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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 431 - (show annotations)
Wed Sep 2 09:23:21 2009 UTC (5 years, 11 months ago) by ph10
File size: 170156 byte(s)
Error occurred while calculating annotation data.
Updage ChangeLog for another bug, already fixed by previous patch.
1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.00 ??-???-??
5 ----------------------
6
7 1. The table for translating pcre_compile() error codes into POSIX error codes
8 was out-of-date, and there was no check on the pcre_compile() error code
9 being within the table. This could lead to an OK return being given in
10 error.
11
12 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
13 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
14 in a Windows environment.
15
16 3. The pcregrep --count option prints the count for each file even when it is
17 zero, as does GNU grep. However, pcregrep was also printing all files when
18 --files-with-matches was added. Now, when both options are given, it prints
19 counts only for those files that have at least one match. (GNU grep just
20 prints the file name in this circumstance, but including the count seems
21 more useful - otherwise, why use --count?) Also ensured that the
22 combination -clh just lists non-zero counts, with no names.
23
24 4. The long form of the pcregrep -F option was incorrectly implemented as
25 --fixed_strings instead of --fixed-strings. This is an incompatible change,
26 but it seems right to fix it, and I didn't think it was worth preserving
27 the old behaviour.
28
29 5. The command line items --regex=pattern and --regexp=pattern were not
30 recognized by pcregrep, which required --regex pattern or --regexp pattern
31 (with a space rather than an '='). The man page documented the '=' forms,
32 which are compatible with GNU grep; these now work.
33
34 6. No libpcreposix.pc file was created for pkg-config; there was just
35 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
36
37 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
38 when UCP support is not needed, by modifying the Python script that
39 generates it from Unicode data files. This should not matter if the module
40 is correctly used as a library, but I received one complaint about 50K of
41 unwanted data. My guess is that the person linked everything into his
42 program rather than using a library. Anyway, it does no harm.
43
44 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
45 was a minimum greater than 1 for a wide character in a possessive
46 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
47 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
48 character. Chaos in the form of incorrect output or a compiling loop could
49 result.
50
51 9. The restrictions on what a pattern can contain when partial matching is
52 requested for pcre_exec() have been removed. All patterns can now be
53 partially matched by this function. In addition, if there are at least two
54 slots in the offset vector, the offsets of the first-encountered partial
55 match are set in them when PCRE_ERROR_PARTIAL is returned.
56
57 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
58 synonymous with PCRE_PARTIAL, for backwards compatibility, and
59 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
60 and may be more useful for multi-segment matching, especially with
61 pcre_exec().
62
63 11. Partial matching with pcre_exec() is now more intuitive. A partial match
64 used to be given if ever the end of the subject was reached; now it is
65 given only if matching could not proceed because another character was
66 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
67 string "Z", which now yields "no match" instead of "partial match". In the
68 case of pcre_dfa_exec(), "no match" is given if every matching path for the
69 final character ended with (*FAIL).
70
71 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
72 if the pattern had a "must contain" character that was already found in the
73 earlier partial match, unless partial matching was again requested. For
74 example, with the pattern /dog.(body)?/, the "must contain" character is
75 "g". If the first part-match was for the string "dog", restarting with
76 "sbody" failed.
77
78 13. Added a pcredemo man page, created automatically from the pcredemo.c file,
79 so that the demonstration program is easily available in environments where
80 PCRE has not been installed from source.
81
82 14. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
83 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
84 library.
85
86
87 Version 7.9 11-Apr-09
88 ---------------------
89
90 1. When building with support for bzlib/zlib (pcregrep) and/or readline
91 (pcretest), all targets were linked against these libraries. This included
92 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
93 libraries. This caused unwanted dependencies to be created. This problem
94 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
95 pcretest is linked with readline.
96
97 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
98 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
99 moved outside it again, because FALSE and TRUE are already defined in AIX,
100 but BOOL is not.
101
102 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
103 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
104
105 4. The pcregrep documentation said spaces were inserted as well as colons (or
106 hyphens) following file names and line numbers when outputting matching
107 lines. This is not true; no spaces are inserted. I have also clarified the
108 wording for the --colour (or --color) option.
109
110 5. In pcregrep, when --colour was used with -o, the list of matching strings
111 was not coloured; this is different to GNU grep, so I have changed it to be
112 the same.
113
114 6. When --colo(u)r was used in pcregrep, only the first matching substring in
115 each matching line was coloured. Now it goes on to look for further matches
116 of any of the test patterns, which is the same behaviour as GNU grep.
117
118 7. A pattern that could match an empty string could cause pcregrep to loop; it
119 doesn't make sense to accept an empty string match in pcregrep, so I have
120 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
121 seems to be how GNU grep behaves.
122
123 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
124 start or after a newline", because the conditional assertion was not being
125 correctly handled. The rule now is that both the assertion and what follows
126 in the first alternative must satisfy the test.
127
128 9. If auto-callout was enabled in a pattern with a conditional group whose
129 condition was an assertion, PCRE could crash during matching, both with
130 pcre_exec() and pcre_dfa_exec().
131
132 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
133 used for matching.
134
135 11. Unicode property support in character classes was not working for
136 characters (bytes) greater than 127 when not in UTF-8 mode.
137
138 12. Added the -M command line option to pcretest.
139
140 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
141
142 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
143
144 16. Added comments and documentation about mis-use of no_arg in the C++
145 wrapper.
146
147 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
148 from Martin Jerabek that uses macro names for all relevant character and
149 string constants.
150
151 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
152 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
153 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
154 these, but not everybody uses configure.
155
156 19. A conditional group that had only one branch was not being correctly
157 recognized as an item that could match an empty string. This meant that an
158 enclosing group might also not be so recognized, causing infinite looping
159 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
160 with the subject "ab", where knowledge that the repeated group can match
161 nothing is needed in order to break the loop.
162
163 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
164 exec(), but without supplying a callout function, matching went wrong.
165
166 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
167 leak if the size of the offset vector was greater than 30. When the vector
168 is smaller, the saved offsets during recursion go onto a local stack
169 vector, but for larger vectors malloc() is used. It was failing to free
170 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
171 error, in fact).
172
173 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
174 heapframe that is used only when UTF-8 support is enabled. This caused no
175 problem, but was untidy.
176
177 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
178 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
179 included within another project.
180
181 24. Steven Van Ingelgem's patches to add more options to the CMake support,
182 slightly modified by me:
183
184 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
185 not building pcregrep.
186
187 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
188 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
189
190 25. Forward references, both numeric and by name, in patterns that made use of
191 duplicate group numbers, could behave incorrectly or give incorrect errors,
192 because when scanning forward to find the reference group, PCRE was not
193 taking into account the duplicate group numbers. A pattern such as
194 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
195
196 26. Changed a few more instances of "const unsigned char *" to USPTR, making
197 the feature of a custom pointer more persuasive (as requested by a user).
198
199 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
200 pcretest.c, inside #ifndefs, because it seems they are sometimes already
201 pre-defined.
202
203 28. Added support for (*UTF8) at the start of a pattern.
204
205 29. Arrange for flags added by the "release type" setting in CMake to be shown
206 in the configuration summary.
207
208
209 Version 7.8 05-Sep-08
210 ---------------------
211
212 1. Replaced UCP searching code with optimized version as implemented for Ad
213 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
214 stage table and inline lookup instead of a function, giving speed ups of 2
215 to 5 times on some simple patterns that I tested. Permission was given to
216 distribute the MultiStage2.py script that generates the tables (it's not in
217 the tarball, but is in the Subversion repository).
218
219 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
220 scripts.
221
222 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
223 a group with a zero qualifier. The result of the study could be incorrect,
224 or the function might crash, depending on the pattern.
225
226 4. Caseless matching was not working for non-ASCII characters in back
227 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
228 It now works when Unicode Property Support is available.
229
230 5. In pcretest, an escape such as \x{de} in the data was always generating
231 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
232 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
233 truncation.
234
235 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
236
237 7. Added two (int) casts to pcregrep when printing the difference of two
238 pointers, in case they are 64-bit values.
239
240 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
241 test 2 if it fails.
242
243 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
244 and a #define of that name to empty if it is not externally set. This is to
245 allow users of MSVC to set it if necessary.
246
247 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
248 the convenience functions in the pcre_get.c source file.
249
250 11. An option change at the start of a pattern that had top-level alternatives
251 could cause overwriting and/or a crash. This command provoked a crash in
252 some environments:
253
254 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
255
256 This potential security problem was recorded as CVE-2008-2371.
257
258 12. For a pattern where the match had to start at the beginning or immediately
259 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
260 pcre_dfa_exec() could read past the end of the passed subject if there was
261 no match. To help with detecting such bugs (e.g. with valgrind), I modified
262 pcretest so that it places the subject at the end of its malloc-ed buffer.
263
264 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
265 exec() might read past the end of the data buffer in UTF-8 mode.
266
267 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
268 the data contained the byte 0x85 as part of a UTF-8 character within its
269 first line. This applied both to normal and DFA matching.
270
271 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
272 /^[^d]*?$/8 failed to match "abc".
273
274 16. Added a missing copyright notice to pcrecpp_internal.h.
275
276 17. Make it more clear in the documentation that values returned from
277 pcre_exec() in ovector are byte offsets, not character counts.
278
279 18. Tidied a few places to stop certain compilers from issuing warnings.
280
281 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
282 supplied by Stefan Weber. I made a further small update for 7.8 because
283 there is a change of source arrangements: the pcre_searchfuncs.c module is
284 replaced by pcre_ucd.c.
285
286
287 Version 7.7 07-May-08
288 ---------------------
289
290 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
291 a string to a long long, pretend we don't even have a long long." This is
292 done by checking for the strtoq, strtoll, and _strtoi64 functions.
293
294 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
295 pre-7.6 versions, which defined a global no_arg variable instead of putting
296 it in the RE class. (See also #8 below.)
297
298 3. Remove a line of dead code, identified by coverity and reported by Nuno
299 Lopes.
300
301 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
302
303 (1) The include/exclude patterns were being applied to the whole pathnames
304 of files, instead of just to the final components.
305
306 (2) If there was more than one level of directory, the subdirectories were
307 skipped unless they satisfied the include/exclude conditions. This is
308 inconsistent with GNU grep (and could even be seen as contrary to the
309 pcregrep specification - which I improved to make it absolutely clear).
310 The action now is always to scan all levels of directory, and just
311 apply the include/exclude patterns to regular files.
312
313 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
314 --exclude_dir in the tests to avoid scanning .svn directories.
315
316 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
317 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
318 doesn't support NULs in patterns.
319
320 7. Added some missing "const"s to declarations of static tables in
321 pcre_compile.c and pcre_dfa_exec.c.
322
323 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
324 caused by fix #2 above. (Subsequently also a second patch to fix the
325 first patch. And a third patch - this was a messy problem.)
326
327 9. Applied Craig's patch to remove the use of push_back().
328
329 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
330 matching function regexec().
331
332 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
333 which, however, unlike Perl's \g{...}, are subroutine calls, not back
334 references. PCRE supports relative numbers with this syntax (I don't think
335 Oniguruma does).
336
337 12. Previously, a group with a zero repeat such as (...){0} was completely
338 omitted from the compiled regex. However, this means that if the group
339 was called as a subroutine from elsewhere in the pattern, things went wrong
340 (an internal error was given). Such groups are now left in the compiled
341 pattern, with a new opcode that causes them to be skipped at execution
342 time.
343
344 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
345 to the way PCRE behaves:
346
347 (a) A lone ] character is dis-allowed (Perl treats it as data).
348
349 (b) A back reference to an unmatched subpattern matches an empty string
350 (Perl fails the current match path).
351
352 (c) A data ] in a character class must be notated as \] because if the
353 first data character in a class is ], it defines an empty class. (In
354 Perl it is not possible to have an empty class.) The empty class []
355 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
356 The negative empty class [^] matches any one character, independently
357 of the DOTALL setting.
358
359 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
360 non-existent subpattern following a character class starting with ']' and
361 containing () gave an internal compiling error instead of "reference to
362 non-existent subpattern". Fortunately, when the pattern did exist, the
363 compiled code was correct. (When scanning forwards to check for the
364 existencd of the subpattern, it was treating the data ']' as terminating
365 the class, so got the count wrong. When actually compiling, the reference
366 was subsequently set up correctly.)
367
368 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
369 it was being rejected as not supported by pcre_dfa_exec(), even though
370 other assertions are supported. I have made pcre_dfa_exec() support
371 (*FAIL).
372
373 16. The implementation of 13c above involved the invention of a new opcode,
374 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
375 cannot be changed at match time, I realized I could make a small
376 improvement to matching performance by compiling OP_ALLANY instead of
377 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
378 on the OP_ANY path.
379
380 17. Compiling pcretest on Windows with readline support failed without the
381 following two fixes: (1) Make the unistd.h include conditional on
382 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
383
384 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
385 ncurses library to be included for pcretest when ReadLine support is
386 requested, but also to allow for it to be overridden. This patch came from
387 Daniel Bergström.
388
389 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
390 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
391 any errors with the current Unicode tables. Thanks to Peter Kankowski for
392 spotting this.
393
394
395 Version 7.6 28-Jan-08
396 ---------------------
397
398 1. A character class containing a very large number of characters with
399 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
400 overflow.
401
402 2. Patch to cut out the "long long" test in pcrecpp_unittest when
403 HAVE_LONG_LONG is not defined.
404
405 3. Applied Christian Ehrlicher's patch to update the CMake build files to
406 bring them up to date and include new features. This patch includes:
407
408 - Fixed PH's badly added libz and libbz2 support.
409 - Fixed a problem with static linking.
410 - Added pcredemo. [But later removed - see 7 below.]
411 - Fixed dftables problem and added an option.
412 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
413 HAVE_LONG_LONG.
414 - Added readline support for pcretest.
415 - Added an listing of the option settings after cmake has run.
416
417 4. A user submitted a patch to Makefile that makes it easy to create
418 "pcre.dll" under mingw when using Configure/Make. I added stuff to
419 Makefile.am that cause it to include this special target, without
420 affecting anything else. Note that the same mingw target plus all
421 the other distribution libraries and programs are now supported
422 when configuring with CMake (see 6 below) instead of with
423 Configure/Make.
424
425 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
426 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
427 exported in the Windows port". It has not yet been confirmed that the patch
428 solves the problem, but it does no harm.
429
430 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
431 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
432 with CMake, and also correct the comment about stack recursion.
433
434 7. Remove the automatic building of pcredemo from the ./configure system and
435 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
436 of a program that users should build themselves after PCRE is installed, so
437 building it automatically is not really right. What is more, it gave
438 trouble in some build environments.
439
440 8. Further tidies to CMakeLists.txt from Sheri and Christian.
441
442
443 Version 7.5 10-Jan-08
444 ---------------------
445
446 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
447 values in parens when parsing an RE using the C++ wrapper."
448
449 2. Negative specials like \S did not work in character classes in UTF-8 mode.
450 Characters greater than 255 were excluded from the class instead of being
451 included.
452
453 3. The same bug as (2) above applied to negated POSIX classes such as
454 [:^space:].
455
456 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
457 defined or documented. It seems to have been a typo for PCRE_STATIC, so
458 I have changed it.
459
460 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
461 first named subpattern) and a construct such as (?&a) would reference the
462 first named subpattern whose name started with "a" (in other words, the
463 length check was missing). Both these problems are fixed. "Subpattern name
464 expected" is now given for (?&) (a zero-length name), and this patch also
465 makes it give the same error for \k'' (previously it complained that that
466 was a reference to a non-existent subpattern).
467
468 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
469 this is right because (?- can be followed by option settings as well as by
470 digits. I have, however, made the messages clearer.
471
472 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
473 than the number used in the conditional) now cause a compile-time error.
474 This is actually not compatible with Perl, which accepts such patterns, but
475 treats the conditional as always being FALSE (as PCRE used to), but it
476 seems to me that giving a diagnostic is better.
477
478 8. Change "alphameric" to the more common word "alphanumeric" in comments
479 and messages.
480
481 9. Fix two occurrences of "backslash" in comments that should have been
482 "backspace".
483
484 10. Remove two redundant lines of code that can never be obeyed (their function
485 was moved elsewhere).
486
487 11. The program that makes PCRE's Unicode character property table had a bug
488 which caused it to generate incorrect table entries for sequences of
489 characters that have the same character type, but are in different scripts.
490 It amalgamated them into a single range, with the script of the first of
491 them. In other words, some characters were in the wrong script. There were
492 thirteen such cases, affecting characters in the following ranges:
493
494 U+002b0 - U+002c1
495 U+0060c - U+0060d
496 U+0061e - U+00612
497 U+0064b - U+0065e
498 U+0074d - U+0076d
499 U+01800 - U+01805
500 U+01d00 - U+01d77
501 U+01d9b - U+01dbf
502 U+0200b - U+0200f
503 U+030fc - U+030fe
504 U+03260 - U+0327f
505 U+0fb46 - U+0fbb1
506 U+10450 - U+1049d
507
508 12. The -o option (show only the matching part of a line) for pcregrep was not
509 compatible with GNU grep in that, if there was more than one match in a
510 line, it showed only the first of them. It now behaves in the same way as
511 GNU grep.
512
513 13. If the -o and -v options were combined for pcregrep, it printed a blank
514 line for every non-matching line. GNU grep prints nothing, and pcregrep now
515 does the same. The return code can be used to tell if there were any
516 non-matching lines.
517
518 14. Added --file-offsets and --line-offsets to pcregrep.
519
520 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
521 infinitely looping recursion. The bug was that positive lookaheads were not
522 being skipped when checking for a possible empty match (negative lookaheads
523 and both kinds of lookbehind were skipped).
524
525 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
526 inclusion of <windows.h> to before rather than after the definition of
527 INVALID_FILE_ATTRIBUTES (patch from David Byron).
528
529 17. Specifying a possessive quantifier with a specific limit for a Unicode
530 character property caused pcre_compile() to compile bad code, which led at
531 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
532 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
533 caused the error; without that there was no problem.
534
535 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
536
537 19. Added --enable-pcretest-libreadline.
538
539 20. In pcrecpp.cc, the variable 'count' was incremented twice in
540 RE::GlobalReplace(). As a result, the number of replacements returned was
541 double what it should be. I removed one of the increments, but Craig sent a
542 later patch that removed the other one (the right fix) and added unit tests
543 that check the return values (which was not done before).
544
545 21. Several CMake things:
546
547 (1) Arranged that, when cmake is used on Unix, the libraries end up with
548 the names libpcre and libpcreposix, not just pcre and pcreposix.
549
550 (2) The above change means that pcretest and pcregrep are now correctly
551 linked with the newly-built libraries, not previously installed ones.
552
553 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
554
555 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
556 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
557 UTF-8 newline character). The key issue is that the pattern starts .*;
558 this means that the match must be either at the beginning, or after a
559 newline. The bug was in the code for advancing after a failed match and
560 checking that the new position followed a newline. It was not taking
561 account of UTF-8 characters correctly.
562
563 23. PCRE was behaving differently from Perl in the way it recognized POSIX
564 character classes. PCRE was not treating the sequence [:...:] as a
565 character class unless the ... were all letters. Perl, however, seems to
566 allow any characters between [: and :], though of course it rejects as
567 unknown any "names" that contain non-letters, because all the known class
568 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
569 for example, whereas PCRE did not - it did not recognize a POSIX character
570 class. This seemed a bit dangerous, so the code has been changed to be
571 closer to Perl. The behaviour is not identical to Perl, because PCRE will
572 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
573 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
574 Perl does, and where it didn't before.
575
576 24. Rewrite so as to remove the single use of %n from pcregrep because in some
577 Windows environments %n is disabled by default.
578
579
580 Version 7.4 21-Sep-07
581 ---------------------
582
583 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
584 means that a class such as [\s] counted as "explicit reference to CR or
585 LF". That isn't really right - the whole point of the change was to try to
586 help when there was an actual mention of one of the two characters. So now
587 the change happens only if \r or \n (or a literal CR or LF) character is
588 encountered.
589
590 2. The 32-bit options word was also used for 6 internal flags, but the numbers
591 of both had grown to the point where there were only 3 bits left.
592 Fortunately, there was spare space in the data structure, and so I have
593 moved the internal flags into a new 16-bit field to free up more option
594 bits.
595
596 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
597 but did not set the internal JCHANGED flag - either of these is enough to
598 control the way the "get" function works - but the PCRE_INFO_JCHANGED
599 facility is supposed to tell if (?J) was ever used, so now (?J) at the
600 start sets both bits.
601
602 4. Added options (at build time, compile time, exec time) to change \R from
603 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
604
605 5. doc/pcresyntax.html was missing from the distribution.
606
607 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
608 compatibility, even though it is no longer used.
609
610 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
611 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
612 windows.h file is present (where different names are used). [This was
613 reversed later after testing - see 16 below.]
614
615 8. Changed all #include <config.h> to #include "config.h". There were also
616 some further <pcre.h> cases that I changed to "pcre.h".
617
618 9. When pcregrep was used with the --colour option, it missed the line ending
619 sequence off the lines that it output.
620
621 10. It was pointed out to me that arrays of string pointers cause lots of
622 relocations when a shared library is dynamically loaded. A technique of
623 using a single long string with a table of offsets can drastically reduce
624 these. I have refactored PCRE in four places to do this. The result is
625 dramatic:
626
627 Originally: 290
628 After changing UCP table: 187
629 After changing error message table: 43
630 After changing table of "verbs" 36
631 After changing table of Posix names 22
632
633 Thanks to the folks working on Gregex for glib for this insight.
634
635 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
636 unicode-properties was also set.
637
638 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
639
640 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
641 checked only for CRLF.
642
643 14. Added casts to pcretest.c to avoid compiler warnings.
644
645 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
646
647 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
648 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
649 entirely. This removes changes made in 7 above.
650
651 17. The CMake files have been updated, and there is now more information about
652 building with CMake in the NON-UNIX-USE document.
653
654
655 Version 7.3 28-Aug-07
656 ---------------------
657
658 1. In the rejigging of the build system that eventually resulted in 7.1, the
659 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
660 brackets there is not right, since it causes compilers to look for an
661 installed pcre.h, not the version that is in the source that is being
662 compiled (which of course may be different). I have changed it back to:
663
664 #include "pcre.h"
665
666 I have a vague recollection that the change was concerned with compiling in
667 different directories, but in the new build system, that is taken care of
668 by the VPATH setting the Makefile.
669
670 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
671 when the subject happened to end in the byte 0x85 (e.g. if the last
672 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
673 characters but of course it shouldn't be taken as a newline when it is part
674 of another character. The bug was that, for an unlimited repeat of . in
675 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
676 characters when looking for a newline.
677
678 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
679
680 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
681 in debug output.
682
683 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
684 long printing in the pcrecpp unittest when running under MinGW.
685
686 6. ESC_K was left out of the EBCDIC table.
687
688 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
689 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
690 limit also applies to "virtual nesting" when a pattern is recursive, and in
691 this case 1000 isn't so big. I have been able to remove this limit at the
692 expense of backing off one optimization in certain circumstances. Normally,
693 when pcre_exec() would call its internal match() function recursively and
694 immediately return the result unconditionally, it uses a "tail recursion"
695 feature to save stack. However, when a subpattern that can match an empty
696 string has an unlimited repetition quantifier, it no longer makes this
697 optimization. That gives it a stack frame in which to save the data for
698 checking that an empty string has been matched. Previously this was taken
699 from the 1000-entry workspace that had been reserved. So now there is no
700 explicit limit, but more stack is used.
701
702 8. Applied Daniel's patches to solve problems with the import/export magic
703 syntax that is required for Windows, and which was going wrong for the
704 pcreposix and pcrecpp parts of the library. These were overlooked when this
705 problem was solved for the main library.
706
707 9. There were some crude static tests to avoid integer overflow when computing
708 the size of patterns that contain repeated groups with explicit upper
709 limits. As the maximum quantifier is 65535, the maximum group length was
710 set at 30,000 so that the product of these two numbers did not overflow a
711 32-bit integer. However, it turns out that people want to use groups that
712 are longer than 30,000 bytes (though not repeat them that many times).
713 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
714 made it possible to implement the integer overflow checks in a much more
715 dynamic way, which I have now done. The artificial limitation on group
716 length has been removed - we now have only the limit on the total length of
717 the compiled pattern, which depends on the LINK_SIZE setting.
718
719 10. Fixed a bug in the documentation for get/copy named substring when
720 duplicate names are permitted. If none of the named substrings are set, the
721 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
722 empty string.
723
724 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
725 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
726 because the ] is interpreted as the first data character and the
727 terminating ] is not found. PCRE has been made compatible with Perl in this
728 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
729 cause memory overwriting.
730
731 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
732 string has been matched (to stop an infinite loop). It was not recognizing
733 a conditional subpattern that could match an empty string if that
734 subpattern was within another subpattern. For example, it looped when
735 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
736 condition was not nested. This bug has been fixed.
737
738 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
739 past the start of the subject in the presence of bytes with the top bit
740 set, for example "\x8aBCD".
741
742 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
743 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
744
745 14. Optimized (?!) to (*FAIL).
746
747 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
748 This restricts code points to be within the range 0 to 0x10FFFF, excluding
749 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
750 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
751 does: it's just the validity check that is more restrictive.
752
753 16. Inserted checks for integer overflows during escape sequence (backslash)
754 processing, and also fixed erroneous offset values for syntax errors during
755 backslash processing.
756
757 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
758 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
759
760 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
761 caused an overrun.
762
763 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
764 something other than just ASCII characters) inside a group that had an
765 unlimited repeat caused a loop at compile time (while checking to see
766 whether the group could match an empty string).
767
768 20. Debugging a pattern containing \p or \P could cause a crash. For example,
769 [\P{Any}] did so. (Error in the code for printing property names.)
770
771 21. An orphan \E inside a character class could cause a crash.
772
773 22. A repeated capturing bracket such as (A)? could cause a wild memory
774 reference during compilation.
775
776 23. There are several functions in pcre_compile() that scan along a compiled
777 expression for various reasons (e.g. to see if it's fixed length for look
778 behind). There were bugs in these functions when a repeated \p or \P was
779 present in the pattern. These operators have additional parameters compared
780 with \d, etc, and these were not being taken into account when moving along
781 the compiled data. Specifically:
782
783 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
784 length.
785
786 (b) An item such as \pL+ within a repeated group could cause crashes or
787 loops.
788
789 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
790 "reference to non-existent subpattern" error.
791
792 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
793
794 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
795 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
796
797 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
798
799 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
800 character were causing crashes (broken optimization).
801
802 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
803 \p or \P) caused a compile-time loop.
804
805 28. More problems have arisen in unanchored patterns when CRLF is a valid line
806 break. For example, the unstudied pattern [\r\n]A does not match the string
807 "\r\nA" because change 7.0/46 below moves the current point on by two
808 characters after failing to match at the start. However, the pattern \nA
809 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
810 the same is true. There doesn't seem any very clean way out of this, but
811 what I have chosen to do makes the common cases work: PCRE now takes note
812 of whether there can be an explicit match for \r or \n anywhere in the
813 pattern, and if so, 7.0/46 no longer applies. As part of this change,
814 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
815 pattern has explicit CR or LF references.
816
817 29. Added (*CR) etc for changing newline setting at start of pattern.
818
819
820 Version 7.2 19-Jun-07
821 ---------------------
822
823 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
824 which is apparently normally available under Windows.
825
826 2. Re-jig the pcregrep tests with different newline settings in an attempt
827 to make them independent of the local environment's newline setting.
828
829 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
830
831 4. Some of the "internals" tests were previously cut out when the link size
832 was not 2, because the output contained actual offsets. The recent new
833 "Z" feature of pcretest means that these can be cut out, making the tests
834 usable with all link sizes.
835
836 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
837 stack recursion. This gives a massive performance boost under BSD, but just
838 a small improvement under Linux. However, it saves one field in the frame
839 in all cases.
840
841 6. Added more features from the forthcoming Perl 5.10:
842
843 (a) (?-n) (where n is a string of digits) is a relative subroutine or
844 recursion call. It refers to the nth most recently opened parentheses.
845
846 (b) (?+n) is also a relative subroutine call; it refers to the nth next
847 to be opened parentheses.
848
849 (c) Conditions that refer to capturing parentheses can be specified
850 relatively, for example, (?(-2)... or (?(+3)...
851
852 (d) \K resets the start of the current match so that everything before
853 is not part of it.
854
855 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
856
857 (f) \g{name} is another synonym - part of Perl 5.10's unification of
858 reference syntax.
859
860 (g) (?| introduces a group in which the numbering of parentheses in each
861 alternative starts with the same number.
862
863 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
864
865 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
866 PCRE_INFO_JCHANGED.
867
868 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
869 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
870 for detecting groups that can match an empty string.
871
872 9. A pattern with a very large number of alternatives (more than several
873 hundred) was running out of internal workspace during the pre-compile
874 phase, where pcre_compile() figures out how much memory will be needed. A
875 bit of new cunning has reduced the workspace needed for groups with
876 alternatives. The 1000-alternative test pattern now uses 12 bytes of
877 workspace instead of running out of the 4096 that are available.
878
879 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
880
881 11. Applied patch from Google to remove an optimization that didn't quite work.
882 The report of the bug said:
883
884 pcrecpp::RE("a*").FullMatch("aaa") matches, while
885 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
886 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
887
888 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
889 it matched the wrong number of bytes.
890
891
892 Version 7.1 24-Apr-07
893 ---------------------
894
895 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
896 that is more "standard", making use of automake and other Autotools. There
897 is some re-arrangement of the files and adjustment of comments consequent
898 on this.
899
900 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
901 for recursive directory scanning broke on some systems because the files
902 are not scanned in any specific order and on different systems the order
903 was different. A call to "sort" has been inserted into RunGrepTest for the
904 approprate test as a short-term fix. In the longer term there may be an
905 alternative.
906
907 3. I had an email from Eric Raymond about problems translating some of PCRE's
908 man pages to HTML (despite the fact that I distribute HTML pages, some
909 people do their own conversions for various reasons). The problems
910 concerned the use of low-level troff macros .br and .in. I have therefore
911 removed all such uses from the man pages (some were redundant, some could
912 be replaced by .nf/.fi pairs). The 132html script that I use to generate
913 HTML has been updated to handle .nf/.fi and to complain if it encounters
914 .br or .in.
915
916 4. Updated comments in configure.ac that get placed in config.h.in and also
917 arranged for config.h to be included in the distribution, with the name
918 config.h.generic, for the benefit of those who have to compile without
919 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
920
921 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
922 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
923 makevp.bat for the current PCRE, using the additional files
924 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
925
926 6. A Windows user reported a minor discrepancy with test 2, which turned out
927 to be caused by a trailing space on an input line that had got lost in his
928 copy. The trailing space was an accident, so I've just removed it.
929
930 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
931 that is needed.
932
933 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
934 as "const" (a) because they are and (b) because it helps the PHP
935 maintainers who have recently made a script to detect big data structures
936 in the php code that should be moved to the .rodata section. I remembered
937 to update Builducptable as well, so it won't revert if ucptable.h is ever
938 re-created.
939
940 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
941 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
942 order to be able to cut out the UTF-8 tables in the latter when UTF-8
943 support is not required. This saves 1.5-2K of code, which is important in
944 some applications.
945
946 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
947 so as not to refer to the tables, even though these functions will never be
948 called when UTF-8 support is disabled. Otherwise there are problems with a
949 shared library.
950
951 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
952
953 (a) It was defining its arguments as char * instead of void *.
954
955 (b) It was assuming that all moves were upwards in memory; this was true
956 a long time ago when I wrote it, but is no longer the case.
957
958 The emulated memove() is provided for those environments that have neither
959 memmove() nor bcopy(). I didn't think anyone used it these days, but that
960 is clearly not the case, as these two bugs were recently reported.
961
962 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
963 and Detrail to create the HTML documentation, the .txt form of the man
964 pages, and it removes trailing spaces from listed files. It also creates
965 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
966 case, it wraps all the #defines with #ifndefs. This script should be run
967 before "make dist".
968
969 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
970 with Unicode property support.
971
972 (a) For a maximizing quantifier, if the two different cases of the
973 character were of different lengths in their UTF-8 codings (there are
974 some cases like this - I found 11), and the matching function had to
975 back up over a mixture of the two cases, it incorrectly assumed they
976 were both the same length.
977
978 (b) When PCRE was configured to use the heap rather than the stack for
979 recursion during matching, it was not correctly preserving the data for
980 the other case of a UTF-8 character when checking ahead for a match
981 while processing a minimizing repeat. If the check also involved
982 matching a wide character, but failed, corruption could cause an
983 erroneous result when trying to check for a repeat of the original
984 character.
985
986 13. Some tidying changes to the testing mechanism:
987
988 (a) The RunTest script now detects the internal link size and whether there
989 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
990 values substituted by "configure". (The RunGrepTest script already did
991 this for UTF-8.) The configure.ac script no longer substitutes the
992 relevant variables.
993
994 (b) The debugging options /B and /D in pcretest show the compiled bytecode
995 with length and offset values. This means that the output is different
996 for different internal link sizes. Test 2 is skipped for link sizes
997 other than 2 because of this, bypassing the problem. Unfortunately,
998 there was also a test in test 3 (the locale tests) that used /B and
999 failed for link sizes other than 2. Rather than cut the whole test out,
1000 I have added a new /Z option to pcretest that replaces the length and
1001 offset values with spaces. This is now used to make test 3 independent
1002 of link size. (Test 2 will be tidied up later.)
1003
1004 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1005 segmentation fault instead of returning the appropriate error message.
1006
1007 15. In multiline mode when the newline sequence was set to "any", the pattern
1008 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1009 This doesn't seem right; it now treats the CRLF combination as the line
1010 ending, and so does not match in that case. It's only a pattern such as ^$
1011 that would hit this one: something like ^ABC$ would have failed after \r
1012 and then tried again after \r\n.
1013
1014 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1015 in an attempt to make files that differ only in their line terminators
1016 compare equal. This works on Linux.
1017
1018 17. Under certain error circumstances pcregrep might try to free random memory
1019 as it exited. This is now fixed, thanks to valgrind.
1020
1021 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1022 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1023 was because its rules for how to advance for /g after matching an empty
1024 string at the end of a line did not allow for this case. They now check for
1025 it specially.
1026
1027 20. pcretest is supposed to handle patterns and data of any length, by
1028 extending its buffers when necessary. It was getting this wrong when the
1029 buffer for a data line had to be extended.
1030
1031 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1032 CRLF as a newline sequence.
1033
1034 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1035 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1036 I have nevertheless tidied it up.
1037
1038 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1039
1040 24. Added a man page for pcre-config.
1041
1042
1043 Version 7.0 19-Dec-06
1044 ---------------------
1045
1046 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1047 moving to gcc 4.1.1.
1048
1049 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1050 sys/time.h, which is documented as needed for this function. It doesn't
1051 seem to matter on Linux, but it showed up on some releases of OS X.
1052
1053 3. It seems that there are systems where bytes whose values are greater than
1054 127 match isprint() in the "C" locale. The "C" locale should be the
1055 default when a C program starts up. In most systems, only ASCII printing
1056 characters match isprint(). This difference caused the output from pcretest
1057 to vary, making some of the tests fail. I have changed pcretest so that:
1058
1059 (a) When it is outputting text in the compiled version of a pattern, bytes
1060 other than 32-126 are always shown as hex escapes.
1061
1062 (b) When it is outputting text that is a matched part of a subject string,
1063 it does the same, unless a different locale has been set for the match
1064 (using the /L modifier). In this case, it uses isprint() to decide.
1065
1066 4. Fixed a major bug that caused incorrect computation of the amount of memory
1067 required for a compiled pattern when options that changed within the
1068 pattern affected the logic of the preliminary scan that determines the
1069 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1070 that the computed length was too small. The symptoms of this bug were
1071 either the PCRE error "internal error: code overflow" from pcre_compile(),
1072 or a glibc crash with a message such as "pcretest: free(): invalid next
1073 size (fast)". Examples of patterns that provoked this bug (shown in
1074 pcretest format) are:
1075
1076 /(?-x: )/x
1077 /(?x)(?-x: \s*#\s*)/
1078 /((?i)[\x{c0}])/8
1079 /(?i:[\x{c0}])/8
1080
1081 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1082 is now done differently.
1083
1084 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1085 wrapper classes; (b) implement a new function in the C++ scanner that is
1086 more efficient than the old way of doing things because it avoids levels of
1087 recursion in the regex matching; (c) add a paragraph to the documentation
1088 for the FullMatch() function.
1089
1090 6. The escape sequence \n was being treated as whatever was defined as
1091 "newline". Not only was this contrary to the documentation, which states
1092 that \n is character 10 (hex 0A), but it also went horribly wrong when
1093 "newline" was defined as CRLF. This has been fixed.
1094
1095 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1096 was being set to -1 for the "end of line" case (supposedly a value that no
1097 character can have). Though this value is never used (the check for end of
1098 line is "zero bytes in current character"), it caused compiler complaints.
1099 I've changed it to 0xffffffff.
1100
1101 8. In pcre_version.c, the version string was being built by a sequence of
1102 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1103 string (as it is for production releases) called a macro with an empty
1104 argument. The C standard says the result of this is undefined. The gcc
1105 compiler treats it as an empty string (which was what was wanted) but it is
1106 reported that Visual C gives an error. The source has been hacked around to
1107 avoid this problem.
1108
1109 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1110 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1111 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1112 of them did).
1113
1114 10. Originally, pcretest opened its input and output without "b"; then I was
1115 told that "b" was needed in some environments, so it was added for release
1116 5.0 to both the input and output. (It makes no difference on Unix-like
1117 systems.) Later I was told that it is wrong for the input on Windows. I've
1118 now abstracted the modes into two macros, to make it easier to fiddle with
1119 them, and removed "b" from the input mode under Windows.
1120
1121 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1122
1123 12. Added -help and --help to pcretest as an official way of being reminded
1124 of the options.
1125
1126 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1127 and pcrecpp.cc because they annoy compilers at high warning levels.
1128
1129 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1130
1131 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1132 scripts are not C programs :-) and which was not noticed because it works
1133 on Linux.
1134
1135 16. pcretest is supposed to handle any length of pattern and data line (as one
1136 line or as a continued sequence of lines) by extending its input buffer if
1137 necessary. This feature was broken for very long pattern lines, leading to
1138 a string of junk being passed to pcre_compile() if the pattern was longer
1139 than about 50K.
1140
1141 17. I have done a major re-factoring of the way pcre_compile() computes the
1142 amount of memory needed for a compiled pattern. Previously, there was code
1143 that made a preliminary scan of the pattern in order to do this. That was
1144 OK when PCRE was new, but as the facilities have expanded, it has become
1145 harder and harder to keep it in step with the real compile phase, and there
1146 have been a number of bugs (see for example, 4 above). I have now found a
1147 cunning way of running the real compile function in a "fake" mode that
1148 enables it to compute how much memory it would need, while actually only
1149 ever using a few hundred bytes of working memory and without too many
1150 tests of the mode. This should make future maintenance and development
1151 easier. A side effect of this work is that the limit of 200 on the nesting
1152 depth of parentheses has been removed (though this was never a serious
1153 limitation, I suspect). However, there is a downside: pcre_compile() now
1154 runs more slowly than before (30% or more, depending on the pattern). I
1155 hope this isn't a big issue. There is no effect on runtime performance.
1156
1157 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1158 newline (only possible for the last line of a file) and it was a
1159 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1160
1161 19. Added additional timing features to pcretest. (1) The -tm option now times
1162 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1163 separate command line item, by a number that specifies the number of
1164 repeats to use when timing. The default is 50000; this gives better
1165 precision, but takes uncomfortably long for very large patterns.
1166
1167 20. Extended pcre_study() to be more clever in cases where a branch of a
1168 subpattern has no definite first character. For example, (a*|b*)[cd] would
1169 previously give no result from pcre_study(). Now it recognizes that the
1170 first character must be a, b, c, or d.
1171
1172 21. There was an incorrect error "recursive call could loop indefinitely" if
1173 a subpattern (or the entire pattern) that was being tested for matching an
1174 empty string contained only one non-empty item after a nested subpattern.
1175 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1176 incorrectly, because the \d was being skipped in the check.
1177
1178 22. The pcretest program now has a new pattern option /B and a command line
1179 option -b, which is equivalent to adding /B to every pattern. This causes
1180 it to show the compiled bytecode, without the additional information that
1181 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1182 is the same as /B/I).
1183
1184 23. A new optimization is now able automatically to treat some sequences such
1185 as a*b as a*+b. More specifically, if something simple (such as a character
1186 or a simple class like \d) has an unlimited quantifier, and is followed by
1187 something that cannot possibly match the quantified thing, the quantifier
1188 is automatically "possessified".
1189
1190 24. A recursive reference to a subpattern whose number was greater than 39
1191 went wrong under certain circumstances in UTF-8 mode. This bug could also
1192 have affected the operation of pcre_study().
1193
1194 25. Realized that a little bit of performance could be had by replacing
1195 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1196
1197 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1198
1199 27. Possessive quantifiers such as a++ were previously implemented by turning
1200 them into atomic groups such as ($>a+). Now they have their own opcodes,
1201 which improves performance. This includes the automatically created ones
1202 from 23 above.
1203
1204 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1205 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1206 the first matched character to be a colon. This applied both to named and
1207 numbered groups.
1208
1209 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1210
1211 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1212 building PCRE on Windows easier, so I have included it in the distribution.
1213
1214 31. There is now a check in pcretest against a ridiculously large number being
1215 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1216 loop, the loop is abandoned.
1217
1218 32. Forward references to subpatterns in conditions such as (?(2)...) where
1219 subpattern 2 is defined later cause pcre_compile() to search forwards in
1220 the pattern for the relevant set of parentheses. This search went wrong
1221 when there were unescaped parentheses in a character class, parentheses
1222 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1223
1224 33. "Subroutine" calls and backreferences were previously restricted to
1225 referencing subpatterns earlier in the regex. This restriction has now
1226 been removed.
1227
1228 34. Added a number of extra features that are going to be in Perl 5.10. On the
1229 whole, these are just syntactic alternatives for features that PCRE had
1230 previously implemented using the Python syntax or my own invention. The
1231 other formats are all retained for compatibility.
1232
1233 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1234 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1235 also .NET compatible.
1236
1237 (b) A recursion or subroutine call to a named group can now be defined as
1238 (?&name) as well as (?P>name).
1239
1240 (c) A backreference to a named group can now be defined as \k<name> or
1241 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1242 5.10, are also .NET compatible.
1243
1244 (d) A conditional reference to a named group can now use the syntax
1245 (?(<name>) or (?('name') as well as (?(name).
1246
1247 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1248 groups (named and numbered) that are never evaluated inline, but can be
1249 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1250 is always false. There may be only one alternative in such a group.
1251
1252 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1253 as the simple (?(R). The condition is true only if the most recent
1254 recursion is that of the given number or name. It does not search out
1255 through the entire recursion stack.
1256
1257 (g) The escape \gN or \g{N} has been added, where N is a positive or
1258 negative number, specifying an absolute or relative reference.
1259
1260 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1261 some "unreachable code" warnings.
1262
1263 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1264 things, this adds five new scripts.
1265
1266 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1267 There were also incompatibilities regarding the handling of \Q..\E inside
1268 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1269 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1270
1271 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1272 matches an empty string, and forcibly breaks the loop. There were bugs in
1273 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1274 against aaaa the result was just "a" rather than "aaaa", for example. Two
1275 separate and independent bugs (that affected different cases) have been
1276 fixed.
1277
1278 39. Refactored the code to abolish the use of different opcodes for small
1279 capturing bracket numbers. This is a tidy that I avoided doing when I
1280 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1281 The new approach is not only tidier, it makes it possible to reduce the
1282 memory needed to fix the previous bug (38).
1283
1284 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1285 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1286 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1287 mode.
1288
1289 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1290 report.
1291
1292 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1293 copy construction and assignment in the C++ wrapper.
1294
1295 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1296 couple of bugs that could have given wrong results in the "--newline=crlf"
1297 case.
1298
1299 44. Added a number of casts and did some reorganization of signed/unsigned int
1300 variables following suggestions from Dair Grant. Also renamed the variable
1301 "this" as "item" because it is a C++ keyword.
1302
1303 45. Arranged for dftables to add
1304
1305 #include "pcre_internal.h"
1306
1307 to pcre_chartables.c because without it, gcc 4.x may remove the array
1308 definition from the final binary if PCRE is built into a static library and
1309 dead code stripping is activated.
1310
1311 46. For an unanchored pattern, if a match attempt fails at the start of a
1312 newline sequence, and the newline setting is CRLF or ANY, and the next two
1313 characters are CRLF, advance by two characters instead of one.
1314
1315
1316 Version 6.7 04-Jul-06
1317 ---------------------
1318
1319 1. In order to handle tests when input lines are enormously long, pcretest has
1320 been re-factored so that it automatically extends its buffers when
1321 necessary. The code is crude, but this _is_ just a test program. The
1322 default size has been increased from 32K to 50K.
1323
1324 2. The code in pcre_study() was using the value of the re argument before
1325 testing it for NULL. (Of course, in any sensible call of the function, it
1326 won't be NULL.)
1327
1328 3. The memmove() emulation function in pcre_internal.h, which is used on
1329 systems that lack both memmove() and bcopy() - that is, hardly ever -
1330 was missing a "static" storage class specifier.
1331
1332 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1333 containing an extended class (one that cannot be represented by a bitmap
1334 because it contains high-valued characters or Unicode property items, e.g.
1335 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1336 pattern, but PCRE should not loop if you do not (it no longer does).
1337 [Detail: two cases were found: (a) a repeated subpattern containing an
1338 extended class; (b) a recursive reference to a subpattern that followed a
1339 previous extended class. It wasn't skipping over the extended class
1340 correctly when UTF-8 mode was not set.]
1341
1342 5. A negated single-character class was not being recognized as fixed-length
1343 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1344 compile error "lookbehind assertion is not fixed length".
1345
1346 6. The RunPerlTest auxiliary script was showing an unexpected difference
1347 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1348 write a Perl script that can interpret lines of an input file either as
1349 byte characters or as UTF-8, which is what "perltest" was being required to
1350 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1351 can't do is switch easily at run time between having the "use utf8;" pragma
1352 or not. In the end, I fudged it by using the RunPerlTest script to insert
1353 "use utf8;" explicitly for the UTF-8 tests.
1354
1355 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1356 the end of the subject string, contrary to the documentation and to what
1357 Perl does. This was true of both matching functions. Now it matches only at
1358 the start of the subject and immediately after *internal* newlines.
1359
1360 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1361 a pointer to an int instead of a pointer to an unsigned long int. This
1362 caused problems on 64-bit systems.
1363
1364 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1365 instance of the 'standard' template library not being so standard".
1366
1367 10. There was no check on the number of named subpatterns nor the maximum
1368 length of a subpattern name. The product of these values is used to compute
1369 the size of the memory block for a compiled pattern. By supplying a very
1370 long subpattern name and a large number of named subpatterns, the size
1371 computation could be caused to overflow. This is now prevented by limiting
1372 the length of names to 32 characters, and the number of named subpatterns
1373 to 10,000.
1374
1375 11. Subpatterns that are repeated with specific counts have to be replicated in
1376 the compiled pattern. The size of memory for this was computed from the
1377 length of the subpattern and the repeat count. The latter is limited to
1378 65535, but there was no limit on the former, meaning that integer overflow
1379 could in principle occur. The compiled length of a repeated subpattern is
1380 now limited to 30,000 bytes in order to prevent this.
1381
1382 12. Added the optional facility to have named substrings with the same name.
1383
1384 13. Added the ability to use a named substring as a condition, using the
1385 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1386 are numbers (not recommended). Forward references are permitted.
1387
1388 14. Added forward references in named backreferences (if you see what I mean).
1389
1390 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1391 pattern could run off the end of the subject. For example, the pattern
1392 "(?s)(.{1,5})"8 did this with the subject "ab".
1393
1394 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1395 PCRE_CASELESS was set when matching characters that were quantified with ?
1396 or *.
1397
1398 17. A character class other than a single negated character that had a minimum
1399 but no maximum quantifier - for example [ab]{6,} - was not handled
1400 correctly by pce_dfa_exec(). It would match only one character.
1401
1402 18. A valid (though odd) pattern that looked like a POSIX character
1403 class but used an invalid character after [ (for example [[,abc,]]) caused
1404 pcre_compile() to give the error "Failed: internal error: code overflow" or
1405 in some cases to crash with a glibc free() error. This could even happen if
1406 the pattern terminated after [[ but there just happened to be a sequence of
1407 letters, a binary zero, and a closing ] in the memory that followed.
1408
1409 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1410 over the years. Originally (before any Unicode support), just the bottom 8
1411 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1412 output from "man perlunicode" includes this:
1413
1414 The regular expression compiler produces polymorphic opcodes. That
1415 is, the pattern adapts to the data and automatically switches to
1416 the Unicode character scheme when presented with Unicode data--or
1417 instead uses a traditional byte scheme when presented with byte
1418 data.
1419
1420 Sadly, a wide octal escape does not cause a switch, and in a string with
1421 no other multibyte characters, these octal escapes are treated as before.
1422 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1423 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1424 Unicode string.
1425
1426 I have not perpetrated such confusion in PCRE. Up till now, it took just
1427 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1428 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1429 translate to the appropriate multibyte character.
1430
1431 29. Applied some refactoring to reduce the number of warnings from Microsoft
1432 and Borland compilers. This has included removing the fudge introduced
1433 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1434 a warning about an unused variable.
1435
1436 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1437 characters since release 4.0, because Perl (from release 5.004) does not.
1438 [Or at least, is documented not to: some releases seem to be in conflict
1439 with the documentation.] However, when a pattern was studied with
1440 pcre_study() and all its branches started with \s, PCRE still included VT
1441 as a possible starting character. Of course, this did no harm; it just
1442 caused an unnecessary match attempt.
1443
1444 22. Removed a now-redundant internal flag bit that recorded the fact that case
1445 dependency changed within the pattern. This was once needed for "required
1446 byte" processing, but is no longer used. This recovers a now-scarce options
1447 bit. Also moved the least significant internal flag bit to the most-
1448 significant bit of the word, which was not previously used (hangover from
1449 the days when it was an int rather than a uint) to free up another bit for
1450 the future.
1451
1452 23. Added support for CRLF line endings as well as CR and LF. As well as the
1453 default being selectable at build time, it can now be changed at runtime
1454 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1455 specify that it is scanning data with non-default line endings.
1456
1457 24. Changed the definition of CXXLINK to make it agree with the definition of
1458 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1459
1460 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1461 recursions. This makes a big different to stack usage for some patterns.
1462
1463 26. If a subpattern containing a named recursion or subroutine reference such
1464 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1465 the space required for the compiled pattern went wrong and gave too small a
1466 value. Depending on the environment, this could lead to "Failed: internal
1467 error: code overflow at offset 49" or "glibc detected double free or
1468 corruption" errors.
1469
1470 27. Applied patches from Google (a) to support the new newline modes and (b) to
1471 advance over multibyte UTF-8 characters in GlobalReplace.
1472
1473 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1474 difference for some implementation of PCRE in some Windows version.
1475
1476 29. Added some extra testing facilities to pcretest:
1477
1478 \q<number> in a data line sets the "match limit" value
1479 \Q<number> in a data line sets the "match recursion limt" value
1480 -S <number> sets the stack size, where <number> is in megabytes
1481
1482 The -S option isn't available for Windows.
1483
1484
1485 Version 6.6 06-Feb-06
1486 ---------------------
1487
1488 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1489 in pcreposix.h. I have copied the definition from pcre.h.
1490
1491 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1492 because pcre.h is no longer a built file.
1493
1494 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1495 not normally included in the compiled code.
1496
1497
1498 Version 6.5 01-Feb-06
1499 ---------------------
1500
1501 1. When using the partial match feature with pcre_dfa_exec(), it was not
1502 anchoring the second and subsequent partial matches at the new starting
1503 point. This could lead to incorrect results. For example, with the pattern
1504 /1234/, partially matching against "123" and then "a4" gave a match.
1505
1506 2. Changes to pcregrep:
1507
1508 (a) All non-match returns from pcre_exec() were being treated as failures
1509 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1510 error message is output. Some extra information is given for the
1511 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1512 probably the only errors that are likely to be caused by users (by
1513 specifying a regex that has nested indefinite repeats, for instance).
1514 If there are more than 20 of these errors, pcregrep is abandoned.
1515
1516 (b) A binary zero was treated as data while matching, but terminated the
1517 output line if it was written out. This has been fixed: binary zeroes
1518 are now no different to any other data bytes.
1519
1520 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1521 used to set a locale for matching. The --locale=xxxx long option has
1522 been added (no short equivalent) to specify a locale explicitly on the
1523 pcregrep command, overriding the environment variables.
1524
1525 (d) When -B was used with -n, some line numbers in the output were one less
1526 than they should have been.
1527
1528 (e) Added the -o (--only-matching) option.
1529
1530 (f) If -A or -C was used with -c (count only), some lines of context were
1531 accidentally printed for the final match.
1532
1533 (g) Added the -H (--with-filename) option.
1534
1535 (h) The combination of options -rh failed to suppress file names for files
1536 that were found from directory arguments.
1537
1538 (i) Added the -D (--devices) and -d (--directories) options.
1539
1540 (j) Added the -F (--fixed-strings) option.
1541
1542 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1543
1544 (l) Added the --colo(u)r option.
1545
1546 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1547 is not present by default.
1548
1549 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1550 items such as (?R) or (?1), when the recursion could match a number of
1551 alternatives. If it matched one of the alternatives, but subsequently,
1552 outside the recursion, there was a failure, the code tried to back up into
1553 the recursion. However, because of the way PCRE is implemented, this is not
1554 possible, and the result was an incorrect result from the match.
1555
1556 In order to prevent this happening, the specification of recursion has
1557 been changed so that all such subpatterns are automatically treated as
1558 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1559
1560 4. I had overlooked the fact that, in some locales, there are characters for
1561 which isalpha() is true but neither isupper() nor islower() are true. In
1562 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1563 and ordfeminine) are like this. This affected the treatment of \w and \W
1564 when they appeared in character classes, but not when they appeared outside
1565 a character class. The bit map for "word" characters is now created
1566 separately from the results of isalnum() instead of just taking it from the
1567 upper, lower, and digit maps. (Plus the underscore character, of course.)
1568
1569 5. The above bug also affected the handling of POSIX character classes such as
1570 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1571 permanent tables. Instead, the bit maps for such a class were previously
1572 created as the appropriate unions of the upper, lower, and digit bitmaps.
1573 Now they are created by subtraction from the [[:word:]] class, which has
1574 its own bitmap.
1575
1576 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1577 It is created by subtracting the vertical space characters (\x09, \x0a,
1578 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1579 subtraction was done in the overall bitmap for a character class, meaning
1580 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1581 be recognized. This bug has been fixed.
1582
1583 7. Patches from the folks at Google:
1584
1585 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1586 real life, but is still worth protecting against".
1587
1588 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1589 regular expressions".
1590
1591 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1592 have it.
1593
1594 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1595 "configure" and the latter not, in order to fix a problem somebody had
1596 with compiling the Arg class on HP-UX.
1597
1598 (e) Improve the error-handling of the C++ wrapper a little bit.
1599
1600 (f) New tests for checking recursion limiting.
1601
1602 8. The pcre_memmove() function, which is used only if the environment does not
1603 have a standard memmove() function (and is therefore rarely compiled),
1604 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1605 returning a result (though PCRE never actually uses the result).
1606
1607 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1608 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1609 returned instead of calling malloc() with an overflowing number that would
1610 most likely cause subsequent chaos.
1611
1612 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1613
1614 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1615 with this option is matched, the nmatch and pmatch options of regexec() are
1616 ignored.
1617
1618 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1619 provided in case anyone wants to the the POSIX interface with UTF-8
1620 strings.
1621
1622 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1623 C++ linking (needed for some HP-UX environments).
1624
1625 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1626 (unused parameter) and in the pcre_printint() function (omitted "default"
1627 switch label when the default is to do nothing).
1628
1629 15. Added some code to make it possible, when PCRE is compiled as a C++
1630 library, to replace subject pointers for pcre_exec() with a smart pointer
1631 class, thus making it possible to process discontinuous strings.
1632
1633 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1634 much the same function. They were added by different people who were trying
1635 to make PCRE easy to compile on non-Unix systems. It has been suggested
1636 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1637 for compiling on Windows systems. I have therefore replaced it with
1638 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1639 defaults to "extern" for C or "extern C" for C++, which works fine on
1640 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1641 SCOPE with something explicit in config.h. In addition:
1642
1643 (a) pcreposix.h still had just "extern" instead of either of these macros;
1644 I have replaced it with PCRE_DATA_SCOPE.
1645
1646 (b) Functions such as _pcre_xclass(), which are internal to the library,
1647 but external in the C sense, all had PCRE_EXPORT in their definitions.
1648 This is apparently wrong for the Windows case, so I have removed it.
1649 (It makes no difference on Unix-like systems.)
1650
1651 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1652 of recursive calls to match(). This is different to MATCH_LIMIT because
1653 that limits the total number of calls to match(), not all of which increase
1654 the depth of recursion. Limiting the recursion depth limits the amount of
1655 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1656 when PCRE is compiled, and changed at run time. A patch from Google adds
1657 this functionality to the C++ interface.
1658
1659 18. Changes to the handling of Unicode character properties:
1660
1661 (a) Updated the table to Unicode 4.1.0.
1662
1663 (b) Recognize characters that are not in the table as "Cn" (undefined).
1664
1665 (c) I revised the way the table is implemented to a much improved format
1666 which includes recognition of ranges. It now supports the ranges that
1667 are defined in UnicodeData.txt, and it also amalgamates other
1668 characters into ranges. This has reduced the number of entries in the
1669 table from around 16,000 to around 3,000, thus reducing its size
1670 considerably. I realized I did not need to use a tree structure after
1671 all - a binary chop search is just as efficient. Having reduced the
1672 number of entries, I extended their size from 6 bytes to 8 bytes to
1673 allow for more data.
1674
1675 (d) Added support for Unicode script names via properties such as \p{Han}.
1676
1677 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1678 matching that character.
1679
1680 20. When matching a repeated Unicode property with a minimum greater than zero,
1681 (for example \pL{2,}), PCRE could look past the end of the subject if it
1682 reached it while seeking the minimum number of characters. This could
1683 happen only if some of the characters were more than one byte long, because
1684 there is a check for at least the minimum number of bytes.
1685
1686 21. Refactored the implementation of \p and \P so as to be more general, to
1687 allow for more different types of property in future. This has changed the
1688 compiled form incompatibly. Anybody with saved compiled patterns that use
1689 \p or \P will have to recompile them.
1690
1691 22. Added "Any" and "L&" to the supported property types.
1692
1693 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1694 but give a compile time error if the value is greater than 0xff.
1695
1696 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1697 accidentally not being installed or uninstalled.
1698
1699 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1700 made were to insert the current release number. This seemed silly, because
1701 it made things harder for people building PCRE on systems that don't run
1702 "configure". I have turned pcre.h into a distributed file, no longer built
1703 by "configure", with the version identification directly included. There is
1704 no longer a pcre.h.in file.
1705
1706 However, this change necessitated a change to the pcre-config script as
1707 well. It is built from pcre-config.in, and one of the substitutions was the
1708 release number. I have updated configure.ac so that ./configure now finds
1709 the release number by grepping pcre.h.
1710
1711 26. Added the ability to run the tests under valgrind.
1712
1713
1714 Version 6.4 05-Sep-05
1715 ---------------------
1716
1717 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1718 "--" to be printed when multiple files were scanned, even when none of the
1719 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1720 consider it to be a bug, and have restored the previous behaviour.
1721
1722 2. A couple of code tidies to get rid of compiler warnings.
1723
1724 3. The pcretest program used to cheat by referring to symbols in the library
1725 whose names begin with _pcre_. These are internal symbols that are not
1726 really supposed to be visible externally, and in some environments it is
1727 possible to suppress them. The cheating is now confined to including
1728 certain files from the library's source, which is a bit cleaner.
1729
1730 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1731 file's purpose clearer.
1732
1733 5. Reorganized pcre_ucp_findchar().
1734
1735
1736 Version 6.3 15-Aug-05
1737 ---------------------
1738
1739 1. The file libpcre.pc.in did not have general read permission in the tarball.
1740
1741 2. There were some problems when building without C++ support:
1742
1743 (a) If C++ support was not built, "make install" and "make test" still
1744 tried to test it.
1745
1746 (b) There were problems when the value of CXX was explicitly set. Some
1747 changes have been made to try to fix these, and ...
1748
1749 (c) --disable-cpp can now be used to explicitly disable C++ support.
1750
1751 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1752 backslash in a target when C++ was disabled. This confuses some
1753 versions of "make", apparently. Using an intermediate variable solves
1754 this. (Same for CPP_LOBJ.)
1755
1756 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1757 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1758 necessary on certain architectures.
1759
1760 4. Added a setting of -export-symbols-regex to the link command to remove
1761 those symbols that are exported in the C sense, but actually are local
1762 within the library, and not documented. Their names all begin with
1763 "_pcre_". This is not a perfect job, because (a) we have to except some
1764 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1765 available (and never for static libraries). I have made a note to try to
1766 find a way round (a) in the future.
1767
1768
1769 Version 6.2 01-Aug-05
1770 ---------------------
1771
1772 1. There was no test for integer overflow of quantifier values. A construction
1773 such as {1111111111111111} would give undefined results. What is worse, if
1774 a minimum quantifier for a parenthesized subpattern overflowed and became
1775 negative, the calculation of the memory size went wrong. This could have
1776 led to memory overwriting.
1777
1778 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1779
1780 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1781 operating environments where this matters.
1782
1783 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1784 PCRE options from within the C++ wrapper.
1785
1786 5. Named capturing subpatterns were not being correctly counted when a pattern
1787 was compiled. This caused two problems: (a) If there were more than 100
1788 such subpatterns, the calculation of the memory needed for the whole
1789 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1790 back references of the form \12, where the number was greater than 9, were
1791 not recognized as back references, even though there were sufficient
1792 previous subpatterns.
1793
1794 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1795 versions of gcc, e.g. 2.95.4.
1796
1797
1798 Version 6.1 21-Jun-05
1799 ---------------------
1800
1801 1. There was one reference to the variable "posix" in pcretest.c that was not
1802 surrounded by "#if !defined NOPOSIX".
1803
1804 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1805 the cross-check on the old pcre_info() function, for the benefit of the
1806 cut-down version of PCRE that is currently imported into Exim.
1807
1808 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1809 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1810 patterns that start (?i) but I don't think that matters. The use of (?i) is
1811 just an example; this all applies to the other options as well.
1812
1813 4. Since libtool seems to echo the compile commands it is issuing, the output
1814 from "make" can be reduced a bit by putting "@" in front of each libtool
1815 compile command.
1816
1817 5. Patch from the folks at Google for configure.in to be a bit more thorough
1818 in checking for a suitable C++ installation before trying to compile the
1819 C++ stuff. This should fix a reported problem when a compiler was present,
1820 but no suitable headers.
1821
1822 6. The man pages all had just "PCRE" as their title. I have changed them to
1823 be the relevant file name. I have also arranged that these names are
1824 retained in the file doc/pcre.txt, which is a concatenation in text format
1825 of all the man pages except the little individual ones for each function.
1826
1827 7. The NON-UNIX-USE file had not been updated for the different set of source
1828 files that come with release 6. I also added a few comments about the C++
1829 wrapper.
1830
1831
1832 Version 6.0 07-Jun-05
1833 ---------------------
1834
1835 1. Some minor internal re-organization to help with my DFA experiments.
1836
1837 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1838 didn't matter for the library itself when fully configured, but did matter
1839 when compiling without UCP support, or within Exim, where the ucp files are
1840 not imported.
1841
1842 3. Refactoring of the library code to split up the various functions into
1843 different source modules. The addition of the new DFA matching code (see
1844 below) to a single monolithic source would have made it really too
1845 unwieldy, quite apart from causing all the code to be include in a
1846 statically linked application, when only some functions are used. This is
1847 relevant even without the DFA addition now that patterns can be compiled in
1848 one application and matched in another.
1849
1850 The downside of splitting up is that there have to be some external
1851 functions and data tables that are used internally in different modules of
1852 the library but which are not part of the API. These have all had their
1853 names changed to start with "_pcre_" so that they are unlikely to clash
1854 with other external names.
1855
1856 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1857 a different (DFA) algorithm. Although it is slower than the original
1858 function, it does have some advantages for certain types of matching
1859 problem.
1860
1861 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1862 including restarting after a partial match.
1863
1864 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1865 defined when compiling for Windows was sent to me. I have put it into the
1866 code, though I have no means of testing or verifying it.
1867
1868 7. Added the pcre_refcount() auxiliary function.
1869
1870 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1871 match before or at the first newline in the subject string. In pcretest,
1872 the /f option on a pattern can be used to set this.
1873
1874 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1875 would behave wrongly. This has been present in PCRE since release 4.0.
1876
1877 10. A number of changes to the pcregrep command:
1878
1879 (a) Refactored how -x works; insert ^(...)$ instead of setting
1880 PCRE_ANCHORED and checking the length, in preparation for adding
1881 something similar for -w.
1882
1883 (b) Added the -w (match as a word) option.
1884
1885 (c) Refactored the way lines are read and buffered so as to have more
1886 than one at a time available.
1887
1888 (d) Implemented a pcregrep test script.
1889
1890 (e) Added the -M (multiline match) option. This allows patterns to match
1891 over several lines of the subject. The buffering ensures that at least
1892 8K, or the rest of the document (whichever is the shorter) is available
1893 for matching (and similarly the previous 8K for lookbehind assertions).
1894
1895 (f) Changed the --help output so that it now says
1896
1897 -w, --word-regex(p)
1898
1899 instead of two lines, one with "regex" and the other with "regexp"
1900 because that confused at least one person since the short forms are the
1901 same. (This required a bit of code, as the output is generated
1902 automatically from a table. It wasn't just a text change.)
1903
1904 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1905 option but starts with a hyphen. Could be a pattern or a path name
1906 starting with a hyphen, for instance.
1907
1908 (h) "-" can be given as a file name to represent stdin.
1909
1910 (i) When file names are being printed, "(standard input)" is used for
1911 the standard input, for compatibility with GNU grep. Previously
1912 "<stdin>" was used.
1913
1914 (j) The option --label=xxx can be used to supply a name to be used for
1915 stdin when file names are being printed. There is no short form.
1916
1917 (k) Re-factored the options decoding logic because we are going to add
1918 two more options that take data. Such options can now be given in four
1919 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1920
1921 (l) Added the -A, -B, and -C options for requesting that lines of context
1922 around matches be printed.
1923
1924 (m) Added the -L option to print the names of files that do not contain
1925 any matching lines, that is, the complement of -l.
1926
1927 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1928 continue to scan other files.
1929
1930 (o) The -s option was incorrectly implemented. For compatibility with other
1931 greps, it now suppresses the error message for a non-existent or non-
1932 accessible file (but not the return code). There is a new option called
1933 -q that suppresses the output of matching lines, which was what -s was
1934 previously doing.
1935
1936 (p) Added --include and --exclude options to specify files for inclusion
1937 and exclusion when recursing.
1938
1939 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1940 Hopefully, it now does.
1941
1942 12. Missing cast in pcre_study().
1943
1944 13. Added an "uninstall" target to the makefile.
1945
1946 14. Replaced "extern" in the function prototypes in Makefile.in with
1947 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1948 world, but is set differently for Windows.
1949
1950 15. Added a second compiling function called pcre_compile2(). The only
1951 difference is that it has an extra argument, which is a pointer to an
1952 integer error code. When there is a compile-time failure, this is set
1953 non-zero, in addition to the error test pointer being set to point to an
1954 error message. The new argument may be NULL if no error number is required
1955 (but then you may as well call pcre_compile(), which is now just a
1956 wrapper). This facility is provided because some applications need a
1957 numeric error indication, but it has also enabled me to tidy up the way
1958 compile-time errors are handled in the POSIX wrapper.
1959
1960 16. Added VPATH=.libs to the makefile; this should help when building with one
1961 prefix path and installing with another. (Or so I'm told by someone who
1962 knows more about this stuff than I do.)
1963
1964 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1965 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1966 match everything, including newlines. This is not POSIX-compatible, but
1967 somebody wanted the feature. From pcretest it can be activated by using
1968 both the P and the s flags.
1969
1970 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1971
1972 19. libpcre.pc was being incorrectly installed as executable.
1973
1974 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1975 it now also looks for '\r' so that it will work unmodified on Windows.
1976
1977 21. Added Google's contributed C++ wrapper to the distribution.
1978
1979 22. Added some untidy missing memory free() calls in pcretest, to keep
1980 Electric Fence happy when testing.
1981
1982
1983
1984 Version 5.0 13-Sep-04
1985 ---------------------
1986
1987 1. Internal change: literal characters are no longer packed up into items
1988 containing multiple characters in a single byte-string. Each character
1989 is now matched using a separate opcode. However, there may be more than one
1990 byte in the character in UTF-8 mode.
1991
1992 2. The pcre_callout_block structure has two new fields: pattern_position and
1993 next_item_length. These contain the offset in the pattern to the next match
1994 item, and its length, respectively.
1995
1996 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1997 insertion of callouts before each pattern item. Added the /C option to
1998 pcretest to make use of this.
1999
2000 4. On the advice of a Windows user, the lines
2001
2002 #if defined(_WIN32) || defined(WIN32)
2003 _setmode( _fileno( stdout ), 0x8000 );
2004 #endif /* defined(_WIN32) || defined(WIN32) */
2005
2006 have been added to the source of pcretest. This apparently does useful
2007 magic in relation to line terminators.
2008
2009 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2010 for the benefit of those environments where the "b" makes a difference.
2011
2012 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2013 to know about it. I have put a hack into configure.in that adds in code
2014 to set GCC=yes if CC=icc. This seems to end up at a point in the
2015 generated configure script that is early enough to affect the setting of
2016 compiler options, which is what is needed, but I have no means of testing
2017 whether it really works. (The user who reported this had patched the
2018 generated configure script, which of course I cannot do.)
2019
2020 LATER: After change 22 below (new libtool files), the configure script
2021 seems to know about icc (and also ecc). Therefore, I have commented out
2022 this hack in configure.in.
2023
2024 7. Added support for pkg-config (2 patches were sent in).
2025
2026 8. Negated POSIX character classes that used a combination of internal tables
2027 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2028 [[:^ascii]]. Typically, they would match almost any characters. The other
2029 POSIX classes were not broken in this way.
2030
2031 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2032 to find the match, as PCRE was deluded into thinking that the match had to
2033 start at the start point or following a newline. The same bug applied to
2034 patterns with negative forward assertions or any backward assertions
2035 preceding ".*" at the start, unless the pattern required a fixed first
2036 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2037
2038 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2039 starting at the last subject character, bytes beyond the end of the subject
2040 string were read.
2041
2042 11. Renamed the variable "class" as "classbits" to make life easier for C++
2043 users. (Previously there was a macro definition, but it apparently wasn't
2044 enough.)
2045
2046 12. Added the new field "tables" to the extra data so that tables can be passed
2047 in at exec time, or the internal tables can be re-selected. This allows
2048 a compiled regex to be saved and re-used at a later time by a different
2049 program that might have everything at different addresses.
2050
2051 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2052 -R library as well as a -L library.
2053
2054 14. The debugging options of pcretest (-d on the command line or D on a
2055 pattern) showed incorrect output for anything following an extended class
2056 that contained multibyte characters and which was followed by a quantifier.
2057
2058 15. Added optional support for general category Unicode character properties
2059 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2060 support. It adds about 90K to the size of the library. The meanings of the
2061 inbuilt class escapes such as \d and \s have NOT been changed.
2062
2063 16. Updated pcredemo.c to include calls to free() to release the memory for the
2064 compiled pattern.
2065
2066 17. The generated file chartables.c was being created in the source directory
2067 instead of in the building directory. This caused the build to fail if the
2068 source directory was different from the building directory, and was
2069 read-only.
2070
2071 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2072 file. No doubt somebody will tell me if they don't make sense... Also added
2073 Dan Mooney's comments about building on OpenVMS.
2074
2075 19. Added support for partial matching via the PCRE_PARTIAL option for
2076 pcre_exec() and the \P data escape in pcretest.
2077
2078 20. Extended pcretest with 3 new pattern features:
2079
2080 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2081 write the compiled pattern to the file whose name is "rest-of-line".
2082 This is a straight binary dump of the data, with the saved pointer to
2083 the character tables forced to be NULL. The study data, if any, is
2084 written too. After writing, pcretest reads a new pattern.
2085
2086 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2087 compiled pattern from the given file. There must not be any
2088 occurrences of "<" in the file name (pretty unlikely); if there are,
2089 pcretest will instead treat the initial "<" as a pattern delimiter.
2090 After reading in the pattern, pcretest goes on to read data lines as
2091 usual.
2092
2093 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2094 and 16-bit fields in a compiled pattern, to simulate a pattern that
2095 was compiled on a host of opposite endianness.
2096
2097 21. The pcre-exec() function can now cope with patterns that were compiled on
2098 hosts of opposite endianness, with this restriction:
2099
2100 As for any compiled expression that is saved and used later, the tables
2101 pointer field cannot be preserved; the extra_data field in the arguments
2102 to pcre_exec() should be used to pass in a tables address if a value
2103 other than the default internal tables were used at compile time.
2104
2105 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2106 now diagnosed as an error. Previously, most of the time, a negative number
2107 would have been treated as zero, but if in addition "ovector" was passed as
2108 NULL, a crash could occur.
2109
2110 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2111 new versions from the libtool 1.5 distribution (the last one is a copy of
2112 a file called libtool.m4). This seems to have fixed the need to patch
2113 "configure" to support Darwin 1.3 (which I used to do). However, I still
2114 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2115 workstation).
2116
2117 24. Changed the PCRE licence to be the more standard "BSD" licence.
2118
2119
2120 Version 4.5 01-Dec-03
2121 ---------------------
2122
2123 1. There has been some re-arrangement of the code for the match() function so
2124 that it can be compiled in a version that does not call itself recursively.
2125 Instead, it keeps those local variables that need separate instances for
2126 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2127 needs to "recurse". Keeping track of where control must go is done by means
2128 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2129 hide most of the details from the main code, and operates only if
2130 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2131 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2132 operating.
2133
2134 To make it easier for callers to provide specially tailored get/free
2135 functions for this usage, two new functions, pcre_stack_malloc, and
2136 pcre_stack_free, are used. They are always called in strict stacking order,
2137 and the size of block requested is always the same.
2138
2139 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2140 PCRE has been compiled to use the stack or the heap for recursion. The
2141 -C option of pcretest uses this to show which version is compiled.
2142
2143 A new data escape \S, is added to pcretest; it causes the amounts of store
2144 obtained and freed by both kinds of malloc/free at match time to be added
2145 to the output.
2146
2147 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2148 what's available on my current Linux desktop machine.
2149
2150 3. When matching a UTF-8 string, the test for a valid string at the start has
2151 been extended. If start_offset is not zero, PCRE now checks that it points
2152 to a byte that is the start of a UTF-8 character. If not, it returns
2153 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2154 this is necessary because there may be backward assertions in the pattern.
2155 When matching the same subject several times, it may save resources to use
2156 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2157
2158 4. The code for checking the validity of UTF-8 strings has been tightened so
2159 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2160 containing "overlong sequences".
2161
2162 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2163 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2164 should have been "&", but it just so happened that all the cases this let
2165 through by mistake were picked up later in the function.
2166
2167 6. I had used a variable called "isblank" - this is a C99 function, causing
2168 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2169
2170 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2171 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2172 so the version is shown; (c) stop "make test" echoing "./RunTest".
2173
2174 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2175
2176 9. The prototype for memmove() for systems that don't have it was using
2177 size_t, but the inclusion of the header that defines size_t was later. I've
2178 moved the #includes for the C headers earlier to avoid this.
2179
2180 10. Added some adjustments to the code to make it easier to compiler on certain
2181 special systems:
2182
2183 (a) Some "const" qualifiers were missing.
2184 (b) Added the macro EXPORT before all exported functions; by default this
2185 is defined to be empty.
2186 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2187 that it reads its output file name as an argument instead of writing
2188 to the standard output and assuming this can be redirected.
2189
2190 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2191 class containing characters with values greater than 255, PCRE compilation
2192 went into a loop.
2193
2194 12. A recursive reference to a subpattern that was within another subpattern
2195 that had a minimum quantifier of zero caused PCRE to crash. For example,
2196 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2197 recursion. If the recursively-called subpattern itself had a zero repeat,
2198 that was OK.
2199
2200 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2201 buffer into which it was copied (for escape processing) was still set at
2202 1024, so long lines caused crashes.
2203
2204 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2205 "internal error: code overflow...". This applied to any character class
2206 that was followed by a possessive quantifier.
2207
2208 15. Modified the Makefile to add libpcre.la as a prerequisite for
2209 libpcreposix.la because I was told this is needed for a parallel build to
2210 work.
2211
2212 16. If a pattern that contained .* following optional items at the start was
2213 studied, the wrong optimizing data was generated, leading to matching
2214 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2215 matching string must start with a or b or c. The correct conclusion for
2216 this pattern is that a match can start with any character.
2217
2218
2219 Version 4.4 13-Aug-03
2220 ---------------------
2221
2222 1. In UTF-8 mode, a character class containing characters with values between
2223 127 and 255 was not handled correctly if the compiled pattern was studied.
2224 In fixing this, I have also improved the studying algorithm for such
2225 classes (slightly).
2226
2227 2. Three internal functions had redundant arguments passed to them. Removal
2228 might give a very teeny performance improvement.
2229
2230 3. Documentation bug: the value of the capture_top field in a callout is *one
2231 more than* the number of the hightest numbered captured substring.
2232
2233 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2234 in incorrectly linking with a previously installed version. They now link
2235 explicitly with libpcre.la.
2236
2237 5. configure.in no longer needs to recognize Cygwin specially.
2238
2239 6. A problem in pcre.in for Windows platforms is fixed.
2240
2241 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2242 pcretest, it used to include the size of the study block as part of its
2243 output. Unfortunately, the structure contains a field that has a different
2244 size on different hardware architectures. This meant that the tests that
2245 showed this size failed. As the block is currently always of a fixed size,
2246 this information isn't actually particularly useful in pcretest output, so
2247 I have just removed it.
2248
2249 8. Three pre-processor statements accidentally did not start in column 1.
2250 Sadly, there are *still* compilers around that complain, even though
2251 standard C has not required this for well over a decade. Sigh.
2252
2253 9. In pcretest, the code for checking callouts passed small integers in the
2254 callout_data field, which is a void * field. However, some picky compilers
2255 complained about the casts involved for this on 64-bit systems. Now
2256 pcretest passes the address of the small integer instead, which should get
2257 rid of the warnings.
2258
2259 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2260 both compile and run time, and gives an error if an invalid UTF-8 sequence
2261 is found. There is a option for disabling this check in cases where the
2262 string is known to be correct and/or the maximum performance is wanted.
2263
2264 11. In response to a bug report, I changed one line in Makefile.in from
2265
2266 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2267 to
2268 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2269
2270 to look similar to other lines, but I have no way of telling whether this
2271 is the right thing to do, as I do not use Windows. No doubt I'll get told
2272 if it's wrong...
2273
2274
2275 Version 4.3 21-May-03
2276 ---------------------
2277
2278 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2279 Makefile.
2280
2281 2. Some refactoring to improve the quality of the code:
2282
2283 (i) The utf8_table... variables are now declared "const".
2284
2285 (ii) The code for \cx, which used the "case flipping" table to upper case
2286 lower case letters, now just substracts 32. This is ASCII-specific,
2287 but the whole concept of \cx is ASCII-specific, so it seems
2288 reasonable.
2289
2290 (iii) PCRE was using its character types table to recognize decimal and
2291 hexadecimal digits in the pattern. This is silly, because it handles
2292 only 0-9, a-f, and A-F, but the character types table is locale-
2293 specific, which means strange things might happen. A private
2294 table is now used for this - though it costs 256 bytes, a table is
2295 much faster than multiple explicit tests. Of course, the standard
2296 character types table is still used for matching digits in subject
2297 strings against \d.
2298
2299 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2300 ending in _t are). So I've renamed it as ESC_tee.
2301
2302 3. The first argument for regexec() in the POSIX wrapper should have been
2303 defined as "const".
2304
2305 4. Changed pcretest to use malloc() for its buffers so that they can be
2306 Electric Fenced for debugging.
2307
2308 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2309 to read one or more bytes before the start of the subject string. Often this
2310 had no effect on PCRE's behaviour, but in some circumstances it could
2311 provoke a segmentation fault.
2312
2313 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2314 to try to read one or more bytes before the start of the subject string.
2315
2316 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2317 UTF-8 support could misbehave in various ways if the subject string
2318 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2319 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2320 back over UTF-8 characters.)
2321
2322
2323 Version 4.2 14-Apr-03
2324 ---------------------
2325
2326 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2327
2328 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2329 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2330 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2331 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2332 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2333 and BUILD_EXEEXT
2334 Note: automatic setting of the BUILD variables is not yet working
2335 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2336 compile-time but not at link-time
2337 [LINK]: use for linking executables only
2338 make different versions for Windows and non-Windows
2339 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2340 libraries
2341 [LINK_FOR_BUILD]: new variable
2342 [OBJEXT]: use throughout
2343 [EXEEXT]: use throughout
2344 <winshared>: new target
2345 <wininstall>: new target
2346 <dftables.o>: use native compiler
2347 <dftables>: use native linker
2348 <install>: handle Windows platform correctly
2349 <clean>: ditto
2350 <check>: ditto
2351 copy DLL to top builddir before testing
2352
2353 As part of these changes, -no-undefined was removed again. This was reported
2354 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2355 in any case.
2356
2357 3. Some tidies to get rid of compiler warnings:
2358
2359 . In the match_data structure, match_limit was an unsigned long int, whereas
2360 match_call_count was an int. I've made them both unsigned long ints.
2361
2362 . In pcretest the fact that a const uschar * doesn't automatically cast to
2363 a void * provoked a warning.
2364
2365 . Turning on some more compiler warnings threw up some "shadow" variables
2366 and a few more missing casts.
2367
2368 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2369 option, a class that contained a single character with a value between 128
2370 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2371
2372 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2373 option, a class that contained several characters, but with at least one
2374 whose value was between 128 and 255 caused PCRE to crash.
2375
2376
2377 Version 4.1 12-Mar-03
2378 ---------------------
2379
2380 1. Compiling with gcc -pedantic found a couple of places where casts were
2381 needed, and a string in dftables.c that was longer than standard compilers are
2382 required to support.
2383
2384 2. Compiling with Sun's compiler found a few more places where the code could
2385 be tidied up in order to avoid warnings.
2386
2387 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2388 first of these names is deprecated in the latest Autoconf in favour of the name
2389 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2390 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2391 analogy I have changed it to CFLAGS_FOR_BUILD.
2392
2393 4. Added -no-undefined to the linking command in the Makefile, because this is
2394 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2395 linking step for the pcreposix library.
2396
2397 5. PCRE was failing to diagnose the case of two named groups with the same
2398 name.
2399
2400 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2401 literal character that is needed in the subject for a match, and scans along to
2402 ensure that it is present before embarking on the full matching process. This
2403 saves time in cases of nested unlimited repeats that are never going to match.
2404 Problem: the scan can take a lot of time if the subject is very long (e.g.
2405 megabytes), thus penalizing straightforward matches. It is now done only if the
2406 amount of subject to be scanned is less than 1000 bytes.
2407
2408 7. A lesser problem with the same optimization is that it was recording the
2409 first character of an anchored pattern as "needed", thus provoking a search
2410 right along the subject, even when the first match of the pattern was going to
2411 fail. The "needed" character is now not set for anchored patterns, unless it
2412 follows something in the pattern that is of non-fixed length. Thus, it still
2413 fulfils its original purpose of finding quick non-matches in cases of nested
2414 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2415
2416
2417 Version 4.0 17-Feb-03
2418 ---------------------
2419
2420 1. If a comment in an extended regex that started immediately after a meta-item
2421 extended to the end of string, PCRE compiled incorrect data. This could lead to
2422 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2423
2424 2. Moved to autoconf 2.53 and libtool 1.4.2.
2425
2426 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2427 the special perltest8 script is no longer needed - all the tests can be run
2428 from a single perltest script.
2429
2430 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2431 by \s. It has now been removed in PCRE. This means it isn't recognized as
2432 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2433 class [:space:] *does* include VT, thereby creating a mess.
2434
2435 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2436 space and tab.
2437
2438 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2439 its new features into the main test script, reducing the number of scripts.
2440
2441 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2442 were backward compatible, and made the (?i) apply to the whole pattern, as if
2443 /i were given. Now it behaves more logically, and applies the option setting
2444 only to what follows. PCRE has been changed to follow suit. However, if it
2445 finds options settings right at the start of the pattern, it extracts them into
2446 the global options, as before. Thus, they show up in the info data.
2447
2448 8. Added support for the \Q...\E escape sequence. Characters in between are
2449 treated as literals. This is slightly different from Perl in that $ and @ are
2450 also handled as literals inside the quotes. In Perl, they will cause variable
2451 interpolation. Note the following examples:
2452
2453 Pattern PCRE matches Perl matches
2454
2455 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2456 \Qabc\$xyz\E abc\$xyz abc\$xyz
2457 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2458
2459 For compatibility with Perl, \Q...\E sequences are recognized inside character
2460 classes as well as outside them.
2461
2462 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2463 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2464 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2465 signed/unsigned warnings.
2466
2467 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2468 option for pcretest, so I've replaced it by a simple function that does just
2469 that job.
2470
2471 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2472 "pcregrep -".
2473
2474 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2475 Java package. This provides some syntactic sugar for simple cases of what my
2476 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2477 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2478 item, you can use this simplified notation. Note that only makes sense with
2479 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2480 greediness, whatever the setting of the PCRE_UNGREEDY option.
2481
2482 13. A change of greediness default within a pattern was not taking effect at
2483 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2484 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2485 was abstracted outside.
2486
2487 14. PCRE now supports the \G assertion. It is true when the current matching
2488 position is at the start point of the match. This differs from \A when the
2489 starting offset is non-zero. Used with the /g option of pcretest (or similar
2490 code), it works in the same way as it does for Perl's /g option. If all
2491 alternatives of a regex begin with \G, the expression is anchored to the start
2492 match position, and the "anchored" flag is set in the compiled expression.
2493
2494 15. Some bugs concerning the handling of certain option changes within patterns
2495 have been fixed. These applied to options other than (?ims). For example,
2496 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2497 been the other way round. Some of this was related to change 7 above.
2498
2499 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2500 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2501 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2502 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2503
2504 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2505 mode. Unlike ".", it always matches newline, whatever the setting of
2506 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2507 assertions. Perl allows it, but it doesn't (in general) work because it can't
2508 calculate the length of the lookbehind. At least, that's the case for Perl
2509 5.8.0 - I've been told they are going to document that it doesn't work in
2510 future.
2511
2512 18. Added an error diagnosis for escapes that PCRE does not support: these are
2513 \L, \l, \N, \P, \p, \U, \u, and \X.
2514
2515 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2516 reading past the end of the pattern in cases such as /[abcd/.
2517
2518 20. PCRE was getting more memory than necessary for patterns with classes that
2519 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2520
2521 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2522 compiling PCRE for use with Virtual Pascal.
2523
2524 22. Small fix to the Makefile to make it work properly if the build is done
2525 outside the source tree.
2526
2527 23. Added a new extension: a condition to go with recursion. If a conditional
2528 subpattern starts with (?(R) the "true" branch is used if recursion has
2529 happened, whereas the "false" branch is used only at the top level.
2530
2531 24. When there was a very long string of literal characters (over 255 bytes
2532 without UTF support, over 250 bytes with UTF support), the computation of how
2533 much memory was required could be incorrect, leading to segfaults or other
2534 strange effects.
2535
2536 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2537 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2538 there was a subsequent back reference to those brackets. This meant that, for
2539 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2540 possible to check for precisely this case. All we can do is abandon the
2541 optimization if .* occurs inside capturing brackets when there are any back
2542 references whatsoever. (See below for a better fix that came later.)
2543
2544 26. The handling of the optimization for finding the first character of a
2545 non-anchored pattern, and for finding a character that is required later in the
2546 match were failing in some cases. This didn't break the matching; it just
2547 failed to optimize when it could. The way this is done has been re-implemented.
2548
2549 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2550
2551 28. Added a new feature that provides some of the functionality that Perl
2552 provides with (?{...}). The facility is termed a "callout". The way it is done
2553 in PCRE is for the caller to provide an optional function, by setting
2554 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2555 global variable. By default it is unset, which disables all calling out. To get
2556 the function called, the regex must include (?C) at appropriate points. This
2557 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2558 This provides a means of identifying different callout points. When PCRE
2559 reaches such a point in the regex, if pcre_callout has been set, the external
2560 function is called. It is provided with data in a structure called
2561 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2562 matching continues; if it returns a non-zero value, the match at the current
2563 point fails. However, backtracking will occur if possible. [This was changed
2564 later and other features added - see item 49 below.]
2565
2566 29. pcretest is upgraded to test the callout functionality. It provides a
2567 callout function that displays information. By default, it shows the start of
2568 the match and the current position in the text. There are some new data escapes
2569 to vary what happens:
2570
2571 \C+ in addition, show current contents of captured substrings
2572 \C- do not supply a callout function
2573 \C!n return 1 when callout number n is reached
2574 \C!n!m return 1 when callout number n is reached for the mth time
2575
2576 30. If pcregrep was called with the -l option and just a single file name, it
2577 output "<stdin>" if a match was found, instead of the file name.
2578
2579 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2580 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2581 pcre_exec(). This saves a malloc/free per call. The default value of
2582 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2583 when configuring.
2584
2585 32. The default maximum size of a compiled pattern is 64K. There have been a
2586 few cases of people hitting this limit. The code now uses macros to handle the
2587 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2588 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2589 configuring. Tests 2 and 5 work only with 2-byte links because they output
2590 debugging information about compiled patterns.
2591
2592 33. Internal code re-arrangements:
2593
2594 (a) Moved the debugging function for printing out a compiled regex into
2595 its own source file (printint.c) and used #include to pull it into
2596 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2597 separate copies.
2598
2599 (b) Defined the list of op-code names for debugging as a macro in
2600 internal.h so that it is next to the definition of the opcodes.
2601
2602 (c) Defined a table of op-code lengths for simpler skipping along compiled
2603 code. This is again a macro in internal.h so that it is next to the
2604 definition of the opcodes.
2605
2606 34. Added support for recursive calls to individual subpatterns, along the
2607 lines of Robin Houston's patch (but implemented somewhat differently).
2608
2609 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2610 allow it to read and process whole directories in Win32. This code was
2611 contributed by Lionel Fourquaux; it has not been tested by me.
2612
2613 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2614 used to name a group. Names consist of alphanumerics and underscores, and must
2615 be unique. Back references use the syntax (?P=name) and recursive calls use
2616 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2617 numbers. The function pcre_fullinfo() can be used after compilation to extract
2618 a name/number map. There are three relevant calls:
2619
2620 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2621 PCRE_INFO_NAMECOUNT yields the number of entries
2622 PCRE_INFO_NAMETABLE yields a pointer to the map.
2623
2624 The map is a vector of fixed-size entries. The size of each entry depends on
2625 the length of the longest name used. The first two bytes of each entry are the
2626 group number, most significant byte first. There follows the corresponding
2627 name, zero terminated. The names are in alphabetical order.
2628
2629 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2630 case instead of 255. Making it the same both with and without UTF-8 support
2631 means that the same test output works with both.
2632
2633 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2634 calling malloc() with a zero argument.
2635
2636 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2637 optimization. I've improved things by keeping a bitmap of backreferences with
2638 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2639 fact referenced, the optimization can be applied. It is unlikely that a
2640 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2641 the match to follow \n) will appear inside brackets with a number greater than
2642 31, but if it does, any back reference > 31 suppresses the optimization.
2643
2644 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2645 of disabling numbered capturing parentheses. Any opening parenthesis that is
2646 not followed by ? behaves as if it were followed by ?: but named parentheses
2647 can still be used for capturing (and they will acquire numbers in the usual
2648 way).
2649
2650 41. Redesigned the return codes from the match() function into yes/no/error so
2651 that errors can be passed back from deep inside the nested calls. A malloc
2652 failure while inside a recursive subpattern call now causes the
2653 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2654
2655 42. It is now possible to set a limit on the number of times the match()
2656 function is called in a call to pcre_exec(). This facility makes it possible to
2657 limit the amount of recursion and backtracking, though not in a directly
2658 obvious way, because the match() function is used in a number of different
2659 circumstances. The count starts from zero for each position in the subject
2660 string (for non-anchored patterns). The default limit is, for compatibility, a
2661 large number, namely 10 000 000. You can change this in two ways:
2662
2663 (a) When configuring PCRE before making, you can use --with-match-limit=n
2664 to set a default value for the compiled library.
2665
2666 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2667 a different value is set. See 45 below.
2668
2669 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2670
2671 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2672 of things that can be changed at compile time. The first argument specifies
2673 what is wanted and the second points to where the information is to be placed.
2674 The current list of available information is:
2675
2676 PCRE_CONFIG_UTF8
2677
2678 The output is an integer that is set to one if UTF-8 support is available;
2679 otherwise it is set to zero.
2680
2681 PCRE_CONFIG_NEWLINE
2682
2683 The output is an integer that it set to the value of the code that is used for
2684 newline. It is either LF (10) or CR (13).
2685
2686 PCRE_CONFIG_LINK_SIZE
2687
2688 The output is an integer that contains the number of bytes used for internal
2689 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2690
2691 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2692
2693 The output is an integer that contains the threshold above which the POSIX
2694 interface uses malloc() for output vectors. See item 31 above.
2695
2696 PCRE_CONFIG_MATCH_LIMIT
2697
2698 The output is an unsigned integer that contains the default limit of the number
2699 of match() calls in a pcre_exec() execution. See 42 above.
2700
2701 44. pcretest has been upgraded by the addition of the -C option. This causes it
2702 to extract all the available output from the new pcre_config() function, and to
2703 output it. The program then exits immediately.
2704
2705 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2706 order to support additional features. One way would have been to define
2707 pcre_exec2() (for example) with extra arguments, but this would not have been
2708 extensible, and would also have required all calls to the original function to
2709 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2710 is used for passing in "extra" data from pcre_study().
2711
2712 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2713 contains the following fields:
2714
2715 flags a bitmap indicating which of the following fields are set
2716 study_data opaque data from pcre_study()
2717 match_limit a way of specifying a limit on match() calls for a specific
2718 call to pcre_exec()
2719 callout_data data for callouts (see 49 below)
2720
2721 The flag bits are also defined in pcre.h, and are
2722
2723 PCRE_EXTRA_STUDY_DATA
2724 PCRE_EXTRA_MATCH_LIMIT
2725 PCRE_EXTRA_CALLOUT_DATA
2726
2727 The pcre_study() function now returns one of these new pcre_extra blocks, with
2728 the actual study data pointed to by the study_data field, and the
2729 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2730 before. That is, this change is entirely upwards-compatible and requires no
2731 change to existing code.
2732
2733 If you want to pass in additional data to pcre_exec(), you can either place it
2734 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2735 block.
2736
2737 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2738 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2739 times with different match limits, until it finds the minimum value needed for
2740 pcre_exec() to complete. The value is then output. This can be instructive; for
2741 most simple matches the number is quite small, but for pathological cases it
2742 gets very large very quickly.
2743
2744 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2745 returns the size of the data block pointed to by the study_data field in a
2746 pcre_extra block, that is, the value that was passed as the argument to
2747 pcre_malloc() when PCRE was getting memory in which to place the information
2748 created by pcre_study(). The fourth argument should point to a size_t variable.
2749 pcretest has been extended so that this information is shown after a successful
2750 pcre_study() call when information about the compiled regex is being displayed.
2751
2752 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2753 because what follows is always an absolute path. (Later: it turns out that this
2754 is more than cosmetic for MinGW, because it doesn't like empty path
2755 components.)
2756
2757 49. Some changes have been made to the callout feature (see 28 above):
2758
2759 (i) A callout function now has three choices for what it returns:
2760
2761 0 => success, carry on matching
2762 > 0 => failure at this point, but backtrack if possible
2763 < 0 => serious error, return this value from pcre_exec()
2764
2765 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2766 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2767 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2768 use by callout functions. It will never be used by PCRE itself.
2769
2770 (ii) The pcre_extra structure (see 45 above) has a void * field called
2771 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2772 pcre_callout_block structure has a field of the same name. The contents of
2773 the field passed in the pcre_extra structure are passed to the callout
2774 function in the corresponding field in the callout block. This makes it
2775 easier to use the same callout-containing regex from multiple threads. For
2776 testing, the pcretest program has a new data escape
2777
2778 \C*n pass the number n (may be negative) as callout_data
2779
2780 If the callout function in pcretest receives a non-zero value as
2781 callout_data, it returns that value.
2782
2783 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2784 there were some redundant $(CFLAGS) in commands that are now specified as
2785 $(LINK), which already includes $(CFLAGS).
2786
2787 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2788 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2789 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2790 one-byte characters throughout. Note that case-insensitive matching applies
2791 only to characters whose values are less than 256. PCRE doesn't support the
2792 notion of cases for higher-valued characters.
2793
2794 (i) A character class whose characters are all within 0-255 is handled as
2795 a bit map, and the map is inverted for negative classes. Previously, a
2796 character > 255 always failed to match such a class; however it should
2797 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2798
2799 (ii) A negated character class with a single character < 255 is coded as
2800 "not this character" (OP_NOT). This wasn't working properly when the test
2801 character was multibyte, either singly or repeated.
2802
2803 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2804 mode, for example: \x{100}{2,3}.
2805
2806 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2807 singly or repeated) now correctly test multibyte characters. However,
2808 PCRE doesn't recognize any characters with values greater than 255 as
2809 digits, spaces, or word characters. Such characters always match \D, \S,
2810 and \W, and never match \d, \s, or \w.
2811
2812 (v) Classes may now contain characters and character ranges with values
2813 greater than 255. For example: [ab\x{100}-\x{400}].
2814
2815 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2816 PCRE in UTF-8 mode.
2817
2818 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2819 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2820 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2821 value.)
2822
2823 53. The single man page has become too large. I have therefore split it up into
2824 a number of separate man pages. These also give rise to individual HTML pages;
2825 these are now put in a separate directory, and there is an index.html page that
2826 lists them all. Some hyperlinking between the pages has been installed.
2827
2828 54. Added convenience functions for handling named capturing parentheses.
2829
2830 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2831 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2832 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2833 are faulted.
2834
2835 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2836 calling configure. These values are used when compiling the dftables.c program
2837 which is run to generate the source of the default character tables. They
2838 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2839 you will need to set these values.
2840
2841 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2842
2843
2844 Version 3.9 02-Jan-02
2845 ---------------------
2846
2847 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2848
2849 2. If --disable-static was given, the building process failed when trying to
2850 build pcretest and pcregrep. (For some reason it was using libtool to compile
2851 them, which is not right, as they aren't part of the library.)
2852
2853
2854 Version 3.8 18-Dec-01
2855 ---------------------
2856
2857 1. The experimental UTF-8 code was completely screwed up. It was packing the
2858 bytes in the wrong order. How dumb can you get?
2859
2860
2861 Version 3.7 29-Oct-01
2862 ---------------------
2863
2864 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2865 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2866 this didn't happen under Solaris 8, where I normally test things.
2867
2868 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2869 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2870 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2871 this fix an hour or so after the initial 3.7 release.)
2872
2873
2874 Version 3.6 23-Oct-01
2875 ---------------------
2876
2877 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2878 offsets passed as NULL with zero offset count.
2879
2880 2. The config.guess and config.sub files had not been updated when I moved to
2881 the latest autoconf.
2882
2883
2884 Version 3.5 15-Aug-01
2885 ---------------------
2886
2887 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2888 had been forgotten.
2889
2890 2. By using declared but undefined structures, we can avoid using "void"
2891 definitions in pcre.h while keeping the internal definitions of the structures
2892 private.
2893
2894 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2895 user point of view, this means that both static and shared libraries are built
2896 by default, but this can be individually controlled. More of the work of
2897 handling this static/shared cases is now inside libtool instead of PCRE's make
2898 file.
2899
2900 4. The pcretest utility is now installed along with pcregrep because it is
2901 useful for users (to test regexs) and by doing this, it automatically gets
2902 relinked by libtool. The documentation has been turned into a man page, so
2903 there are now .1, .txt, and .html versions in /doc.
2904
2905 5. Upgrades to pcregrep:
2906 (i) Added long-form option names like gnu grep.
2907 (ii) Added --help to list all options with an explanatory phrase.
2908 (iii) Added -r, --recursive to recurse into sub-directories.
2909 (iv) Added -f, --file to read patterns from a file.
2910
2911 6. pcre_exec() was referring to its "code" argument before testing that
2912 argument for NULL (and giving an error if it was NULL).
2913
2914 7. Upgraded Makefile.in to allow for compiling in a different directory from
2915 the source directory.
2916
2917 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2918 options bits, the pointer it was passed was to an int instead of to an unsigned
2919 long int. This mattered only on 64-bit systems.
2920
2921 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2922 generated) instead of pcre.in, which it its source. Also made the same change
2923 in several of the .c files.
2924
2925 10. A new release of gcc defines printf() as a macro, which broke pcretest
2926 because it had an ifdef in the middle of a string argument for printf(). Fixed
2927 by using separate calls to printf().
2928
2929 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2930 script, to force use of CR or LF instead of \n in the source. On non-Unix
2931 systems, the value can be set in config.h.
2932
2933 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2934 absolute limit. Changed the text of the error message to make this clear, and
2935 likewise updated the man page.
2936
2937 13. The limit of 99 on the number of capturing subpatterns has been removed.
2938 The new limit is 65535, which I hope will not be a "real" limit.
2939
2940
2941 Version 3.4 22-Aug-00
2942 ---------------------
2943
2944 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2945
2946 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2947
2948
2949 Version 3.3 01-Aug-00
2950 ---------------------
2951
2952 1. If an octal character was given, but the value was greater than \377, it
2953 was not getting masked to the least significant bits, as documented. This could
2954 lead to crashes in some systems.
2955
2956 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2957 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2958
2959 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2960 These just pass their arguments on to (pcre_free)(), but they are provided
2961 because some uses of PCRE bind it to non-C systems that can call its functions,
2962 but cannot call free() or pcre_free() directly.
2963
2964 4. Add "make test" as a synonym for "make check". Corrected some comments in
2965 the Makefile.
2966
2967 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2968 Makefile.
2969
2970 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2971 command called pgrep for grepping around the active processes.
2972
2973 7. Added the beginnings of support for UTF-8 character strings.
2974
2975 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2976 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2977 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2978 out for the ar command.)
2979
2980
2981 Version 3.2 12-May-00
2982 ---------------------
2983
2984 This is purely a bug fixing release.
2985
2986 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2987 of ZA. This was just one example of several cases that could provoke this bug,
2988 which was introduced by change 9 of version 2.00. The code for breaking
2989 infinite loops after an iteration that matches an empty string was't working
2990 correctly.
2991
2992 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2993 when matched against abbab (for example). After matching an empty string, it
2994 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2995 caused it to match further down the string than it should.
2996
2997 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2998 was there because it doesn't seem to be needed, and it causes trouble on some
2999 systems, as it is not a Standard C header. It has been removed.
3000
3001 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3002 were reported on the Macintosh. The changes were from
3003
3004 while ((c = *(++ptr)) != 0 && c != '\n');
3005 to
3006 while ((c = *(++ptr)) != 0 && c != '\n') ;
3007
3008 Totally extraordinary, but if that's what it takes...
3009
3010 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3011 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3012 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3013 assumes the way PCRE uses memmove() (always moving upwards).
3014
3015 6. PCRE is being used in one environment where strchr() is not available. There
3016 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3017 faster code anyway.
3018
3019
3020 Version 3.1 09-Feb-00
3021 ---------------------
3022
3023 The only change in this release is the fixing of some bugs in Makefile.in for
3024 the "install" target:
3025
3026 (1) It was failing to install pcreposix.h.
3027
3028 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3029
3030
3031 Version 3.0 01-Feb-00
3032 ---------------------
3033
3034 1. Add support for the /+ modifier to perltest (to output $` like it does in
3035 pcretest).
3036
3037 2. Add support for the /g modifier to perltest.
3038
3039 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3040 matches null strings.
3041
3042 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3043 pattern. Perl treats empty patterns specially - it reuses the most recent
3044 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3045 effect.
3046
3047 5. The POSIX interface was broken in that it was just handing over the POSIX
3048 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3049 required a bigger vector, with some working space on the end. This means that
3050 the POSIX wrapper now has to get and free some memory, and copy the results.
3051
3052 6. Added some simple autoconf support, placing the test data and the
3053 documentation in separate directories, re-organizing some of the
3054 information files, and making it build pcre-config (a GNU standard). Also added
3055 libtool support for building PCRE as a shared library, which is now the
3056 default.
3057
3058 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3059 09 are not valid octal constants. Single digits will be used for minor values
3060 less than 10.
3061
3062 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3063 existing programs that set these in the POSIX interface can use PCRE without
3064 modification.
3065
3066 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3067 return all that pcre_info() returns, plus additional data. The pcre_info()
3068 function is retained for compatibility, but is considered to be obsolete.
3069
3070 10. Added experimental recursion feature (?R) to handle one common case that
3071 Perl 5.6 will be able to do with (?p{...}).
3072
3073 11. Added support for POSIX character classes like [:alpha:], which Perl is
3074 adopting.
3075
3076
3077 Version 2.08 31-Aug-99
3078 ----------------------
3079
3080 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3081 trying to match at the startoffset position, but instead was moving forward to
3082 the next newline as if a previous match had failed.
3083
3084 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3085 and could get into a loop if a null string was matched other than at the start
3086 of the subject.
3087
3088 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3089 be distinguished at compile time, and for completeness also added PCRE_DATE.
3090
3091 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3092 in GnuWin32 environments.
3093
3094
3095 Version 2.07 29-Jul-99
3096 ----------------------
3097
3098 1. The documentation is now supplied in plain text form and HTML as well as in
3099 the form of man page sources.
3100
3101 2. C++ compilers don't like assigning (void *) values to other pointer types.
3102 In particular this affects malloc(). Although there is no problem in Standard
3103 C, I've put in casts to keep C++ compilers happy.
3104
3105 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3106 should be (const char *).
3107
3108 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3109 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3110 However, I haven't made this a standard facility. The documentation doesn't
3111 mention it, and the Makefile doesn't support it.
3112
3113 5. The Makefile now contains an "install" target, with editable destinations at
3114 the top of the file. The pcretest program is not installed.
3115
3116 6. pgrep -V now gives the PCRE version number and date.
3117
3118 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3119 causing the entire string to be ignored, instead of just the last character.
3120
3121 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3122 non-matching string, it can take a very, very long time, even for strings of
3123 quite modest length, because of the nested recursion. PCRE now does better in
3124 some of these cases. It does this by remembering the last required literal
3125 character in the pattern, and pre-searching the subject to ensure it is present
3126 before running the real match. In other words, it applies a heuristic to detect
3127 some types of certain failure quickly, and in the above example, if presented
3128 with a string that has no trailing " it gives "no match" very quickly.
3129
3130 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3131 other alternatives are tried instead.
3132
3133
3134 Version 2.06 09-Jun-99
3135 ----------------------
3136
3137 1. Change pcretest's output for amount of store used to show just the code
3138 space, because the remainder (the data block) varies in size between 32-bit and
3139 64-bit systems.
3140
3141 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3142 start matching at. This allows lookbehinds to work when searching for multiple
3143 occurrences in a string.
3144
3145 3. Added additional options to pcretest for testing multiple occurrences:
3146
3147 /+ outputs the rest of the string that follows a match
3148 /g loops for multiple occurrences, using the new startoffset argument
3149 /G loops for multiple occurrences by passing an incremented pointer
3150
3151 4. PCRE wasn't doing the "first character" optimization for patterns starting
3152 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3153 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3154 the letter 'x'. On long subject strings, this gives a significant speed-up.
3155
3156
3157 Version 2.05 21-Apr-99
3158 ----------------------
3159
3160 1. Changed the type of magic_number from int to long int so that it works
3161 properly on 16-bit systems.
3162
3163 2. Fixed a bug which caused patterns starting with .* not to work correctly
3164 when the subject string contained newline characters. PCRE was assuming
3165 anchoring for such patterns in all cases, which is not correct because .* will
3166 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3167 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3168 must be retried after every newline in the subject.
3169
3170
3171 Version 2.04 18-Feb-99
3172 ----------------------
3173
3174 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3175 computation of the store needed to hold the pattern was incorrect (too large).
3176 If such patterns were nested a few deep, this could multiply and become a real
3177 problem.
3178
3179 2. Added /M option to pcretest to show the memory requirement of a specific
3180 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3181
3182 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3183 compiled in such a way that the backtracking after subsequent failure was
3184 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3185 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3186
3187
3188 Version 2.03 02-Feb-99
3189 ----------------------
3190
3191 1. Fixed typo and small mistake in man page.
3192
3193 2. Added 4th condition (GPL supersedes if conflict) and created separate
3194 LICENCE file containing the conditions.
3195
3196 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3197 Perl, that is the internal \ allows the delimiter to be included in the
3198 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3199 the final delimiter, add \ to the end of the pattern (to test the error).
3200
3201 4. Added the convenience functions for extracting substrings after a successful
3202 match. Updated pcretest to make it able to test these functions.
3203
3204
3205 Version 2.02 14-Jan-99
3206 ----------------------
3207
3208 1. Initialized the working variables associated with each extraction so that
3209 their saving and restoring doesn't refer to uninitialized store.
3210
3211 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3212 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3213 fix the problem.
3214
3215 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3216 calls, and wasn't printing the correct value for compiling calls. Increased the
3217 default value of LOOPREPEAT, and the number of significant figures in the
3218 times.
3219
3220 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3221
3222 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3223 a building problem on Windows NT with a FAT file system.
3224
3225
3226 Version 2.01 21-Oct-98
3227 ----------------------
3228
3229 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3230 to character tables built by pcre_maketables() in the current locale. If NULL
3231 is passed, the default tables are used.
3232
3233
3234 Version 2.00 24-Sep-98
3235 ----------------------
3236
3237 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3238 it any more.
3239
3240 2. Allow quantification of (?>) groups, and make it work correctly.
3241
3242 3. The first character computation wasn't working for (?>) groups.
3243
3244 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3245 end of the subject) and add 5.005's \z, which really does match only at the
3246 very end of the subject.
3247
3248 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3249
3250 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3251 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3252 localized options. All options to pcre_study() were also removed.
3253
3254 7. Add other new features from 5.005:
3255
3256 $(?<= positive lookbehind
3257 $(?<! negative lookbehind
3258 (?imsx-imsx) added the unsetting capability
3259 such a setting is global if at outer level; local otherwise
3260 (?imsx-imsx:) non-capturing groups with option setting
3261 (?(cond)re|re) conditional pattern matching
3262
3263 A backreference to itself in a repeated group matches the previous
3264 captured string.
3265
3266 8. General tidying up of studying (both automatic and via "study")
3267 consequential on the addition of new assertions.
3268
3269 9. As in 5.005, unlimited repeated groups that could match an empty substring
3270 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3271 runtime if any iteration does actually match an empty substring.
3272
3273 10. Include the RunTest script in the distribution.
3274
3275 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3276 discrepancies, some of which were old and were also with respect to 5.004. They
3277 have now been fixed.
3278
3279
3280 Version 1.09 28-Apr-98
3281 ----------------------
3282
3283 1. A negated single character class followed by a quantifier with a minimum
3284 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3285 program crashes, or just wrong answers. This did not apply to negated classes
3286 containing more than one character, or to minima other than one.
3287
3288
3289 Version 1.08 27-Mar-98
3290 ----------------------
3291
3292 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3293
3294 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3295 latter must appear before anything that relies on it in the pattern.
3296
3297
3298 Version 1.07 16-Feb-98
3299 ----------------------
3300
3301 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3302 repeat of a potentially empty string).
3303
3304
3305 Version 1.06 23-Jan-98
3306 ----------------------
3307
3308 1. Added Markus Oberhumer's little patches for C++.
3309
3310 2. Literal strings longer than 255 characters were broken.
3311
3312
3313 Version 1.05 23-Dec-97
3314 ----------------------
3315
3316 1. Negated character classes containing more than one character were failing if
3317 PCRE_CASELESS was set at run time.
3318
3319
3320 Version 1.04 19-Dec-97
3321 ----------------------
3322
3323 1. Corrected the man page, where some "const" qualifiers had been omitted.
3324
3325 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3326 input syntax.
3327
3328 3. Fixed memory leak which occurred when a regex with back references was
3329 matched with an offsets vector that wasn't big enough. The temporary memory
3330 that is used in this case wasn't being freed if the match failed.
3331
3332 4. Tidied pcretest to ensure it frees memory that it gets.
3333
3334 5. Temporary memory was being obtained in the case where the passed offsets
3335 vector was exactly big enough.
3336
3337 6. Corrected definition of offsetof() from change 5 below.
3338
3339 7. I had screwed up change 6 below and broken the rules for the use of
3340 setjmp(). Now fixed.
3341
3342
3343 Version 1.03 18-Dec-97
3344 ----------------------
3345
3346 1. A erroneous regex with a missing opening parenthesis was correctly
3347 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3348 on some systems.
3349
3350 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3351 it was reported that one broken compiler failed on the former because "code" is
3352 also an independent variable.
3353
3354 3. The erroneous regex a[]b caused an array overrun reference.
3355
3356 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3357 fail on data ending with that character. (It was going on too far, and checking
3358 the next character, typically a binary zero.) This was specific to the
3359 optimized code for single-character negative classes.
3360
3361 5. Added a contributed patch from the TIN world which does the following:
3362
3363 + Add an undef for memmove, in case the the system defines a macro for it.
3364
3365 + Add a definition of offsetof(), in case there isn't one. (I don't know
3366 the reason behind this - offsetof() is part of the ANSI standard - but
3367 it does no harm).
3368
3369 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3370 most of the places where whitespace preceded '#'. I have given up and
3371 allowed the remaining 2 cases to be at the margin.
3372
3373 + Rename some variables in pcre to eliminate shadowing. This seems very
3374 pedantic, but does no harm, of course.
3375
3376 6. Moved the call to setjmp() into its own function, to get rid of warnings
3377 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3378
3379 7. Constructs such as \d{8,} were compiling into the equivalent of
3380 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3381 outcome, but in this particular case used more store than had been allocated,
3382 which caused the bug to be discovered because it threw up an internal error.
3383
3384 8. The debugging code in both pcre and pcretest for outputting the compiled
3385 form of a regex was going wrong in the case of back references followed by
3386 curly-bracketed repeats.
3387
3388
3389 Version 1.02 12-Dec-97
3390 ----------------------
3391
3392 1. Typos in pcre.3 and comments in the source fixed.
3393
3394 2. Applied a contributed patch to get rid of places where it used to remove
3395 'const' from variables, and fixed some signed/unsigned and uninitialized
3396 variable warnings.
3397
3398 3. Added the "runtest" target to Makefile.
3399
3400 4. Set default compiler flag to -O2 rather than just -O.
3401
3402
3403 Version 1.01 19-Nov-97
3404 ----------------------
3405
3406 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3407 like /([ab]*)*/, that is, for classes with more than one character in them.
3408
3409 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3410 as /((?>a*))*/ (a PCRE_EXTRA facility).
3411
3412
3413 Version 1.00 18-Nov-97
3414 ----------------------
3415
3416 1. Added compile-time macros to support systems such as SunOS4 which don't have
3417 memmove() or strerror() but have other things that can be used instead.
3418
3419 2. Arranged that "make clean" removes the executables.
3420
3421
3422 Version 0.99 27-Oct-97
3423 ----------------------
3424
3425 1. Fixed bug in code for optimizing classes with only one character. It was
3426 initializing a 32-byte map regardless, which could cause it to run off the end
3427 of the memory it had got.
3428
3429 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3430
3431
3432 Version 0.98 22-Oct-97
3433 ----------------------
3434
3435 1. Fixed bug in code for handling temporary memory usage when there are more
3436 back references than supplied space in the ovector. This could cause segfaults.
3437
3438
3439 Version 0.97 21-Oct-97
3440 ----------------------
3441
3442 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3443
3444 2. Optimized negated single characters not to use a bit map.
3445
3446 3. Brought error texts together as macro definitions; clarified some of them;
3447 fixed one that was wrong - it said "range out of order" when it meant "invalid
3448 escape sequence".
3449
3450 4. Changed some char * arguments to const char *.
3451
3452 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3453
3454 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3455 pcretest.
3456
3457
3458 Version 0.96 16-Oct-97
3459 ----------------------
3460
3461 1. Added a simple "pgrep" utility to the distribution.
3462
3463 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3464 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3465 where "ddd" means "one or more decimal digits".
3466
3467 3. Fixed serious bug. If a pattern had a back reference, but the call to
3468 pcre_exec() didn't supply a large enough ovector to record the related
3469 identifying subpattern, the match always failed. PCRE now remembers the number
3470 of the largest back reference, and gets some temporary memory in which to save
3471 the offsets during matching if necessary, in order to ensure that
3472 backreferences always work.
3473
3474 4. Increased the compatibility with Perl in a number of ways:
3475
3476 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3477 to request this handling. The option can be set at compile or exec time.
3478
3479 (b) $ matches before a terminating newline by default; an option
3480 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3481 mode). The option can be set at compile or exec time.
3482
3483 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3484 the same as Perl's. If the decimal number it represents is less than 10
3485 or there aren't that many previous left capturing parentheses, an octal
3486 escape is read. Inside a character class, it's always an octal escape,
3487 even if it is a single digit.
3488
3489 (d) An escaped but undefined alphabetic character is taken as a literal,
3490 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3491 escapes.
3492
3493 (e) {0} is now permitted. (The previous item is removed from the compiled
3494 pattern).
3495
3496 5. Changed all the names of code files so that the basic parts are no longer
3497 than 10 characters, and abolished the teeny "globals.c" file.
3498
3499 6. Changed the handling of character classes; they are now done with a 32-byte
3500 bit map always.
3501
3502 7. Added the -d and /D options to pcretest to make it possible to look at the
3503 internals of compilation without having to recompile pcre.
3504
3505
3506 Version 0.95 23-Sep-97
3507 ----------------------
3508
3509 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3510 \x20 at the start of a run of normal characters. These were being treated as
3511 real characters, instead of the source characters being re-checked.
3512
3513
3514 Version 0.94 18-Sep-97
3515 ----------------------
3516
3517 1. The functions are now thread-safe, with the caveat that the global variables
3518 containing pointers to malloc() and free() or alternative functions are the
3519 same for all threads.
3520
3521 2. Get pcre_study() to generate a bitmap of initial characters for non-
3522 anchored patterns when this is possible, and use it if passed to pcre_exec().
3523
3524
3525 Version 0.93 15-Sep-97
3526 ----------------------
3527
3528 1. /(b)|(:+)/ was computing an incorrect first character.
3529
3530 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3531 but not actually doing anything yet.
3532
3533 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3534 as Perl does (e.g. [-az] or [az-]).
3535
3536 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3537 all possible positions.
3538
3539 5. Split up into different modules to avoid including unneeded functions in a
3540 compiled binary. However, compile and exec are still in one module. The "study"
3541 function is split off.
3542
3543 6. The character tables are now in a separate module whose source is generated
3544 by an auxiliary program - but can then be edited by hand if required. There are
3545 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3546 toupper() in the code.
3547
3548 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3549 make them global. Abolish the function for setting them, as the caller can now
3550 set them directly.
3551
3552
3553 Version 0.92 11-Sep-97
3554 ----------------------
3555
3556 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3557 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3558
3559 2. Caseless matching was not working in character classes if the characters in
3560 the pattern were in upper case.
3561
3562 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3563
3564 4. Make PCRE_ANCHORED public and accept as a compile option.
3565
3566 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3567 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3568 pass them.
3569
3570 6. Give an error if bad option bits passed at compile or run time.
3571
3572 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3573 pcretest to cause it to pass that flag.
3574
3575 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3576 options, and the first character, if set.
3577
3578 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3579
3580
3581 Version 0.91 10-Sep-97
3582 ----------------------
3583
3584 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3585 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3586
3587 2. PCRE was looping on encountering an indefinitely repeated back reference to
3588 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3589 Perl does - treats the match as successful.
3590
3591 ****

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5