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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 448 - (show annotations) (download)
Tue Sep 15 19:11:48 2009 UTC (5 years, 2 months ago) by ph10
File size: 172900 byte(s)
Re-organize tests to make is possible to test 5.10 features against Perl.

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8 was out-of-date, and there was no check on the pcre_compile() error code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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12