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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 438 - (show annotations) (download)
Sun Sep 6 20:00:47 2009 UTC (4 years, 10 months ago) by ph10
File size: 171937 byte(s)
Fix internal error for forward reference with [^m] interposing.

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8 was out-of-date, and there was no check on the pcre_compile() error code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

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