4. Added -no-undefined to the linking command in the Makefile, because this is
2580 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2581 linking step for the pcreposix library.
2583 5. PCRE was failing to diagnose the case of two named groups with the same
2584 name.
2586 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2587 literal character that is needed in the subject for a match, and scans along to
2588 ensure that it is present before embarking on the full matching process. This
2589 saves time in cases of nested unlimited repeats that are never going to match.
2590 Problem: the scan can take a lot of time if the subject is very long (e.g.
2591 megabytes), thus penalizing straightforward matches. It is now done only if the
2592 amount of subject to be scanned is less than 1000 bytes.
2594 7. A lesser problem with the same optimization is that it was recording the
2595 first character of an anchored pattern as "needed", thus provoking a search
2596 right along the subject, even when the first match of the pattern was going to
2597 fail. The "needed" character is now not set for anchored patterns, unless it
2598 follows something in the pattern that is of non-fixed length. Thus, it still
2599 fulfils its original purpose of finding quick non-matches in cases of nested
2600 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2603 Version 4.0 17-Feb-03
2604 ---------------------
2606 1. If a comment in an extended regex that started immediately after a meta-item
2607 extended to the end of string, PCRE compiled incorrect data. This could lead to
2608 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2610 2. Moved to autoconf 2.53 and libtool 1.4.2.
2612 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2613 the special perltest8 script is no longer needed - all the tests can be run
2614 from a single perltest script.
2616 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2617 by \s. It has now been removed in PCRE. This means it isn't recognized as
2618 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2619 class [:space:] *does* include VT, thereby creating a mess.
2621 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2622 space and tab.
2624 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2625 its new features into the main test script, reducing the number of scripts.
2627 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2628 were backward compatible, and made the (?i) apply to the whole pattern, as if
2629 /i were given. Now it behaves more logically, and applies the option setting
2630 only to what follows. PCRE has been changed to follow suit. However, if it
2631 finds options settings right at the start of the pattern, it extracts them into
2632 the global options, as before. Thus, they show up in the info data.
2634 8. Added support for the \Q...\E escape sequence. Characters in between are
2635 treated as literals. This is slightly different from Perl in that $ and @ are
2636 also handled as literals inside the quotes. In Perl, they will cause variable
2637 interpolation. Note the following examples:
2639 Pattern PCRE matches Perl matches
2641 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2642 \Qabc\$xyz\E abc\$xyz abc\$xyz
2643 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2645 For compatibility with Perl, \Q...\E sequences are recognized inside character
2646 classes as well as outside them.
2648 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2649 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2650 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2651 signed/unsigned warnings.
2653 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2654 option for pcretest, so I've replaced it by a simple function that does just
2655 that job.
2657 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2658 "pcregrep -".
2660 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2661 Java package. This provides some syntactic sugar for simple cases of what my
2662 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2663 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2664 item, you can use this simplified notation. Note that only makes sense with
2665 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2666 greediness, whatever the setting of the PCRE_UNGREEDY option.
2668 13. A change of greediness default within a pattern was not taking effect at
2669 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2670 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2671 was abstracted outside.
2673 14. PCRE now supports the \G assertion. It is true when the current matching
2674 position is at the start point of the match. This differs from \A when the
2675 starting offset is non-zero. Used with the /g option of pcretest (or similar
2676 code), it works in the same way as it does for Perl's /g option. If all
2677 alternatives of a regex begin with \G, the expression is anchored to the start
2678 match position, and the "anchored" flag is set in the compiled expression.
2680 15. Some bugs concerning the handling of certain option changes within patterns
2681 have been fixed. These applied to options other than (?ims). For example,
2682 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2683 been the other way round. Some of this was related to change 7 above.
2685 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2686 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2687 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2688 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2690 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2691 mode. Unlike ".", it always matches newline, whatever the setting of
2692 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2693 assertions. Perl allows it, but it doesn't (in general) work because it can't
2694 calculate the length of the lookbehind. At least, that's the case for Perl
2695 5.8.0 - I've been told they are going to document that it doesn't work in
2696 future.
2698 18. Added an error diagnosis for escapes that PCRE does not support: these are
2699 \L, \l, \N, \P, \p, \U, \u, and \X.
2701 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2702 reading past the end of the pattern in cases such as /[abcd/.
2704 20. PCRE was getting more memory than necessary for patterns with classes that
2705 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2707 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2708 compiling PCRE for use with Virtual Pascal.
2710 22. Small fix to the Makefile to make it work properly if the build is done
2711 outside the source tree.
2713 23. Added a new extension: a condition to go with recursion. If a conditional
2714 subpattern starts with (?(R) the "true" branch is used if recursion has
2715 happened, whereas the "false" branch is used only at the top level.
2717 24. When there was a very long string of literal characters (over 255 bytes
2718 without UTF support, over 250 bytes with UTF support), the computation of how
2719 much memory was required could be incorrect, leading to segfaults or other
2720 strange effects.
2722 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2723 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2724 there was a subsequent back reference to those brackets. This meant that, for
2725 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2726 possible to check for precisely this case. All we can do is abandon the
2727 optimization if .* occurs inside capturing brackets when there are any back
2728 references whatsoever. (See below for a better fix that came later.)
2730 26. The handling of the optimization for finding the first character of a
2731 non-anchored pattern, and for finding a character that is required later in the
2732 match were failing in some cases. This didn't break the matching; it just
2733 failed to optimize when it could. The way this is done has been re-implemented.
2735 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2737 28. Added a new feature that provides some of the functionality that Perl
2738 provides with (?{...}). The facility is termed a "callout". The way it is done
2739 in PCRE is for the caller to provide an optional function, by setting
2740 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2741 global variable. By default it is unset, which disables all calling out. To get
2742 the function called, the regex must include (?C) at appropriate points. This
2743 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2744 This provides a means of identifying different callout points. When PCRE
2745 reaches such a point in the regex, if pcre_callout has been set, the external
2746 function is called. It is provided with data in a structure called
2747 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2748 matching continues; if it returns a non-zero value, the match at the current
2749 point fails. However, backtracking will occur if possible. [This was changed
2750 later and other features added - see item 49 below.]
2752 29. pcretest is upgraded to test the callout functionality. It provides a
2753 callout function that displays information. By default, it shows the start of
2754 the match and the current position in the text. There are some new data escapes
2755 to vary what happens:
2757 \C+ in addition, show current contents of captured substrings
2758 \C- do not supply a callout function
2759 \C!n return 1 when callout number n is reached
2760 \C!n!m return 1 when callout number n is reached for the mth time
2762 30. If pcregrep was called with the -l option and just a single file name, it
2763 output "<stdin>" if a match was found, instead of the file name.
2765 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2766 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2767 pcre_exec(). This saves a malloc/free per call. The default value of
2768 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2769 when configuring.
2771 32. The default maximum size of a compiled pattern is 64K. There have been a
2772 few cases of people hitting this limit. The code now uses macros to handle the
2773 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2774 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2775 configuring. Tests 2 and 5 work only with 2-byte links because they output
2776 debugging information about compiled patterns.
2778 33. Internal code re-arrangements:
2780 (a) Moved the debugging function for printing out a compiled regex into
2781 its own source file (printint.c) and used #include to pull it into
2782 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2783 separate copies.
2785 (b) Defined the list of op-code names for debugging as a macro in
2786 internal.h so that it is next to the definition of the opcodes.
2788 (c) Defined a table of op-code lengths for simpler skipping along compiled
2789 code. This is again a macro in internal.h so that it is next to the
2790 definition of the opcodes.
2792 34. Added support for recursive calls to individual subpatterns, along the
2793 lines of Robin Houston's patch (but implemented somewhat differently).
2795 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2796 allow it to read and process whole directories in Win32. This code was
2797 contributed by Lionel Fourquaux; it has not been tested by me.
2799 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2800 used to name a group. Names consist of alphanumerics and underscores, and must
2801 be unique. Back references use the syntax (?P=name) and recursive calls use
2802 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2803 numbers. The function pcre_fullinfo() can be used after compilation to extract
2804 a name/number map. There are three relevant calls:
2806 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2807 PCRE_INFO_NAMECOUNT yields the number of entries
2808 PCRE_INFO_NAMETABLE yields a pointer to the map.
2810 The map is a vector of fixed-size entries. The size of each entry depends on
2811 the length of the longest name used. The first two bytes of each entry are the
2812 group number, most significant byte first. There follows the corresponding
2813 name, zero terminated. The names are in alphabetical order.
2815 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2816 case instead of 255. Making it the same both with and without UTF-8 support
2817 means that the same test output works with both.
2819 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2820 calling malloc() with a zero argument.
2822 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2823 optimization. I've improved things by keeping a bitmap of backreferences with
2824 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2825 fact referenced, the optimization can be applied. It is unlikely that a
2826 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2827 the match to follow \n) will appear inside brackets with a number greater than
2828 31, but if it does, any back reference > 31 suppresses the optimization.
2830 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2831 of disabling numbered capturing parentheses. Any opening parenthesis that is
2832 not followed by ? behaves as if it were followed by ?: but named parentheses
2833 can still be used for capturing (and they will acquire numbers in the usual
2834 way).
2836 41. Redesigned the return codes from the match() function into yes/no/error so
2837 that errors can be passed back from deep inside the nested calls. A malloc
2838 failure while inside a recursive subpattern call now causes the
2839 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2841 42. It is now possible to set a limit on the number of times the match()
2842 function is called in a call to pcre_exec(). This facility makes it possible to
2843 limit the amount of recursion and backtracking, though not in a directly
2844 obvious way, because the match() function is used in a number of different
2845 circumstances. The count starts from zero for each position in the subject
2846 string (for non-anchored patterns). The default limit is, for compatibility, a
2847 large number, namely 10 000 000. You can change this in two ways:
2849 (a) When configuring PCRE before making, you can use --with-match-limit=n
2850 to set a default value for the compiled library.
2852 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2853 a different value is set. See 45 below.
2855 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2857 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2858 of things that can be changed at compile time. The first argument specifies
2859 what is wanted and the second points to where the information is to be placed.
2860 The current list of available information is:
2864 The output is an integer that is set to one if UTF-8 support is available;
2865 otherwise it is set to zero.
2869 The output is an integer that it set to the value of the code that is used for
2870 newline. It is either LF (10) or CR (13).
2874 The output is an integer that contains the number of bytes used for internal
2875 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2879 The output is an integer that contains the threshold above which the POSIX
2880 interface uses malloc() for output vectors. See item 31 above.
2884 The output is an unsigned integer that contains the default limit of the number
2885 of match() calls in a pcre_exec() execution. See 42 above.
2887 44. pcretest has been upgraded by the addition of the -C option. This causes it
2888 to extract all the available output from the new pcre_config() function, and to
2889 output it. The program then exits immediately.
2891 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2892 order to support additional features. One way would have been to define
2893 pcre_exec2() (for example) with extra arguments, but this would not have been
2894 extensible, and would also have required all calls to the original function to
2895 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2896 is used for passing in "extra" data from pcre_study().
2898 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2899 contains the following fields:
2901 flags a bitmap indicating which of the following fields are set
2902 study_data opaque data from pcre_study()
2903 match_limit a way of specifying a limit on match() calls for a specific
2904 call to pcre_exec()
2905 callout_data data for callouts (see 49 below)
2907 The flag bits are also defined in pcre.h, and are
2913 The pcre_study() function now returns one of these new pcre_extra blocks, with
2914 the actual study data pointed to by the study_data field, and the
2915 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2916 before. That is, this change is entirely upwards-compatible and requires no
2917 change to existing code.
2919 If you want to pass in additional data to pcre_exec(), you can either place it
2920 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2921 block.
2923 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2924 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2925 times with different match limits, until it finds the minimum value needed for
2926 pcre_exec() to complete. The value is then output. This can be instructive; for
2927 most simple matches the number is quite small, but for pathological cases it
2928 gets very large very quickly.
2930 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2931 returns the size of the data block pointed to by the study_data field in a
2932 pcre_extra block, that is, the value that was passed as the argument to
2933 pcre_malloc() when PCRE was getting memory in which to place the information
2934 created by pcre_study(). The fourth argument should point to a size_t variable.
2935 pcretest has been extended so that this information is shown after a successful
2936 pcre_study() call when information about the compiled regex is being displayed.
2938 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2939 because what follows is always an absolute path. (Later: it turns out that this
2940 is more than cosmetic for MinGW, because it doesn't like empty path
2941 components.)
2943 49. Some changes have been made to the callout feature (see 28 above):
2945 (i) A callout function now has three choices for what it returns:
2947 0 => success, carry on matching
2948 > 0 => failure at this point, but backtrack if possible
2949 < 0 => serious error, return this value from pcre_exec()
2951 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2952 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2953 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2954 use by callout functions. It will never be used by PCRE itself.
2956 (ii) The pcre_extra structure (see 45 above) has a void * field called
2957 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2958 pcre_callout_block structure has a field of the same name. The contents of
2959 the field passed in the pcre_extra structure are passed to the callout
2960 function in the corresponding field in the callout block. This makes it
2961 easier to use the same callout-containing regex from multiple threads. For
2962 testing, the pcretest program has a new data escape
2964 \C*n pass the number n (may be negative) as callout_data
2966 If the callout function in pcretest receives a non-zero value as
2967 callout_data, it returns that value.
2969 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2970 there were some redundant $(CFLAGS) in commands that are now specified as
2971 $(LINK), which already includes $(CFLAGS).
2973 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2974 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2975 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2976 one-byte characters throughout. Note that case-insensitive matching applies
2977 only to characters whose values are less than 256. PCRE doesn't support the
2978 notion of cases for higher-valued characters.
2980 (i) A character class whose characters are all within 0-255 is handled as
2981 a bit map, and the map is inverted for negative classes. Previously, a
2982 character > 255 always failed to match such a class; however it should
2983 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2985 (ii) A negated character class with a single character < 255 is coded as
2986 "not this character" (OP_NOT). This wasn't working properly when the test
2987 character was multibyte, either singly or repeated.
2989 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2990 mode, for example: \x{100}{2,3}.
2992 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2993 singly or repeated) now correctly test multibyte characters. However,
2994 PCRE doesn't recognize any characters with values greater than 255 as
2995 digits, spaces, or word characters. Such characters always match \D, \S,
2996 and \W, and never match \d, \s, or \w.
2998 (v) Classes may now contain characters and character ranges with values
2999 greater than 255. For example: [ab\x{100}-\x{400}].
3001 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3002 PCRE in UTF-8 mode.
3004 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3005 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3006 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3007 value.)
3009 53. The single man page has become too large. I have therefore split it up into
3010 a number of separate man pages. These also give rise to individual HTML pages;
3011 these are now put in a separate directory, and there is an index.html page that
3012 lists them all. Some hyperlinking between the pages has been installed.
3014 54. Added convenience functions for handling named capturing parentheses.
3016 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3017 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3018 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3019 are faulted.
3021 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3022 calling configure. These values are used when compiling the dftables.c program
3023 which is run to generate the source of the default character tables. They
3024 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3025 you will need to set these values.
3027 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3030 Version 3.9 02-Jan-02
3031 ---------------------
3033 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3035 2. If --disable-static was given, the building process failed when trying to
3036 build pcretest and pcregrep. (For some reason it was using libtool to compile
3037 them, which is not right, as they aren't part of the library.)
3040 Version 3.8 18-Dec-01
3041 ---------------------
3043 1. The experimental UTF-8 code was completely screwed up. It was packing the
3044 bytes in the wrong order. How dumb can you get?
3047 Version 3.7 29-Oct-01
3048 ---------------------
3050 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3051 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3052 this didn't happen under Solaris 8, where I normally test things.
3054 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3055 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3056 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3057 this fix an hour or so after the initial 3.7 release.)
3060 Version 3.6 23-Oct-01
3061 ---------------------
3063 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3064 offsets passed as NULL with zero offset count.
3066 2. The config.guess and config.sub files had not been updated when I moved to
3067 the latest autoconf.
3070 Version 3.5 15-Aug-01
3071 ---------------------
3073 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3074 had been forgotten.
3076 2. By using declared but undefined structures, we can avoid using "void"
3077 definitions in pcre.h while keeping the internal definitions of the structures
3078 private.
3080 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3081 user point of view, this means that both static and shared libraries are built
3082 by default, but this can be individually controlled. More of the work of
3083 handling this static/shared cases is now inside libtool instead of PCRE's make
3084 file.
3086 4. The pcretest utility is now installed along with pcregrep because it is
3087 useful for users (to test regexs) and by doing this, it automatically gets
3088 relinked by libtool. The documentation has been turned into a man page, so
3089 there are now .1, .txt, and .html versions in /doc.
3091 5. Upgrades to pcregrep:
3092 (i) Added long-form option names like gnu grep.
3093 (ii) Added --help to list all options with an explanatory phrase.
3094 (iii) Added -r, --recursive to recurse into sub-directories.
3095 (iv) Added -f, --file to read patterns from a file.
3097 6. pcre_exec() was referring to its "code" argument before testing that
3098 argument for NULL (and giving an error if it was NULL).
3100 7. Upgraded Makefile.in to allow for compiling in a different directory from
3101 the source directory.
3103 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3104 options bits, the pointer it was passed was to an int instead of to an unsigned
3105 long int. This mattered only on 64-bit systems.
3107 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3108 generated) instead of pcre.in, which it its source. Also made the same change
3109 in several of the .c files.
3111 10. A new release of gcc defines printf() as a macro, which broke pcretest
3112 because it had an ifdef in the middle of a string argument for printf(). Fixed
3113 by using separate calls to printf().
3115 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3116 script, to force use of CR or LF instead of \n in the source. On non-Unix
3117 systems, the value can be set in config.h.
3119 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3120 absolute limit. Changed the text of the error message to make this clear, and
3121 likewise updated the man page.
3123 13. The limit of 99 on the number of capturing subpatterns has been removed.
3124 The new limit is 65535, which I hope will not be a "real" limit.
3127 Version 3.4 22-Aug-00
3128 ---------------------
3130 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3132 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3135 Version 3.3 01-Aug-00
3136 ---------------------
3138 1. If an octal character was given, but the value was greater than \377, it
3139 was not getting masked to the least significant bits, as documented. This could
3140 lead to crashes in some systems.
3142 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3143 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3145 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3146 These just pass their arguments on to (pcre_free)(), but they are provided
3147 because some uses of PCRE bind it to non-C systems that can call its functions,
3148 but cannot call free() or pcre_free() directly.
3150 4. Add "make test" as a synonym for "make check". Corrected some comments in
3151 the Makefile.
3153 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3154 Makefile.
3156 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3157 command called pgrep for grepping around the active processes.
3159 7. Added the beginnings of support for UTF-8 character strings.
3161 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3162 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3163 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3164 out for the ar command.)
3167 Version 3.2 12-May-00
3168 ---------------------
3170 This is purely a bug fixing release.
3172 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3173 of ZA. This was just one example of several cases that could provoke this bug,
3174 which was introduced by change 9 of version 2.00. The code for breaking
3175 infinite loops after an iteration that matches an empty string was't working
3176 correctly.
3178 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3179 when matched against abbab (for example). After matching an empty string, it
3180 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3181 caused it to match further down the string than it should.
3183 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3184 was there because it doesn't seem to be needed, and it causes trouble on some
3185 systems, as it is not a Standard C header. It has been removed.
3187 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3188 were reported on the Macintosh. The changes were from
3190 while ((c = *(++ptr)) != 0 && c != '\n');
3191 to
3192 while ((c = *(++ptr)) != 0 && c != '\n') ;
3194 Totally extraordinary, but if that's what it takes...
3196 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3197 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3198 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3199 assumes the way PCRE uses memmove() (always moving upwards).
3201 6. PCRE is being used in one environment where strchr() is not available. There
3202 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3203 faster code anyway.
3206 Version 3.1 09-Feb-00
3207 ---------------------
3209 The only change in this release is the fixing of some bugs in Makefile.in for
3210 the "install" target:
3212 (1) It was failing to install pcreposix.h.
3214 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3217 Version 3.0 01-Feb-00
3218 ---------------------
3220 1. Add support for the /+ modifier to perltest (to output $` like it does in
3221 pcretest).
3223 2. Add support for the /g modifier to perltest.
3225 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3226 matches null strings.
3228 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3229 pattern. Perl treats empty patterns specially - it reuses the most recent
3230 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3231 effect.
3233 5. The POSIX interface was broken in that it was just handing over the POSIX
3234 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3235 required a bigger vector, with some working space on the end. This means that
3236 the POSIX wrapper now has to get and free some memory, and copy the results.
3238 6. Added some simple autoconf support, placing the test data and the
3239 documentation in separate directories, re-organizing some of the
3240 information files, and making it build pcre-config (a GNU standard). Also added
3241 libtool support for building PCRE as a shared library, which is now the
3242 default.
3244 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3245 09 are not valid octal constants. Single digits will be used for minor values
3246 less than 10.
3248 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3249 existing programs that set these in the POSIX interface can use PCRE without
3250 modification.
3252 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3253 return all that pcre_info() returns, plus additional data. The pcre_info()
3254 function is retained for compatibility, but is considered to be obsolete.
3256 10. Added experimental recursion feature (?R) to handle one common case that
3257 Perl 5.6 will be able to do with (?p{...}).
3259 11. Added support for POSIX character classes like [:alpha:], which Perl is
3260 adopting.
3263 Version 2.08 31-Aug-99
3264 ----------------------
3266 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3267 trying to match at the startoffset position, but instead was moving forward to
3268 the next newline as if a previous match had failed.
3270 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3271 and could get into a loop if a null string was matched other than at the start
3272 of the subject.
3274 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3275 be distinguished at compile time, and for completeness also added PCRE_DATE.
3277 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3278 in GnuWin32 environments.
3281 Version 2.07 29-Jul-99
3282 ----------------------
3284 1. The documentation is now supplied in plain text form and HTML as well as in
3285 the form of man page sources.
3287 2. C++ compilers don't like assigning (void *) values to other pointer types.
3288 In particular this affects malloc(). Although there is no problem in Standard
3289 C, I've put in casts to keep C++ compilers happy.
3291 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3292 should be (const char *).
3294 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3295 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3296 However, I haven't made this a standard facility. The documentation doesn't
3297 mention it, and the Makefile doesn't support it.
3299 5. The Makefile now contains an "install" target, with editable destinations at
3300 the top of the file. The pcretest program is not installed.
3302 6. pgrep -V now gives the PCRE version number and date.
3304 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3305 causing the entire string to be ignored, instead of just the last character.
3307 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3308 non-matching string, it can take a very, very long time, even for strings of
3309 quite modest length, because of the nested recursion. PCRE now does better in
3310 some of these cases. It does this by remembering the last required literal
3311 character in the pattern, and pre-searching the subject to ensure it is present
3312 before running the real match. In other words, it applies a heuristic to detect
3313 some types of certain failure quickly, and in the above example, if presented
3314 with a string that has no trailing " it gives "no match" very quickly.
3316 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3317 other alternatives are tried instead.
3320 Version 2.06 09-Jun-99
3321 ----------------------
3323 1. Change pcretest's output for amount of store used to show just the code
3324 space, because the remainder (the data block) varies in size between 32-bit and
3325 64-bit systems.
3327 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3328 start matching at. This allows lookbehinds to work when searching for multiple
3329 occurrences in a string.
3331 3. Added additional options to pcretest for testing multiple occurrences:
3333 /+ outputs the rest of the string that follows a match
3334 /g loops for multiple occurrences, using the new startoffset argument
3335 /G loops for multiple occurrences by passing an incremented pointer
3337 4. PCRE wasn't doing the "first character" optimization for patterns starting
3338 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3339 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3340 the letter 'x'. On long subject strings, this gives a significant speed-up.
3343 Version 2.05 21-Apr-99
3344 ----------------------
3346 1. Changed the type of magic_number from int to long int so that it works
3347 properly on 16-bit systems.
3349 2. Fixed a bug which caused patterns starting with .* not to work correctly
3350 when the subject string contained newline characters. PCRE was assuming
3351 anchoring for such patterns in all cases, which is not correct because .* will
3352 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3353 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3354 must be retried after every newline in the subject.
3357 Version 2.04 18-Feb-99
3358 ----------------------
3360 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3361 computation of the store needed to hold the pattern was incorrect (too large).
3362 If such patterns were nested a few deep, this could multiply and become a real
3363 problem.
3365 2. Added /M option to pcretest to show the memory requirement of a specific
3366 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3368 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3369 compiled in such a way that the backtracking after subsequent failure was
3370 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3371 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3374 Version 2.03 02-Feb-99
3375 ----------------------
3377 1. Fixed typo and small mistake in man page.
3379 2. Added 4th condition (GPL supersedes if conflict) and created separate
3380 LICENCE file containing the conditions.
3382 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3383 Perl, that is the internal \ allows the delimiter to be included in the
3384 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3385 the final delimiter, add \ to the end of the pattern (to test the error).
3387 4. Added the convenience functions for extracting substrings after a successful
3388 match. Updated pcretest to make it able to test these functions.
3391 Version 2.02 14-Jan-99
3392 ----------------------
3394 1. Initialized the working variables associated with each extraction so that
3395 their saving and restoring doesn't refer to uninitialized store.
3397 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3398 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3399 fix the problem.
3401 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3402 calls, and wasn't printing the correct value for compiling calls. Increased the
3403 default value of LOOPREPEAT, and the number of significant figures in the
3404 times.
3406 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3408 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3409 a building problem on Windows NT with a FAT file system.
3412 Version 2.01 21-Oct-98
3413 ----------------------
3415 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3416 to character tables built by pcre_maketables() in the current locale. If NULL
3417 is passed, the default tables are used.
3420 Version 2.00 24-Sep-98
3421 ----------------------
3423 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3424 it any more.
3426 2. Allow quantification of (?>) groups, and make it work correctly.
3428 3. The first character computation wasn't working for (?>) groups.
3430 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3431 end of the subject) and add 5.005's \z, which really does match only at the
3432 very end of the subject.
3434 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3436 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3437 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3438 localized options. All options to pcre_study() were also removed.
3440 7. Add other new features from 5.005:
3442 $(?<= positive lookbehind
3443 $(?<! negative lookbehind
3444 (?imsx-imsx) added the unsetting capability
3445 such a setting is global if at outer level; local otherwise
3446 (?imsx-imsx:) non-capturing groups with option setting
3447 (?(cond)re|re) conditional pattern matching
3449 A backreference to itself in a repeated group matches the previous
3450 captured string.
3452 8. General tidying up of studying (both automatic and via "study")
3453 consequential on the addition of new assertions.
3455 9. As in 5.005, unlimited repeated groups that could match an empty substring
3456 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3457 runtime if any iteration does actually match an empty substring.
3459 10. Include the RunTest script in the distribution.
3461 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3462 discrepancies, some of which were old and were also with respect to 5.004. They
3463 have now been fixed.
3466 Version 1.09 28-Apr-98
3467 ----------------------
3469 1. A negated single character class followed by a quantifier with a minimum
3470 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3471 program crashes, or just wrong answers. This did not apply to negated classes
3472 containing more than one character, or to minima other than one.
3475 Version 1.08 27-Mar-98
3476 ----------------------
3478 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3480 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3481 latter must appear before anything that relies on it in the pattern.
3484 Version 1.07 16-Feb-98
3485 ----------------------
3487 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3488 repeat of a potentially empty string).
3491 Version 1.06 23-Jan-98
3492 ----------------------
3494 1. Added Markus Oberhumer's little patches for C++.
3496 2. Literal strings longer than 255 characters were broken.
3499 Version 1.05 23-Dec-97
3500 ----------------------
3502 1. Negated character classes containing more than one character were failing if
3503 PCRE_CASELESS was set at run time.
3506 Version 1.04 19-Dec-97
3507 ----------------------
3509 1. Corrected the man page, where some "const" qualifiers had been omitted.
3511 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3512 input syntax.
3514 3. Fixed memory leak which occurred when a regex with back references was
3515 matched with an offsets vector that wasn't big enough. The temporary memory
3516 that is used in this case wasn't being freed if the match failed.
3518 4. Tidied pcretest to ensure it frees memory that it gets.
3520 5. Temporary memory was being obtained in the case where the passed offsets
3521 vector was exactly big enough.
3523 6. Corrected definition of offsetof() from change 5 below.
3525 7. I had screwed up change 6 below and broken the rules for the use of
3526 setjmp(). Now fixed.
3529 Version 1.03 18-Dec-97
3530 ----------------------
3532 1. A erroneous regex with a missing opening parenthesis was correctly
3533 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3534 on some systems.
3536 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3537 it was reported that one broken compiler failed on the former because "code" is
3538 also an independent variable.
3540 3. The erroneous regex a[]b caused an array overrun reference.
3542 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3543 fail on data ending with that character. (It was going on too far, and checking
3544 the next character, typically a binary zero.) This was specific to the
3545 optimized code for single-character negative classes.
3547 5. Added a contributed patch from the TIN world which does the following:
3549 + Add an undef for memmove, in case the the system defines a macro for it.
3551 + Add a definition of offsetof(), in case there isn't one. (I don't know
3552 the reason behind this - offsetof() is part of the ANSI standard - but
3553 it does no harm).
3555 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3556 most of the places where whitespace preceded '#'. I have given up and
3557 allowed the remaining 2 cases to be at the margin.
3559 + Rename some variables in pcre to eliminate shadowing. This seems very
3560 pedantic, but does no harm, of course.
3562 6. Moved the call to setjmp() into its own function, to get rid of warnings
3563 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3565 7. Constructs such as \d{8,} were compiling into the equivalent of
3566 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3567 outcome, but in this particular case used more store than had been allocated,
3568 which caused the bug to be discovered because it threw up an internal error.
3570 8. The debugging code in both pcre and pcretest for outputting the compiled
3571 form of a regex was going wrong in the case of back references followed by
3572 curly-bracketed repeats.
3575 Version 1.02 12-Dec-97
3576 ----------------------
3578 1. Typos in pcre.3 and comments in the source fixed.
3580 2. Applied a contributed patch to get rid of places where it used to remove
3581 'const' from variables, and fixed some signed/unsigned and uninitialized
3582 variable warnings.
3584 3. Added the "runtest" target to Makefile.
3586 4. Set default compiler flag to -O2 rather than just -O.
3589 Version 1.01 19-Nov-97
3590 ----------------------
3592 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3593 like /([ab]*)*/, that is, for classes with more than one character in them.
3595 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3596 as /((?>a*))*/ (a PCRE_EXTRA facility).
3599 Version 1.00 18-Nov-97
3600 ----------------------
3602 1. Added compile-time macros to support systems such as SunOS4 which don't have
3603 memmove() or strerror() but have other things that can be used instead.
3605 2. Arranged that "make clean" removes the executables.
3608 Version 0.99 27-Oct-97
3609 ----------------------
3611 1. Fixed bug in code for optimizing classes with only one character. It was
3612 initializing a 32-byte map regardless, which could cause it to run off the end
3613 of the memory it had got.
3615 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3618 Version 0.98 22-Oct-97
3619 ----------------------
3621 1. Fixed bug in code for handling temporary memory usage when there are more
3622 back references than supplied space in the ovector. This could cause segfaults.
3625 Version 0.97 21-Oct-97
3626 ----------------------
3628 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3630 2. Optimized negated single characters not to use a bit map.
3632 3. Brought error texts together as macro definitions; clarified some of them;
3633 fixed one that was wrong - it said "range out of order" when it meant "invalid
3634 escape sequence".
3636 4. Changed some char * arguments to const char *.
3638 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3640 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3641 pcretest.
3644 Version 0.96 16-Oct-97
3645 ----------------------
3647 1. Added a simple "pgrep" utility to the distribution.
3649 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3650 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3651 where "ddd" means "one or more decimal digits".
3653 3. Fixed serious bug. If a pattern had a back reference, but the call to
3654 pcre_exec() didn't supply a large enough ovector to record the related
3655 identifying subpattern, the match always failed. PCRE now remembers the number
3656 of the largest back reference, and gets some temporary memory in which to save
3657 the offsets during matching if necessary, in order to ensure that
3658 backreferences always work.
3660 4. Increased the compatibility with Perl in a number of ways:
3662 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3663 to request this handling. The option can be set at compile or exec time.
3665 (b) $ matches before a terminating newline by default; an option
3666 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3667 mode). The option can be set at compile or exec time.
3669 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3670 the same as Perl's. If the decimal number it represents is less than 10
3671 or there aren't that many previous left capturing parentheses, an octal
3672 escape is read. Inside a character class, it's always an octal escape,
3673 even if it is a single digit.
3675 (d) An escaped but undefined alphabetic character is taken as a literal,
3676 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3677 escapes.
3679 (e) {0} is now permitted. (The previous item is removed from the compiled
3680 pattern).
3682 5. Changed all the names of code files so that the basic parts are no longer
3683 than 10 characters, and abolished the teeny "globals.c" file.
3685 6. Changed the handling of character classes; they are now done with a 32-byte
3686 bit map always.
3688 7. Added the -d and /D options to pcretest to make it possible to look at the
3689 internals of compilation without having to recompile pcre.
3692 Version 0.95 23-Sep-97
3693 ----------------------
3695 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3696 \x20 at the start of a run of normal characters. These were being treated as
3697 real characters, instead of the source characters being re-checked.
3700 Version 0.94 18-Sep-97
3701 ----------------------
3703 1. The functions are now thread-safe, with the caveat that the global variables
3704 containing pointers to malloc() and free() or alternative functions are the
3705 same for all threads.
3707 2. Get pcre_study() to generate a bitmap of initial characters for non-
3708 anchored patterns when this is possible, and use it if passed to pcre_exec().
3711 Version 0.93 15-Sep-97
3712 ----------------------
3714 1. /(b)|(:+)/ was computing an incorrect first character.
3716 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3717 but not actually doing anything yet.
3719 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3720 as Perl does (e.g. [-az] or [az-]).
3722 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3723 all possible positions.
3725 5. Split up into different modules to avoid including unneeded functions in a
3726 compiled binary. However, compile and exec are still in one module. The "study"
3727 function is split off.
3729 6. The character tables are now in a separate module whose source is generated
3730 by an auxiliary program - but can then be edited by hand if required. There are
3731 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3732 toupper() in the code.
3734 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3735 make them global. Abolish the function for setting them, as the caller can now
3736 set them directly.
3739 Version 0.92 11-Sep-97
3740 ----------------------
3742 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3743 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3745 2. Caseless matching was not working in character classes if the characters in
3746 the pattern were in upper case.
3748 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3750 4. Make PCRE_ANCHORED public and accept as a compile option.
3752 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3753 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3754 pass them.
3756 6. Give an error if bad option bits passed at compile or run time.
3758 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3759 pcretest to cause it to pass that flag.
3761 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3762 options, and the first character, if set.
3764 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3767 Version 0.91 10-Sep-97
3768 ----------------------
3770 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3771 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3773 2. PCRE was looping on encountering an indefinitely repeated back reference to
3774 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3775 Perl does - treats the match as successful.
3777 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5