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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 433 - (show annotations)
Wed Sep 2 16:09:13 2009 UTC (5 years, 8 months ago) by ph10
File size: 170612 byte(s)
Error occurred while calculating annotation data.
For regexec(), if pmatch is NULL, force nmatch to be zero.
1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.00 ??-???-??
5 ----------------------
6
7 1. The table for translating pcre_compile() error codes into POSIX error codes
8 was out-of-date, and there was no check on the pcre_compile() error code
9 being within the table. This could lead to an OK return being given in
10 error.
11
12 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
13 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
14 in a Windows environment.
15
16 3. The pcregrep --count option prints the count for each file even when it is
17 zero, as does GNU grep. However, pcregrep was also printing all files when
18 --files-with-matches was added. Now, when both options are given, it prints
19 counts only for those files that have at least one match. (GNU grep just
20 prints the file name in this circumstance, but including the count seems
21 more useful - otherwise, why use --count?) Also ensured that the
22 combination -clh just lists non-zero counts, with no names.
23
24 4. The long form of the pcregrep -F option was incorrectly implemented as
25 --fixed_strings instead of --fixed-strings. This is an incompatible change,
26 but it seems right to fix it, and I didn't think it was worth preserving
27 the old behaviour.
28
29 5. The command line items --regex=pattern and --regexp=pattern were not
30 recognized by pcregrep, which required --regex pattern or --regexp pattern
31 (with a space rather than an '='). The man page documented the '=' forms,
32 which are compatible with GNU grep; these now work.
33
34 6. No libpcreposix.pc file was created for pkg-config; there was just
35 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
36
37 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
38 when UCP support is not needed, by modifying the Python script that
39 generates it from Unicode data files. This should not matter if the module
40 is correctly used as a library, but I received one complaint about 50K of
41 unwanted data. My guess is that the person linked everything into his
42 program rather than using a library. Anyway, it does no harm.
43
44 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
45 was a minimum greater than 1 for a wide character in a possessive
46 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
47 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
48 character. Chaos in the form of incorrect output or a compiling loop could
49 result.
50
51 9. The restrictions on what a pattern can contain when partial matching is
52 requested for pcre_exec() have been removed. All patterns can now be
53 partially matched by this function. In addition, if there are at least two
54 slots in the offset vector, the offsets of the first-encountered partial
55 match are set in them when PCRE_ERROR_PARTIAL is returned.
56
57 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
58 synonymous with PCRE_PARTIAL, for backwards compatibility, and
59 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
60 and may be more useful for multi-segment matching, especially with
61 pcre_exec().
62
63 11. Partial matching with pcre_exec() is now more intuitive. A partial match
64 used to be given if ever the end of the subject was reached; now it is
65 given only if matching could not proceed because another character was
66 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
67 string "Z", which now yields "no match" instead of "partial match". In the
68 case of pcre_dfa_exec(), "no match" is given if every matching path for the
69 final character ended with (*FAIL).
70
71 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
72 if the pattern had a "must contain" character that was already found in the
73 earlier partial match, unless partial matching was again requested. For
74 example, with the pattern /dog.(body)?/, the "must contain" character is
75 "g". If the first part-match was for the string "dog", restarting with
76 "sbody" failed.
77
78 13. Added a pcredemo man page, created automatically from the pcredemo.c file,
79 so that the demonstration program is easily available in environments where
80 PCRE has not been installed from source.
81
82 14. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
83 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
84 library.
85
86 15. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
87 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
88 is not the first non-POSIX option to be added. Clearly some people find
89 these options useful.
90
91 16. If a caller to the POSIX matching function regexec() passes a non-zero
92 value for \fInmatch\fP with a NULL value for \fIpmatch\fP, the value of
93 \fInmatch\fP is forced to zero.
94
95
96 Version 7.9 11-Apr-09
97 ---------------------
98
99 1. When building with support for bzlib/zlib (pcregrep) and/or readline
100 (pcretest), all targets were linked against these libraries. This included
101 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
102 libraries. This caused unwanted dependencies to be created. This problem
103 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
104 pcretest is linked with readline.
105
106 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
107 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
108 moved outside it again, because FALSE and TRUE are already defined in AIX,
109 but BOOL is not.
110
111 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
112 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
113
114 4. The pcregrep documentation said spaces were inserted as well as colons (or
115 hyphens) following file names and line numbers when outputting matching
116 lines. This is not true; no spaces are inserted. I have also clarified the
117 wording for the --colour (or --color) option.
118
119 5. In pcregrep, when --colour was used with -o, the list of matching strings
120 was not coloured; this is different to GNU grep, so I have changed it to be
121 the same.
122
123 6. When --colo(u)r was used in pcregrep, only the first matching substring in
124 each matching line was coloured. Now it goes on to look for further matches
125 of any of the test patterns, which is the same behaviour as GNU grep.
126
127 7. A pattern that could match an empty string could cause pcregrep to loop; it
128 doesn't make sense to accept an empty string match in pcregrep, so I have
129 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
130 seems to be how GNU grep behaves.
131
132 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
133 start or after a newline", because the conditional assertion was not being
134 correctly handled. The rule now is that both the assertion and what follows
135 in the first alternative must satisfy the test.
136
137 9. If auto-callout was enabled in a pattern with a conditional group whose
138 condition was an assertion, PCRE could crash during matching, both with
139 pcre_exec() and pcre_dfa_exec().
140
141 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
142 used for matching.
143
144 11. Unicode property support in character classes was not working for
145 characters (bytes) greater than 127 when not in UTF-8 mode.
146
147 12. Added the -M command line option to pcretest.
148
149 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
150
151 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
152
153 16. Added comments and documentation about mis-use of no_arg in the C++
154 wrapper.
155
156 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
157 from Martin Jerabek that uses macro names for all relevant character and
158 string constants.
159
160 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
161 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
162 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
163 these, but not everybody uses configure.
164
165 19. A conditional group that had only one branch was not being correctly
166 recognized as an item that could match an empty string. This meant that an
167 enclosing group might also not be so recognized, causing infinite looping
168 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
169 with the subject "ab", where knowledge that the repeated group can match
170 nothing is needed in order to break the loop.
171
172 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
173 exec(), but without supplying a callout function, matching went wrong.
174
175 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
176 leak if the size of the offset vector was greater than 30. When the vector
177 is smaller, the saved offsets during recursion go onto a local stack
178 vector, but for larger vectors malloc() is used. It was failing to free
179 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
180 error, in fact).
181
182 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
183 heapframe that is used only when UTF-8 support is enabled. This caused no
184 problem, but was untidy.
185
186 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
187 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
188 included within another project.
189
190 24. Steven Van Ingelgem's patches to add more options to the CMake support,
191 slightly modified by me:
192
193 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
194 not building pcregrep.
195
196 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
197 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
198
199 25. Forward references, both numeric and by name, in patterns that made use of
200 duplicate group numbers, could behave incorrectly or give incorrect errors,
201 because when scanning forward to find the reference group, PCRE was not
202 taking into account the duplicate group numbers. A pattern such as
203 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
204
205 26. Changed a few more instances of "const unsigned char *" to USPTR, making
206 the feature of a custom pointer more persuasive (as requested by a user).
207
208 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
209 pcretest.c, inside #ifndefs, because it seems they are sometimes already
210 pre-defined.
211
212 28. Added support for (*UTF8) at the start of a pattern.
213
214 29. Arrange for flags added by the "release type" setting in CMake to be shown
215 in the configuration summary.
216
217
218 Version 7.8 05-Sep-08
219 ---------------------
220
221 1. Replaced UCP searching code with optimized version as implemented for Ad
222 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
223 stage table and inline lookup instead of a function, giving speed ups of 2
224 to 5 times on some simple patterns that I tested. Permission was given to
225 distribute the MultiStage2.py script that generates the tables (it's not in
226 the tarball, but is in the Subversion repository).
227
228 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
229 scripts.
230
231 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
232 a group with a zero qualifier. The result of the study could be incorrect,
233 or the function might crash, depending on the pattern.
234
235 4. Caseless matching was not working for non-ASCII characters in back
236 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
237 It now works when Unicode Property Support is available.
238
239 5. In pcretest, an escape such as \x{de} in the data was always generating
240 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
241 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
242 truncation.
243
244 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
245
246 7. Added two (int) casts to pcregrep when printing the difference of two
247 pointers, in case they are 64-bit values.
248
249 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
250 test 2 if it fails.
251
252 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
253 and a #define of that name to empty if it is not externally set. This is to
254 allow users of MSVC to set it if necessary.
255
256 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
257 the convenience functions in the pcre_get.c source file.
258
259 11. An option change at the start of a pattern that had top-level alternatives
260 could cause overwriting and/or a crash. This command provoked a crash in
261 some environments:
262
263 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
264
265 This potential security problem was recorded as CVE-2008-2371.
266
267 12. For a pattern where the match had to start at the beginning or immediately
268 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
269 pcre_dfa_exec() could read past the end of the passed subject if there was
270 no match. To help with detecting such bugs (e.g. with valgrind), I modified
271 pcretest so that it places the subject at the end of its malloc-ed buffer.
272
273 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
274 exec() might read past the end of the data buffer in UTF-8 mode.
275
276 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
277 the data contained the byte 0x85 as part of a UTF-8 character within its
278 first line. This applied both to normal and DFA matching.
279
280 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
281 /^[^d]*?$/8 failed to match "abc".
282
283 16. Added a missing copyright notice to pcrecpp_internal.h.
284
285 17. Make it more clear in the documentation that values returned from
286 pcre_exec() in ovector are byte offsets, not character counts.
287
288 18. Tidied a few places to stop certain compilers from issuing warnings.
289
290 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
291 supplied by Stefan Weber. I made a further small update for 7.8 because
292 there is a change of source arrangements: the pcre_searchfuncs.c module is
293 replaced by pcre_ucd.c.
294
295
296 Version 7.7 07-May-08
297 ---------------------
298
299 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
300 a string to a long long, pretend we don't even have a long long." This is
301 done by checking for the strtoq, strtoll, and _strtoi64 functions.
302
303 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
304 pre-7.6 versions, which defined a global no_arg variable instead of putting
305 it in the RE class. (See also #8 below.)
306
307 3. Remove a line of dead code, identified by coverity and reported by Nuno
308 Lopes.
309
310 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
311
312 (1) The include/exclude patterns were being applied to the whole pathnames
313 of files, instead of just to the final components.
314
315 (2) If there was more than one level of directory, the subdirectories were
316 skipped unless they satisfied the include/exclude conditions. This is
317 inconsistent with GNU grep (and could even be seen as contrary to the
318 pcregrep specification - which I improved to make it absolutely clear).
319 The action now is always to scan all levels of directory, and just
320 apply the include/exclude patterns to regular files.
321
322 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
323 --exclude_dir in the tests to avoid scanning .svn directories.
324
325 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
326 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
327 doesn't support NULs in patterns.
328
329 7. Added some missing "const"s to declarations of static tables in
330 pcre_compile.c and pcre_dfa_exec.c.
331
332 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
333 caused by fix #2 above. (Subsequently also a second patch to fix the
334 first patch. And a third patch - this was a messy problem.)
335
336 9. Applied Craig's patch to remove the use of push_back().
337
338 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
339 matching function regexec().
340
341 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
342 which, however, unlike Perl's \g{...}, are subroutine calls, not back
343 references. PCRE supports relative numbers with this syntax (I don't think
344 Oniguruma does).
345
346 12. Previously, a group with a zero repeat such as (...){0} was completely
347 omitted from the compiled regex. However, this means that if the group
348 was called as a subroutine from elsewhere in the pattern, things went wrong
349 (an internal error was given). Such groups are now left in the compiled
350 pattern, with a new opcode that causes them to be skipped at execution
351 time.
352
353 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
354 to the way PCRE behaves:
355
356 (a) A lone ] character is dis-allowed (Perl treats it as data).
357
358 (b) A back reference to an unmatched subpattern matches an empty string
359 (Perl fails the current match path).
360
361 (c) A data ] in a character class must be notated as \] because if the
362 first data character in a class is ], it defines an empty class. (In
363 Perl it is not possible to have an empty class.) The empty class []
364 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
365 The negative empty class [^] matches any one character, independently
366 of the DOTALL setting.
367
368 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
369 non-existent subpattern following a character class starting with ']' and
370 containing () gave an internal compiling error instead of "reference to
371 non-existent subpattern". Fortunately, when the pattern did exist, the
372 compiled code was correct. (When scanning forwards to check for the
373 existencd of the subpattern, it was treating the data ']' as terminating
374 the class, so got the count wrong. When actually compiling, the reference
375 was subsequently set up correctly.)
376
377 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
378 it was being rejected as not supported by pcre_dfa_exec(), even though
379 other assertions are supported. I have made pcre_dfa_exec() support
380 (*FAIL).
381
382 16. The implementation of 13c above involved the invention of a new opcode,
383 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
384 cannot be changed at match time, I realized I could make a small
385 improvement to matching performance by compiling OP_ALLANY instead of
386 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
387 on the OP_ANY path.
388
389 17. Compiling pcretest on Windows with readline support failed without the
390 following two fixes: (1) Make the unistd.h include conditional on
391 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
392
393 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
394 ncurses library to be included for pcretest when ReadLine support is
395 requested, but also to allow for it to be overridden. This patch came from
396 Daniel Bergström.
397
398 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
399 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
400 any errors with the current Unicode tables. Thanks to Peter Kankowski for
401 spotting this.
402
403
404 Version 7.6 28-Jan-08
405 ---------------------
406
407 1. A character class containing a very large number of characters with
408 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
409 overflow.
410
411 2. Patch to cut out the "long long" test in pcrecpp_unittest when
412 HAVE_LONG_LONG is not defined.
413
414 3. Applied Christian Ehrlicher's patch to update the CMake build files to
415 bring them up to date and include new features. This patch includes:
416
417 - Fixed PH's badly added libz and libbz2 support.
418 - Fixed a problem with static linking.
419 - Added pcredemo. [But later removed - see 7 below.]
420 - Fixed dftables problem and added an option.
421 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
422 HAVE_LONG_LONG.
423 - Added readline support for pcretest.
424 - Added an listing of the option settings after cmake has run.
425
426 4. A user submitted a patch to Makefile that makes it easy to create
427 "pcre.dll" under mingw when using Configure/Make. I added stuff to
428 Makefile.am that cause it to include this special target, without
429 affecting anything else. Note that the same mingw target plus all
430 the other distribution libraries and programs are now supported
431 when configuring with CMake (see 6 below) instead of with
432 Configure/Make.
433
434 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
435 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
436 exported in the Windows port". It has not yet been confirmed that the patch
437 solves the problem, but it does no harm.
438
439 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
440 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
441 with CMake, and also correct the comment about stack recursion.
442
443 7. Remove the automatic building of pcredemo from the ./configure system and
444 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
445 of a program that users should build themselves after PCRE is installed, so
446 building it automatically is not really right. What is more, it gave
447 trouble in some build environments.
448
449 8. Further tidies to CMakeLists.txt from Sheri and Christian.
450
451
452 Version 7.5 10-Jan-08
453 ---------------------
454
455 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
456 values in parens when parsing an RE using the C++ wrapper."
457
458 2. Negative specials like \S did not work in character classes in UTF-8 mode.
459 Characters greater than 255 were excluded from the class instead of being
460 included.
461
462 3. The same bug as (2) above applied to negated POSIX classes such as
463 [:^space:].
464
465 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
466 defined or documented. It seems to have been a typo for PCRE_STATIC, so
467 I have changed it.
468
469 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
470 first named subpattern) and a construct such as (?&a) would reference the
471 first named subpattern whose name started with "a" (in other words, the
472 length check was missing). Both these problems are fixed. "Subpattern name
473 expected" is now given for (?&) (a zero-length name), and this patch also
474 makes it give the same error for \k'' (previously it complained that that
475 was a reference to a non-existent subpattern).
476
477 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
478 this is right because (?- can be followed by option settings as well as by
479 digits. I have, however, made the messages clearer.
480
481 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
482 than the number used in the conditional) now cause a compile-time error.
483 This is actually not compatible with Perl, which accepts such patterns, but
484 treats the conditional as always being FALSE (as PCRE used to), but it
485 seems to me that giving a diagnostic is better.
486
487 8. Change "alphameric" to the more common word "alphanumeric" in comments
488 and messages.
489
490 9. Fix two occurrences of "backslash" in comments that should have been
491 "backspace".
492
493 10. Remove two redundant lines of code that can never be obeyed (their function
494 was moved elsewhere).
495
496 11. The program that makes PCRE's Unicode character property table had a bug
497 which caused it to generate incorrect table entries for sequences of
498 characters that have the same character type, but are in different scripts.
499 It amalgamated them into a single range, with the script of the first of
500 them. In other words, some characters were in the wrong script. There were
501 thirteen such cases, affecting characters in the following ranges:
502
503 U+002b0 - U+002c1
504 U+0060c - U+0060d
505 U+0061e - U+00612
506 U+0064b - U+0065e
507 U+0074d - U+0076d
508 U+01800 - U+01805
509 U+01d00 - U+01d77
510 U+01d9b - U+01dbf
511 U+0200b - U+0200f
512 U+030fc - U+030fe
513 U+03260 - U+0327f
514 U+0fb46 - U+0fbb1
515 U+10450 - U+1049d
516
517 12. The -o option (show only the matching part of a line) for pcregrep was not
518 compatible with GNU grep in that, if there was more than one match in a
519 line, it showed only the first of them. It now behaves in the same way as
520 GNU grep.
521
522 13. If the -o and -v options were combined for pcregrep, it printed a blank
523 line for every non-matching line. GNU grep prints nothing, and pcregrep now
524 does the same. The return code can be used to tell if there were any
525 non-matching lines.
526
527 14. Added --file-offsets and --line-offsets to pcregrep.
528
529 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
530 infinitely looping recursion. The bug was that positive lookaheads were not
531 being skipped when checking for a possible empty match (negative lookaheads
532 and both kinds of lookbehind were skipped).
533
534 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
535 inclusion of <windows.h> to before rather than after the definition of
536 INVALID_FILE_ATTRIBUTES (patch from David Byron).
537
538 17. Specifying a possessive quantifier with a specific limit for a Unicode
539 character property caused pcre_compile() to compile bad code, which led at
540 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
541 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
542 caused the error; without that there was no problem.
543
544 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
545
546 19. Added --enable-pcretest-libreadline.
547
548 20. In pcrecpp.cc, the variable 'count' was incremented twice in
549 RE::GlobalReplace(). As a result, the number of replacements returned was
550 double what it should be. I removed one of the increments, but Craig sent a
551 later patch that removed the other one (the right fix) and added unit tests
552 that check the return values (which was not done before).
553
554 21. Several CMake things:
555
556 (1) Arranged that, when cmake is used on Unix, the libraries end up with
557 the names libpcre and libpcreposix, not just pcre and pcreposix.
558
559 (2) The above change means that pcretest and pcregrep are now correctly
560 linked with the newly-built libraries, not previously installed ones.
561
562 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
563
564 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
565 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
566 UTF-8 newline character). The key issue is that the pattern starts .*;
567 this means that the match must be either at the beginning, or after a
568 newline. The bug was in the code for advancing after a failed match and
569 checking that the new position followed a newline. It was not taking
570 account of UTF-8 characters correctly.
571
572 23. PCRE was behaving differently from Perl in the way it recognized POSIX
573 character classes. PCRE was not treating the sequence [:...:] as a
574 character class unless the ... were all letters. Perl, however, seems to
575 allow any characters between [: and :], though of course it rejects as
576 unknown any "names" that contain non-letters, because all the known class
577 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
578 for example, whereas PCRE did not - it did not recognize a POSIX character
579 class. This seemed a bit dangerous, so the code has been changed to be
580 closer to Perl. The behaviour is not identical to Perl, because PCRE will
581 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
582 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
583 Perl does, and where it didn't before.
584
585 24. Rewrite so as to remove the single use of %n from pcregrep because in some
586 Windows environments %n is disabled by default.
587
588
589 Version 7.4 21-Sep-07
590 ---------------------
591
592 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
593 means that a class such as [\s] counted as "explicit reference to CR or
594 LF". That isn't really right - the whole point of the change was to try to
595 help when there was an actual mention of one of the two characters. So now
596 the change happens only if \r or \n (or a literal CR or LF) character is
597 encountered.
598
599 2. The 32-bit options word was also used for 6 internal flags, but the numbers
600 of both had grown to the point where there were only 3 bits left.
601 Fortunately, there was spare space in the data structure, and so I have
602 moved the internal flags into a new 16-bit field to free up more option
603 bits.
604
605 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
606 but did not set the internal JCHANGED flag - either of these is enough to
607 control the way the "get" function works - but the PCRE_INFO_JCHANGED
608 facility is supposed to tell if (?J) was ever used, so now (?J) at the
609 start sets both bits.
610
611 4. Added options (at build time, compile time, exec time) to change \R from
612 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
613
614 5. doc/pcresyntax.html was missing from the distribution.
615
616 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
617 compatibility, even though it is no longer used.
618
619 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
620 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
621 windows.h file is present (where different names are used). [This was
622 reversed later after testing - see 16 below.]
623
624 8. Changed all #include <config.h> to #include "config.h". There were also
625 some further <pcre.h> cases that I changed to "pcre.h".
626
627 9. When pcregrep was used with the --colour option, it missed the line ending
628 sequence off the lines that it output.
629
630 10. It was pointed out to me that arrays of string pointers cause lots of
631 relocations when a shared library is dynamically loaded. A technique of
632 using a single long string with a table of offsets can drastically reduce
633 these. I have refactored PCRE in four places to do this. The result is
634 dramatic:
635
636 Originally: 290
637 After changing UCP table: 187
638 After changing error message table: 43
639 After changing table of "verbs" 36
640 After changing table of Posix names 22
641
642 Thanks to the folks working on Gregex for glib for this insight.
643
644 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
645 unicode-properties was also set.
646
647 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
648
649 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
650 checked only for CRLF.
651
652 14. Added casts to pcretest.c to avoid compiler warnings.
653
654 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
655
656 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
657 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
658 entirely. This removes changes made in 7 above.
659
660 17. The CMake files have been updated, and there is now more information about
661 building with CMake in the NON-UNIX-USE document.
662
663
664 Version 7.3 28-Aug-07
665 ---------------------
666
667 1. In the rejigging of the build system that eventually resulted in 7.1, the
668 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
669 brackets there is not right, since it causes compilers to look for an
670 installed pcre.h, not the version that is in the source that is being
671 compiled (which of course may be different). I have changed it back to:
672
673 #include "pcre.h"
674
675 I have a vague recollection that the change was concerned with compiling in
676 different directories, but in the new build system, that is taken care of
677 by the VPATH setting the Makefile.
678
679 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
680 when the subject happened to end in the byte 0x85 (e.g. if the last
681 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
682 characters but of course it shouldn't be taken as a newline when it is part
683 of another character. The bug was that, for an unlimited repeat of . in
684 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
685 characters when looking for a newline.
686
687 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
688
689 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
690 in debug output.
691
692 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
693 long printing in the pcrecpp unittest when running under MinGW.
694
695 6. ESC_K was left out of the EBCDIC table.
696
697 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
698 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
699 limit also applies to "virtual nesting" when a pattern is recursive, and in
700 this case 1000 isn't so big. I have been able to remove this limit at the
701 expense of backing off one optimization in certain circumstances. Normally,
702 when pcre_exec() would call its internal match() function recursively and
703 immediately return the result unconditionally, it uses a "tail recursion"
704 feature to save stack. However, when a subpattern that can match an empty
705 string has an unlimited repetition quantifier, it no longer makes this
706 optimization. That gives it a stack frame in which to save the data for
707 checking that an empty string has been matched. Previously this was taken
708 from the 1000-entry workspace that had been reserved. So now there is no
709 explicit limit, but more stack is used.
710
711 8. Applied Daniel's patches to solve problems with the import/export magic
712 syntax that is required for Windows, and which was going wrong for the
713 pcreposix and pcrecpp parts of the library. These were overlooked when this
714 problem was solved for the main library.
715
716 9. There were some crude static tests to avoid integer overflow when computing
717 the size of patterns that contain repeated groups with explicit upper
718 limits. As the maximum quantifier is 65535, the maximum group length was
719 set at 30,000 so that the product of these two numbers did not overflow a
720 32-bit integer. However, it turns out that people want to use groups that
721 are longer than 30,000 bytes (though not repeat them that many times).
722 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
723 made it possible to implement the integer overflow checks in a much more
724 dynamic way, which I have now done. The artificial limitation on group
725 length has been removed - we now have only the limit on the total length of
726 the compiled pattern, which depends on the LINK_SIZE setting.
727
728 10. Fixed a bug in the documentation for get/copy named substring when
729 duplicate names are permitted. If none of the named substrings are set, the
730 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
731 empty string.
732
733 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
734 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
735 because the ] is interpreted as the first data character and the
736 terminating ] is not found. PCRE has been made compatible with Perl in this
737 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
738 cause memory overwriting.
739
740 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
741 string has been matched (to stop an infinite loop). It was not recognizing
742 a conditional subpattern that could match an empty string if that
743 subpattern was within another subpattern. For example, it looped when
744 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
745 condition was not nested. This bug has been fixed.
746
747 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
748 past the start of the subject in the presence of bytes with the top bit
749 set, for example "\x8aBCD".
750
751 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
752 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
753
754 14. Optimized (?!) to (*FAIL).
755
756 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
757 This restricts code points to be within the range 0 to 0x10FFFF, excluding
758 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
759 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
760 does: it's just the validity check that is more restrictive.
761
762 16. Inserted checks for integer overflows during escape sequence (backslash)
763 processing, and also fixed erroneous offset values for syntax errors during
764 backslash processing.
765
766 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
767 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
768
769 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
770 caused an overrun.
771
772 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
773 something other than just ASCII characters) inside a group that had an
774 unlimited repeat caused a loop at compile time (while checking to see
775 whether the group could match an empty string).
776
777 20. Debugging a pattern containing \p or \P could cause a crash. For example,
778 [\P{Any}] did so. (Error in the code for printing property names.)
779
780 21. An orphan \E inside a character class could cause a crash.
781
782 22. A repeated capturing bracket such as (A)? could cause a wild memory
783 reference during compilation.
784
785 23. There are several functions in pcre_compile() that scan along a compiled
786 expression for various reasons (e.g. to see if it's fixed length for look
787 behind). There were bugs in these functions when a repeated \p or \P was
788 present in the pattern. These operators have additional parameters compared
789 with \d, etc, and these were not being taken into account when moving along
790 the compiled data. Specifically:
791
792 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
793 length.
794
795 (b) An item such as \pL+ within a repeated group could cause crashes or
796 loops.
797
798 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
799 "reference to non-existent subpattern" error.
800
801 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
802
803 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
804 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
805
806 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
807
808 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
809 character were causing crashes (broken optimization).
810
811 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
812 \p or \P) caused a compile-time loop.
813
814 28. More problems have arisen in unanchored patterns when CRLF is a valid line
815 break. For example, the unstudied pattern [\r\n]A does not match the string
816 "\r\nA" because change 7.0/46 below moves the current point on by two
817 characters after failing to match at the start. However, the pattern \nA
818 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
819 the same is true. There doesn't seem any very clean way out of this, but
820 what I have chosen to do makes the common cases work: PCRE now takes note
821 of whether there can be an explicit match for \r or \n anywhere in the
822 pattern, and if so, 7.0/46 no longer applies. As part of this change,
823 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
824 pattern has explicit CR or LF references.
825
826 29. Added (*CR) etc for changing newline setting at start of pattern.
827
828
829 Version 7.2 19-Jun-07
830 ---------------------
831
832 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
833 which is apparently normally available under Windows.
834
835 2. Re-jig the pcregrep tests with different newline settings in an attempt
836 to make them independent of the local environment's newline setting.
837
838 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
839
840 4. Some of the "internals" tests were previously cut out when the link size
841 was not 2, because the output contained actual offsets. The recent new
842 "Z" feature of pcretest means that these can be cut out, making the tests
843 usable with all link sizes.
844
845 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
846 stack recursion. This gives a massive performance boost under BSD, but just
847 a small improvement under Linux. However, it saves one field in the frame
848 in all cases.
849
850 6. Added more features from the forthcoming Perl 5.10:
851
852 (a) (?-n) (where n is a string of digits) is a relative subroutine or
853 recursion call. It refers to the nth most recently opened parentheses.
854
855 (b) (?+n) is also a relative subroutine call; it refers to the nth next
856 to be opened parentheses.
857
858 (c) Conditions that refer to capturing parentheses can be specified
859 relatively, for example, (?(-2)... or (?(+3)...
860
861 (d) \K resets the start of the current match so that everything before
862 is not part of it.
863
864 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
865
866 (f) \g{name} is another synonym - part of Perl 5.10's unification of
867 reference syntax.
868
869 (g) (?| introduces a group in which the numbering of parentheses in each
870 alternative starts with the same number.
871
872 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
873
874 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
875 PCRE_INFO_JCHANGED.
876
877 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
878 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
879 for detecting groups that can match an empty string.
880
881 9. A pattern with a very large number of alternatives (more than several
882 hundred) was running out of internal workspace during the pre-compile
883 phase, where pcre_compile() figures out how much memory will be needed. A
884 bit of new cunning has reduced the workspace needed for groups with
885 alternatives. The 1000-alternative test pattern now uses 12 bytes of
886 workspace instead of running out of the 4096 that are available.
887
888 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
889
890 11. Applied patch from Google to remove an optimization that didn't quite work.
891 The report of the bug said:
892
893 pcrecpp::RE("a*").FullMatch("aaa") matches, while
894 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
895 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
896
897 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
898 it matched the wrong number of bytes.
899
900
901 Version 7.1 24-Apr-07
902 ---------------------
903
904 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
905 that is more "standard", making use of automake and other Autotools. There
906 is some re-arrangement of the files and adjustment of comments consequent
907 on this.
908
909 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
910 for recursive directory scanning broke on some systems because the files
911 are not scanned in any specific order and on different systems the order
912 was different. A call to "sort" has been inserted into RunGrepTest for the
913 approprate test as a short-term fix. In the longer term there may be an
914 alternative.
915
916 3. I had an email from Eric Raymond about problems translating some of PCRE's
917 man pages to HTML (despite the fact that I distribute HTML pages, some
918 people do their own conversions for various reasons). The problems
919 concerned the use of low-level troff macros .br and .in. I have therefore
920 removed all such uses from the man pages (some were redundant, some could
921 be replaced by .nf/.fi pairs). The 132html script that I use to generate
922 HTML has been updated to handle .nf/.fi and to complain if it encounters
923 .br or .in.
924
925 4. Updated comments in configure.ac that get placed in config.h.in and also
926 arranged for config.h to be included in the distribution, with the name
927 config.h.generic, for the benefit of those who have to compile without
928 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
929
930 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
931 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
932 makevp.bat for the current PCRE, using the additional files
933 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
934
935 6. A Windows user reported a minor discrepancy with test 2, which turned out
936 to be caused by a trailing space on an input line that had got lost in his
937 copy. The trailing space was an accident, so I've just removed it.
938
939 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
940 that is needed.
941
942 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
943 as "const" (a) because they are and (b) because it helps the PHP
944 maintainers who have recently made a script to detect big data structures
945 in the php code that should be moved to the .rodata section. I remembered
946 to update Builducptable as well, so it won't revert if ucptable.h is ever
947 re-created.
948
949 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
950 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
951 order to be able to cut out the UTF-8 tables in the latter when UTF-8
952 support is not required. This saves 1.5-2K of code, which is important in
953 some applications.
954
955 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
956 so as not to refer to the tables, even though these functions will never be
957 called when UTF-8 support is disabled. Otherwise there are problems with a
958 shared library.
959
960 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
961
962 (a) It was defining its arguments as char * instead of void *.
963
964 (b) It was assuming that all moves were upwards in memory; this was true
965 a long time ago when I wrote it, but is no longer the case.
966
967 The emulated memove() is provided for those environments that have neither
968 memmove() nor bcopy(). I didn't think anyone used it these days, but that
969 is clearly not the case, as these two bugs were recently reported.
970
971 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
972 and Detrail to create the HTML documentation, the .txt form of the man
973 pages, and it removes trailing spaces from listed files. It also creates
974 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
975 case, it wraps all the #defines with #ifndefs. This script should be run
976 before "make dist".
977
978 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
979 with Unicode property support.
980
981 (a) For a maximizing quantifier, if the two different cases of the
982 character were of different lengths in their UTF-8 codings (there are
983 some cases like this - I found 11), and the matching function had to
984 back up over a mixture of the two cases, it incorrectly assumed they
985 were both the same length.
986
987 (b) When PCRE was configured to use the heap rather than the stack for
988 recursion during matching, it was not correctly preserving the data for
989 the other case of a UTF-8 character when checking ahead for a match
990 while processing a minimizing repeat. If the check also involved
991 matching a wide character, but failed, corruption could cause an
992 erroneous result when trying to check for a repeat of the original
993 character.
994
995 13. Some tidying changes to the testing mechanism:
996
997 (a) The RunTest script now detects the internal link size and whether there
998 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
999 values substituted by "configure". (The RunGrepTest script already did
1000 this for UTF-8.) The configure.ac script no longer substitutes the
1001 relevant variables.
1002
1003 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1004 with length and offset values. This means that the output is different
1005 for different internal link sizes. Test 2 is skipped for link sizes
1006 other than 2 because of this, bypassing the problem. Unfortunately,
1007 there was also a test in test 3 (the locale tests) that used /B and
1008 failed for link sizes other than 2. Rather than cut the whole test out,
1009 I have added a new /Z option to pcretest that replaces the length and
1010 offset values with spaces. This is now used to make test 3 independent
1011 of link size. (Test 2 will be tidied up later.)
1012
1013 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1014 segmentation fault instead of returning the appropriate error message.
1015
1016 15. In multiline mode when the newline sequence was set to "any", the pattern
1017 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1018 This doesn't seem right; it now treats the CRLF combination as the line
1019 ending, and so does not match in that case. It's only a pattern such as ^$
1020 that would hit this one: something like ^ABC$ would have failed after \r
1021 and then tried again after \r\n.
1022
1023 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1024 in an attempt to make files that differ only in their line terminators
1025 compare equal. This works on Linux.
1026
1027 17. Under certain error circumstances pcregrep might try to free random memory
1028 as it exited. This is now fixed, thanks to valgrind.
1029
1030 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1031 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1032 was because its rules for how to advance for /g after matching an empty
1033 string at the end of a line did not allow for this case. They now check for
1034 it specially.
1035
1036 20. pcretest is supposed to handle patterns and data of any length, by
1037 extending its buffers when necessary. It was getting this wrong when the
1038 buffer for a data line had to be extended.
1039
1040 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1041 CRLF as a newline sequence.
1042
1043 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1044 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1045 I have nevertheless tidied it up.
1046
1047 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1048
1049 24. Added a man page for pcre-config.
1050
1051
1052 Version 7.0 19-Dec-06
1053 ---------------------
1054
1055 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1056 moving to gcc 4.1.1.
1057
1058 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1059 sys/time.h, which is documented as needed for this function. It doesn't
1060 seem to matter on Linux, but it showed up on some releases of OS X.
1061
1062 3. It seems that there are systems where bytes whose values are greater than
1063 127 match isprint() in the "C" locale. The "C" locale should be the
1064 default when a C program starts up. In most systems, only ASCII printing
1065 characters match isprint(). This difference caused the output from pcretest
1066 to vary, making some of the tests fail. I have changed pcretest so that:
1067
1068 (a) When it is outputting text in the compiled version of a pattern, bytes
1069 other than 32-126 are always shown as hex escapes.
1070
1071 (b) When it is outputting text that is a matched part of a subject string,
1072 it does the same, unless a different locale has been set for the match
1073 (using the /L modifier). In this case, it uses isprint() to decide.
1074
1075 4. Fixed a major bug that caused incorrect computation of the amount of memory
1076 required for a compiled pattern when options that changed within the
1077 pattern affected the logic of the preliminary scan that determines the
1078 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1079 that the computed length was too small. The symptoms of this bug were
1080 either the PCRE error "internal error: code overflow" from pcre_compile(),
1081 or a glibc crash with a message such as "pcretest: free(): invalid next
1082 size (fast)". Examples of patterns that provoked this bug (shown in
1083 pcretest format) are:
1084
1085 /(?-x: )/x
1086 /(?x)(?-x: \s*#\s*)/
1087 /((?i)[\x{c0}])/8
1088 /(?i:[\x{c0}])/8
1089
1090 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1091 is now done differently.
1092
1093 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1094 wrapper classes; (b) implement a new function in the C++ scanner that is
1095 more efficient than the old way of doing things because it avoids levels of
1096 recursion in the regex matching; (c) add a paragraph to the documentation
1097 for the FullMatch() function.
1098
1099 6. The escape sequence \n was being treated as whatever was defined as
1100 "newline". Not only was this contrary to the documentation, which states
1101 that \n is character 10 (hex 0A), but it also went horribly wrong when
1102 "newline" was defined as CRLF. This has been fixed.
1103
1104 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1105 was being set to -1 for the "end of line" case (supposedly a value that no
1106 character can have). Though this value is never used (the check for end of
1107 line is "zero bytes in current character"), it caused compiler complaints.
1108 I've changed it to 0xffffffff.
1109
1110 8. In pcre_version.c, the version string was being built by a sequence of
1111 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1112 string (as it is for production releases) called a macro with an empty
1113 argument. The C standard says the result of this is undefined. The gcc
1114 compiler treats it as an empty string (which was what was wanted) but it is
1115 reported that Visual C gives an error. The source has been hacked around to
1116 avoid this problem.
1117
1118 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1119 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1120 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1121 of them did).
1122
1123 10. Originally, pcretest opened its input and output without "b"; then I was
1124 told that "b" was needed in some environments, so it was added for release
1125 5.0 to both the input and output. (It makes no difference on Unix-like
1126 systems.) Later I was told that it is wrong for the input on Windows. I've
1127 now abstracted the modes into two macros, to make it easier to fiddle with
1128 them, and removed "b" from the input mode under Windows.
1129
1130 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1131
1132 12. Added -help and --help to pcretest as an official way of being reminded
1133 of the options.
1134
1135 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1136 and pcrecpp.cc because they annoy compilers at high warning levels.
1137
1138 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1139
1140 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1141 scripts are not C programs :-) and which was not noticed because it works
1142 on Linux.
1143
1144 16. pcretest is supposed to handle any length of pattern and data line (as one
1145 line or as a continued sequence of lines) by extending its input buffer if
1146 necessary. This feature was broken for very long pattern lines, leading to
1147 a string of junk being passed to pcre_compile() if the pattern was longer
1148 than about 50K.
1149
1150 17. I have done a major re-factoring of the way pcre_compile() computes the
1151 amount of memory needed for a compiled pattern. Previously, there was code
1152 that made a preliminary scan of the pattern in order to do this. That was
1153 OK when PCRE was new, but as the facilities have expanded, it has become
1154 harder and harder to keep it in step with the real compile phase, and there
1155 have been a number of bugs (see for example, 4 above). I have now found a
1156 cunning way of running the real compile function in a "fake" mode that
1157 enables it to compute how much memory it would need, while actually only
1158 ever using a few hundred bytes of working memory and without too many
1159 tests of the mode. This should make future maintenance and development
1160 easier. A side effect of this work is that the limit of 200 on the nesting
1161 depth of parentheses has been removed (though this was never a serious
1162 limitation, I suspect). However, there is a downside: pcre_compile() now
1163 runs more slowly than before (30% or more, depending on the pattern). I
1164 hope this isn't a big issue. There is no effect on runtime performance.
1165
1166 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1167 newline (only possible for the last line of a file) and it was a
1168 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1169
1170 19. Added additional timing features to pcretest. (1) The -tm option now times
1171 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1172 separate command line item, by a number that specifies the number of
1173 repeats to use when timing. The default is 50000; this gives better
1174 precision, but takes uncomfortably long for very large patterns.
1175
1176 20. Extended pcre_study() to be more clever in cases where a branch of a
1177 subpattern has no definite first character. For example, (a*|b*)[cd] would
1178 previously give no result from pcre_study(). Now it recognizes that the
1179 first character must be a, b, c, or d.
1180
1181 21. There was an incorrect error "recursive call could loop indefinitely" if
1182 a subpattern (or the entire pattern) that was being tested for matching an
1183 empty string contained only one non-empty item after a nested subpattern.
1184 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1185 incorrectly, because the \d was being skipped in the check.
1186
1187 22. The pcretest program now has a new pattern option /B and a command line
1188 option -b, which is equivalent to adding /B to every pattern. This causes
1189 it to show the compiled bytecode, without the additional information that
1190 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1191 is the same as /B/I).
1192
1193 23. A new optimization is now able automatically to treat some sequences such
1194 as a*b as a*+b. More specifically, if something simple (such as a character
1195 or a simple class like \d) has an unlimited quantifier, and is followed by
1196 something that cannot possibly match the quantified thing, the quantifier
1197 is automatically "possessified".
1198
1199 24. A recursive reference to a subpattern whose number was greater than 39
1200 went wrong under certain circumstances in UTF-8 mode. This bug could also
1201 have affected the operation of pcre_study().
1202
1203 25. Realized that a little bit of performance could be had by replacing
1204 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1205
1206 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1207
1208 27. Possessive quantifiers such as a++ were previously implemented by turning
1209 them into atomic groups such as ($>a+). Now they have their own opcodes,
1210 which improves performance. This includes the automatically created ones
1211 from 23 above.
1212
1213 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1214 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1215 the first matched character to be a colon. This applied both to named and
1216 numbered groups.
1217
1218 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1219
1220 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1221 building PCRE on Windows easier, so I have included it in the distribution.
1222
1223 31. There is now a check in pcretest against a ridiculously large number being
1224 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1225 loop, the loop is abandoned.
1226
1227 32. Forward references to subpatterns in conditions such as (?(2)...) where
1228 subpattern 2 is defined later cause pcre_compile() to search forwards in
1229 the pattern for the relevant set of parentheses. This search went wrong
1230 when there were unescaped parentheses in a character class, parentheses
1231 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1232
1233 33. "Subroutine" calls and backreferences were previously restricted to
1234 referencing subpatterns earlier in the regex. This restriction has now
1235 been removed.
1236
1237 34. Added a number of extra features that are going to be in Perl 5.10. On the
1238 whole, these are just syntactic alternatives for features that PCRE had
1239 previously implemented using the Python syntax or my own invention. The
1240 other formats are all retained for compatibility.
1241
1242 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1243 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1244 also .NET compatible.
1245
1246 (b) A recursion or subroutine call to a named group can now be defined as
1247 (?&name) as well as (?P>name).
1248
1249 (c) A backreference to a named group can now be defined as \k<name> or
1250 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1251 5.10, are also .NET compatible.
1252
1253 (d) A conditional reference to a named group can now use the syntax
1254 (?(<name>) or (?('name') as well as (?(name).
1255
1256 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1257 groups (named and numbered) that are never evaluated inline, but can be
1258 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1259 is always false. There may be only one alternative in such a group.
1260
1261 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1262 as the simple (?(R). The condition is true only if the most recent
1263 recursion is that of the given number or name. It does not search out
1264 through the entire recursion stack.
1265
1266 (g) The escape \gN or \g{N} has been added, where N is a positive or
1267 negative number, specifying an absolute or relative reference.
1268
1269 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1270 some "unreachable code" warnings.
1271
1272 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1273 things, this adds five new scripts.
1274
1275 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1276 There were also incompatibilities regarding the handling of \Q..\E inside
1277 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1278 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1279
1280 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1281 matches an empty string, and forcibly breaks the loop. There were bugs in
1282 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1283 against aaaa the result was just "a" rather than "aaaa", for example. Two
1284 separate and independent bugs (that affected different cases) have been
1285 fixed.
1286
1287 39. Refactored the code to abolish the use of different opcodes for small
1288 capturing bracket numbers. This is a tidy that I avoided doing when I
1289 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1290 The new approach is not only tidier, it makes it possible to reduce the
1291 memory needed to fix the previous bug (38).
1292
1293 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1294 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1295 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1296 mode.
1297
1298 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1299 report.
1300
1301 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1302 copy construction and assignment in the C++ wrapper.
1303
1304 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1305 couple of bugs that could have given wrong results in the "--newline=crlf"
1306 case.
1307
1308 44. Added a number of casts and did some reorganization of signed/unsigned int
1309 variables following suggestions from Dair Grant. Also renamed the variable
1310 "this" as "item" because it is a C++ keyword.
1311
1312 45. Arranged for dftables to add
1313
1314 #include "pcre_internal.h"
1315
1316 to pcre_chartables.c because without it, gcc 4.x may remove the array
1317 definition from the final binary if PCRE is built into a static library and
1318 dead code stripping is activated.
1319
1320 46. For an unanchored pattern, if a match attempt fails at the start of a
1321 newline sequence, and the newline setting is CRLF or ANY, and the next two
1322 characters are CRLF, advance by two characters instead of one.
1323
1324
1325 Version 6.7 04-Jul-06
1326 ---------------------
1327
1328 1. In order to handle tests when input lines are enormously long, pcretest has
1329 been re-factored so that it automatically extends its buffers when
1330 necessary. The code is crude, but this _is_ just a test program. The
1331 default size has been increased from 32K to 50K.
1332
1333 2. The code in pcre_study() was using the value of the re argument before
1334 testing it for NULL. (Of course, in any sensible call of the function, it
1335 won't be NULL.)
1336
1337 3. The memmove() emulation function in pcre_internal.h, which is used on
1338 systems that lack both memmove() and bcopy() - that is, hardly ever -
1339 was missing a "static" storage class specifier.
1340
1341 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1342 containing an extended class (one that cannot be represented by a bitmap
1343 because it contains high-valued characters or Unicode property items, e.g.
1344 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1345 pattern, but PCRE should not loop if you do not (it no longer does).
1346 [Detail: two cases were found: (a) a repeated subpattern containing an
1347 extended class; (b) a recursive reference to a subpattern that followed a
1348 previous extended class. It wasn't skipping over the extended class
1349 correctly when UTF-8 mode was not set.]
1350
1351 5. A negated single-character class was not being recognized as fixed-length
1352 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1353 compile error "lookbehind assertion is not fixed length".
1354
1355 6. The RunPerlTest auxiliary script was showing an unexpected difference
1356 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1357 write a Perl script that can interpret lines of an input file either as
1358 byte characters or as UTF-8, which is what "perltest" was being required to
1359 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1360 can't do is switch easily at run time between having the "use utf8;" pragma
1361 or not. In the end, I fudged it by using the RunPerlTest script to insert
1362 "use utf8;" explicitly for the UTF-8 tests.
1363
1364 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1365 the end of the subject string, contrary to the documentation and to what
1366 Perl does. This was true of both matching functions. Now it matches only at
1367 the start of the subject and immediately after *internal* newlines.
1368
1369 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1370 a pointer to an int instead of a pointer to an unsigned long int. This
1371 caused problems on 64-bit systems.
1372
1373 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1374 instance of the 'standard' template library not being so standard".
1375
1376 10. There was no check on the number of named subpatterns nor the maximum
1377 length of a subpattern name. The product of these values is used to compute
1378 the size of the memory block for a compiled pattern. By supplying a very
1379 long subpattern name and a large number of named subpatterns, the size
1380 computation could be caused to overflow. This is now prevented by limiting
1381 the length of names to 32 characters, and the number of named subpatterns
1382 to 10,000.
1383
1384 11. Subpatterns that are repeated with specific counts have to be replicated in
1385 the compiled pattern. The size of memory for this was computed from the
1386 length of the subpattern and the repeat count. The latter is limited to
1387 65535, but there was no limit on the former, meaning that integer overflow
1388 could in principle occur. The compiled length of a repeated subpattern is
1389 now limited to 30,000 bytes in order to prevent this.
1390
1391 12. Added the optional facility to have named substrings with the same name.
1392
1393 13. Added the ability to use a named substring as a condition, using the
1394 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1395 are numbers (not recommended). Forward references are permitted.
1396
1397 14. Added forward references in named backreferences (if you see what I mean).
1398
1399 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1400 pattern could run off the end of the subject. For example, the pattern
1401 "(?s)(.{1,5})"8 did this with the subject "ab".
1402
1403 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1404 PCRE_CASELESS was set when matching characters that were quantified with ?
1405 or *.
1406
1407 17. A character class other than a single negated character that had a minimum
1408 but no maximum quantifier - for example [ab]{6,} - was not handled
1409 correctly by pce_dfa_exec(). It would match only one character.
1410
1411 18. A valid (though odd) pattern that looked like a POSIX character
1412 class but used an invalid character after [ (for example [[,abc,]]) caused
1413 pcre_compile() to give the error "Failed: internal error: code overflow" or
1414 in some cases to crash with a glibc free() error. This could even happen if
1415 the pattern terminated after [[ but there just happened to be a sequence of
1416 letters, a binary zero, and a closing ] in the memory that followed.
1417
1418 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1419 over the years. Originally (before any Unicode support), just the bottom 8
1420 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1421 output from "man perlunicode" includes this:
1422
1423 The regular expression compiler produces polymorphic opcodes. That
1424 is, the pattern adapts to the data and automatically switches to
1425 the Unicode character scheme when presented with Unicode data--or
1426 instead uses a traditional byte scheme when presented with byte
1427 data.
1428
1429 Sadly, a wide octal escape does not cause a switch, and in a string with
1430 no other multibyte characters, these octal escapes are treated as before.
1431 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1432 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1433 Unicode string.
1434
1435 I have not perpetrated such confusion in PCRE. Up till now, it took just
1436 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1437 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1438 translate to the appropriate multibyte character.
1439
1440 29. Applied some refactoring to reduce the number of warnings from Microsoft
1441 and Borland compilers. This has included removing the fudge introduced
1442 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1443 a warning about an unused variable.
1444
1445 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1446 characters since release 4.0, because Perl (from release 5.004) does not.
1447 [Or at least, is documented not to: some releases seem to be in conflict
1448 with the documentation.] However, when a pattern was studied with
1449 pcre_study() and all its branches started with \s, PCRE still included VT
1450 as a possible starting character. Of course, this did no harm; it just
1451 caused an unnecessary match attempt.
1452
1453 22. Removed a now-redundant internal flag bit that recorded the fact that case
1454 dependency changed within the pattern. This was once needed for "required
1455 byte" processing, but is no longer used. This recovers a now-scarce options
1456 bit. Also moved the least significant internal flag bit to the most-
1457 significant bit of the word, which was not previously used (hangover from
1458 the days when it was an int rather than a uint) to free up another bit for
1459 the future.
1460
1461 23. Added support for CRLF line endings as well as CR and LF. As well as the
1462 default being selectable at build time, it can now be changed at runtime
1463 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1464 specify that it is scanning data with non-default line endings.
1465
1466 24. Changed the definition of CXXLINK to make it agree with the definition of
1467 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1468
1469 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1470 recursions. This makes a big different to stack usage for some patterns.
1471
1472 26. If a subpattern containing a named recursion or subroutine reference such
1473 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1474 the space required for the compiled pattern went wrong and gave too small a
1475 value. Depending on the environment, this could lead to "Failed: internal
1476 error: code overflow at offset 49" or "glibc detected double free or
1477 corruption" errors.
1478
1479 27. Applied patches from Google (a) to support the new newline modes and (b) to
1480 advance over multibyte UTF-8 characters in GlobalReplace.
1481
1482 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1483 difference for some implementation of PCRE in some Windows version.
1484
1485 29. Added some extra testing facilities to pcretest:
1486
1487 \q<number> in a data line sets the "match limit" value
1488 \Q<number> in a data line sets the "match recursion limt" value
1489 -S <number> sets the stack size, where <number> is in megabytes
1490
1491 The -S option isn't available for Windows.
1492
1493
1494 Version 6.6 06-Feb-06
1495 ---------------------
1496
1497 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1498 in pcreposix.h. I have copied the definition from pcre.h.
1499
1500 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1501 because pcre.h is no longer a built file.
1502
1503 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1504 not normally included in the compiled code.
1505
1506
1507 Version 6.5 01-Feb-06
1508 ---------------------
1509
1510 1. When using the partial match feature with pcre_dfa_exec(), it was not
1511 anchoring the second and subsequent partial matches at the new starting
1512 point. This could lead to incorrect results. For example, with the pattern
1513 /1234/, partially matching against "123" and then "a4" gave a match.
1514
1515 2. Changes to pcregrep:
1516
1517 (a) All non-match returns from pcre_exec() were being treated as failures
1518 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1519 error message is output. Some extra information is given for the
1520 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1521 probably the only errors that are likely to be caused by users (by
1522 specifying a regex that has nested indefinite repeats, for instance).
1523 If there are more than 20 of these errors, pcregrep is abandoned.
1524
1525 (b) A binary zero was treated as data while matching, but terminated the
1526 output line if it was written out. This has been fixed: binary zeroes
1527 are now no different to any other data bytes.
1528
1529 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1530 used to set a locale for matching. The --locale=xxxx long option has
1531 been added (no short equivalent) to specify a locale explicitly on the
1532 pcregrep command, overriding the environment variables.
1533
1534 (d) When -B was used with -n, some line numbers in the output were one less
1535 than they should have been.
1536
1537 (e) Added the -o (--only-matching) option.
1538
1539 (f) If -A or -C was used with -c (count only), some lines of context were
1540 accidentally printed for the final match.
1541
1542 (g) Added the -H (--with-filename) option.
1543
1544 (h) The combination of options -rh failed to suppress file names for files
1545 that were found from directory arguments.
1546
1547 (i) Added the -D (--devices) and -d (--directories) options.
1548
1549 (j) Added the -F (--fixed-strings) option.
1550
1551 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1552
1553 (l) Added the --colo(u)r option.
1554
1555 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1556 is not present by default.
1557
1558 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1559 items such as (?R) or (?1), when the recursion could match a number of
1560 alternatives. If it matched one of the alternatives, but subsequently,
1561 outside the recursion, there was a failure, the code tried to back up into
1562 the recursion. However, because of the way PCRE is implemented, this is not
1563 possible, and the result was an incorrect result from the match.
1564
1565 In order to prevent this happening, the specification of recursion has
1566 been changed so that all such subpatterns are automatically treated as
1567 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1568
1569 4. I had overlooked the fact that, in some locales, there are characters for
1570 which isalpha() is true but neither isupper() nor islower() are true. In
1571 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1572 and ordfeminine) are like this. This affected the treatment of \w and \W
1573 when they appeared in character classes, but not when they appeared outside
1574 a character class. The bit map for "word" characters is now created
1575 separately from the results of isalnum() instead of just taking it from the
1576 upper, lower, and digit maps. (Plus the underscore character, of course.)
1577
1578 5. The above bug also affected the handling of POSIX character classes such as
1579 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1580 permanent tables. Instead, the bit maps for such a class were previously
1581 created as the appropriate unions of the upper, lower, and digit bitmaps.
1582 Now they are created by subtraction from the [[:word:]] class, which has
1583 its own bitmap.
1584
1585 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1586 It is created by subtracting the vertical space characters (\x09, \x0a,
1587 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1588 subtraction was done in the overall bitmap for a character class, meaning
1589 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1590 be recognized. This bug has been fixed.
1591
1592 7. Patches from the folks at Google:
1593
1594 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1595 real life, but is still worth protecting against".
1596
1597 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1598 regular expressions".
1599
1600 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1601 have it.
1602
1603 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1604 "configure" and the latter not, in order to fix a problem somebody had
1605 with compiling the Arg class on HP-UX.
1606
1607 (e) Improve the error-handling of the C++ wrapper a little bit.
1608
1609 (f) New tests for checking recursion limiting.
1610
1611 8. The pcre_memmove() function, which is used only if the environment does not
1612 have a standard memmove() function (and is therefore rarely compiled),
1613 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1614 returning a result (though PCRE never actually uses the result).
1615
1616 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1617 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1618 returned instead of calling malloc() with an overflowing number that would
1619 most likely cause subsequent chaos.
1620
1621 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1622
1623 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1624 with this option is matched, the nmatch and pmatch options of regexec() are
1625 ignored.
1626
1627 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1628 provided in case anyone wants to the the POSIX interface with UTF-8
1629 strings.
1630
1631 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1632 C++ linking (needed for some HP-UX environments).
1633
1634 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1635 (unused parameter) and in the pcre_printint() function (omitted "default"
1636 switch label when the default is to do nothing).
1637
1638 15. Added some code to make it possible, when PCRE is compiled as a C++
1639 library, to replace subject pointers for pcre_exec() with a smart pointer
1640 class, thus making it possible to process discontinuous strings.
1641
1642 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1643 much the same function. They were added by different people who were trying
1644 to make PCRE easy to compile on non-Unix systems. It has been suggested
1645 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1646 for compiling on Windows systems. I have therefore replaced it with
1647 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1648 defaults to "extern" for C or "extern C" for C++, which works fine on
1649 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1650 SCOPE with something explicit in config.h. In addition:
1651
1652 (a) pcreposix.h still had just "extern" instead of either of these macros;
1653 I have replaced it with PCRE_DATA_SCOPE.
1654
1655 (b) Functions such as _pcre_xclass(), which are internal to the library,
1656 but external in the C sense, all had PCRE_EXPORT in their definitions.
1657 This is apparently wrong for the Windows case, so I have removed it.
1658 (It makes no difference on Unix-like systems.)
1659
1660 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1661 of recursive calls to match(). This is different to MATCH_LIMIT because
1662 that limits the total number of calls to match(), not all of which increase
1663 the depth of recursion. Limiting the recursion depth limits the amount of
1664 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1665 when PCRE is compiled, and changed at run time. A patch from Google adds
1666 this functionality to the C++ interface.
1667
1668 18. Changes to the handling of Unicode character properties:
1669
1670 (a) Updated the table to Unicode 4.1.0.
1671
1672 (b) Recognize characters that are not in the table as "Cn" (undefined).
1673
1674 (c) I revised the way the table is implemented to a much improved format
1675 which includes recognition of ranges. It now supports the ranges that
1676 are defined in UnicodeData.txt, and it also amalgamates other
1677 characters into ranges. This has reduced the number of entries in the
1678 table from around 16,000 to around 3,000, thus reducing its size
1679 considerably. I realized I did not need to use a tree structure after
1680 all - a binary chop search is just as efficient. Having reduced the
1681 number of entries, I extended their size from 6 bytes to 8 bytes to
1682 allow for more data.
1683
1684 (d) Added support for Unicode script names via properties such as \p{Han}.
1685
1686 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1687 matching that character.
1688
1689 20. When matching a repeated Unicode property with a minimum greater than zero,
1690 (for example \pL{2,}), PCRE could look past the end of the subject if it
1691 reached it while seeking the minimum number of characters. This could
1692 happen only if some of the characters were more than one byte long, because
1693 there is a check for at least the minimum number of bytes.
1694
1695 21. Refactored the implementation of \p and \P so as to be more general, to
1696 allow for more different types of property in future. This has changed the
1697 compiled form incompatibly. Anybody with saved compiled patterns that use
1698 \p or \P will have to recompile them.
1699
1700 22. Added "Any" and "L&" to the supported property types.
1701
1702 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1703 but give a compile time error if the value is greater than 0xff.
1704
1705 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1706 accidentally not being installed or uninstalled.
1707
1708 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1709 made were to insert the current release number. This seemed silly, because
1710 it made things harder for people building PCRE on systems that don't run
1711 "configure". I have turned pcre.h into a distributed file, no longer built
1712 by "configure", with the version identification directly included. There is
1713 no longer a pcre.h.in file.
1714
1715 However, this change necessitated a change to the pcre-config script as
1716 well. It is built from pcre-config.in, and one of the substitutions was the
1717 release number. I have updated configure.ac so that ./configure now finds
1718 the release number by grepping pcre.h.
1719
1720 26. Added the ability to run the tests under valgrind.
1721
1722
1723 Version 6.4 05-Sep-05
1724 ---------------------
1725
1726 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1727 "--" to be printed when multiple files were scanned, even when none of the
1728 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1729 consider it to be a bug, and have restored the previous behaviour.
1730
1731 2. A couple of code tidies to get rid of compiler warnings.
1732
1733 3. The pcretest program used to cheat by referring to symbols in the library
1734 whose names begin with _pcre_. These are internal symbols that are not
1735 really supposed to be visible externally, and in some environments it is
1736 possible to suppress them. The cheating is now confined to including
1737 certain files from the library's source, which is a bit cleaner.
1738
1739 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1740 file's purpose clearer.
1741
1742 5. Reorganized pcre_ucp_findchar().
1743
1744
1745 Version 6.3 15-Aug-05
1746 ---------------------
1747
1748 1. The file libpcre.pc.in did not have general read permission in the tarball.
1749
1750 2. There were some problems when building without C++ support:
1751
1752 (a) If C++ support was not built, "make install" and "make test" still
1753 tried to test it.
1754
1755 (b) There were problems when the value of CXX was explicitly set. Some
1756 changes have been made to try to fix these, and ...
1757
1758 (c) --disable-cpp can now be used to explicitly disable C++ support.
1759
1760 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1761 backslash in a target when C++ was disabled. This confuses some
1762 versions of "make", apparently. Using an intermediate variable solves
1763 this. (Same for CPP_LOBJ.)
1764
1765 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1766 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1767 necessary on certain architectures.
1768
1769 4. Added a setting of -export-symbols-regex to the link command to remove
1770 those symbols that are exported in the C sense, but actually are local
1771 within the library, and not documented. Their names all begin with
1772 "_pcre_". This is not a perfect job, because (a) we have to except some
1773 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1774 available (and never for static libraries). I have made a note to try to
1775 find a way round (a) in the future.
1776
1777
1778 Version 6.2 01-Aug-05
1779 ---------------------
1780
1781 1. There was no test for integer overflow of quantifier values. A construction
1782 such as {1111111111111111} would give undefined results. What is worse, if
1783 a minimum quantifier for a parenthesized subpattern overflowed and became
1784 negative, the calculation of the memory size went wrong. This could have
1785 led to memory overwriting.
1786
1787 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1788
1789 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1790 operating environments where this matters.
1791
1792 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1793 PCRE options from within the C++ wrapper.
1794
1795 5. Named capturing subpatterns were not being correctly counted when a pattern
1796 was compiled. This caused two problems: (a) If there were more than 100
1797 such subpatterns, the calculation of the memory needed for the whole
1798 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1799 back references of the form \12, where the number was greater than 9, were
1800 not recognized as back references, even though there were sufficient
1801 previous subpatterns.
1802
1803 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1804 versions of gcc, e.g. 2.95.4.
1805
1806
1807 Version 6.1 21-Jun-05
1808 ---------------------
1809
1810 1. There was one reference to the variable "posix" in pcretest.c that was not
1811 surrounded by "#if !defined NOPOSIX".
1812
1813 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1814 the cross-check on the old pcre_info() function, for the benefit of the
1815 cut-down version of PCRE that is currently imported into Exim.
1816
1817 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1818 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1819 patterns that start (?i) but I don't think that matters. The use of (?i) is
1820 just an example; this all applies to the other options as well.
1821
1822 4. Since libtool seems to echo the compile commands it is issuing, the output
1823 from "make" can be reduced a bit by putting "@" in front of each libtool
1824 compile command.
1825
1826 5. Patch from the folks at Google for configure.in to be a bit more thorough
1827 in checking for a suitable C++ installation before trying to compile the
1828 C++ stuff. This should fix a reported problem when a compiler was present,
1829 but no suitable headers.
1830
1831 6. The man pages all had just "PCRE" as their title. I have changed them to
1832 be the relevant file name. I have also arranged that these names are
1833 retained in the file doc/pcre.txt, which is a concatenation in text format
1834 of all the man pages except the little individual ones for each function.
1835
1836 7. The NON-UNIX-USE file had not been updated for the different set of source
1837 files that come with release 6. I also added a few comments about the C++
1838 wrapper.
1839
1840
1841 Version 6.0 07-Jun-05
1842 ---------------------
1843
1844 1. Some minor internal re-organization to help with my DFA experiments.
1845
1846 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1847 didn't matter for the library itself when fully configured, but did matter
1848 when compiling without UCP support, or within Exim, where the ucp files are
1849 not imported.
1850
1851 3. Refactoring of the library code to split up the various functions into
1852 different source modules. The addition of the new DFA matching code (see
1853 below) to a single monolithic source would have made it really too
1854 unwieldy, quite apart from causing all the code to be include in a
1855 statically linked application, when only some functions are used. This is
1856 relevant even without the DFA addition now that patterns can be compiled in
1857 one application and matched in another.
1858
1859 The downside of splitting up is that there have to be some external
1860 functions and data tables that are used internally in different modules of
1861 the library but which are not part of the API. These have all had their
1862 names changed to start with "_pcre_" so that they are unlikely to clash
1863 with other external names.
1864
1865 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1866 a different (DFA) algorithm. Although it is slower than the original
1867 function, it does have some advantages for certain types of matching
1868 problem.
1869
1870 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1871 including restarting after a partial match.
1872
1873 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1874 defined when compiling for Windows was sent to me. I have put it into the
1875 code, though I have no means of testing or verifying it.
1876
1877 7. Added the pcre_refcount() auxiliary function.
1878
1879 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1880 match before or at the first newline in the subject string. In pcretest,
1881 the /f option on a pattern can be used to set this.
1882
1883 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1884 would behave wrongly. This has been present in PCRE since release 4.0.
1885
1886 10. A number of changes to the pcregrep command:
1887
1888 (a) Refactored how -x works; insert ^(...)$ instead of setting
1889 PCRE_ANCHORED and checking the length, in preparation for adding
1890 something similar for -w.
1891
1892 (b) Added the -w (match as a word) option.
1893
1894 (c) Refactored the way lines are read and buffered so as to have more
1895 than one at a time available.
1896
1897 (d) Implemented a pcregrep test script.
1898
1899 (e) Added the -M (multiline match) option. This allows patterns to match
1900 over several lines of the subject. The buffering ensures that at least
1901 8K, or the rest of the document (whichever is the shorter) is available
1902 for matching (and similarly the previous 8K for lookbehind assertions).
1903
1904 (f) Changed the --help output so that it now says
1905
1906 -w, --word-regex(p)
1907
1908 instead of two lines, one with "regex" and the other with "regexp"
1909 because that confused at least one person since the short forms are the
1910 same. (This required a bit of code, as the output is generated
1911 automatically from a table. It wasn't just a text change.)
1912
1913 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1914 option but starts with a hyphen. Could be a pattern or a path name
1915 starting with a hyphen, for instance.
1916
1917 (h) "-" can be given as a file name to represent stdin.
1918
1919 (i) When file names are being printed, "(standard input)" is used for
1920 the standard input, for compatibility with GNU grep. Previously
1921 "<stdin>" was used.
1922
1923 (j) The option --label=xxx can be used to supply a name to be used for
1924 stdin when file names are being printed. There is no short form.
1925
1926 (k) Re-factored the options decoding logic because we are going to add
1927 two more options that take data. Such options can now be given in four
1928 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1929
1930 (l) Added the -A, -B, and -C options for requesting that lines of context
1931 around matches be printed.
1932
1933 (m) Added the -L option to print the names of files that do not contain
1934 any matching lines, that is, the complement of -l.
1935
1936 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1937 continue to scan other files.
1938
1939 (o) The -s option was incorrectly implemented. For compatibility with other
1940 greps, it now suppresses the error message for a non-existent or non-
1941 accessible file (but not the return code). There is a new option called
1942 -q that suppresses the output of matching lines, which was what -s was
1943 previously doing.
1944
1945 (p) Added --include and --exclude options to specify files for inclusion
1946 and exclusion when recursing.
1947
1948 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1949 Hopefully, it now does.
1950
1951 12. Missing cast in pcre_study().
1952
1953 13. Added an "uninstall" target to the makefile.
1954
1955 14. Replaced "extern" in the function prototypes in Makefile.in with
1956 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1957 world, but is set differently for Windows.
1958
1959 15. Added a second compiling function called pcre_compile2(). The only
1960 difference is that it has an extra argument, which is a pointer to an
1961 integer error code. When there is a compile-time failure, this is set
1962 non-zero, in addition to the error test pointer being set to point to an
1963 error message. The new argument may be NULL if no error number is required
1964 (but then you may as well call pcre_compile(), which is now just a
1965 wrapper). This facility is provided because some applications need a
1966 numeric error indication, but it has also enabled me to tidy up the way
1967 compile-time errors are handled in the POSIX wrapper.
1968
1969 16. Added VPATH=.libs to the makefile; this should help when building with one
1970 prefix path and installing with another. (Or so I'm told by someone who
1971 knows more about this stuff than I do.)
1972
1973 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1974 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1975 match everything, including newlines. This is not POSIX-compatible, but
1976 somebody wanted the feature. From pcretest it can be activated by using
1977 both the P and the s flags.
1978
1979 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1980
1981 19. libpcre.pc was being incorrectly installed as executable.
1982
1983 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1984 it now also looks for '\r' so that it will work unmodified on Windows.
1985
1986 21. Added Google's contributed C++ wrapper to the distribution.
1987
1988 22. Added some untidy missing memory free() calls in pcretest, to keep
1989 Electric Fence happy when testing.
1990
1991
1992
1993 Version 5.0 13-Sep-04
1994 ---------------------
1995
1996 1. Internal change: literal characters are no longer packed up into items
1997 containing multiple characters in a single byte-string. Each character
1998 is now matched using a separate opcode. However, there may be more than one
1999 byte in the character in UTF-8 mode.
2000
2001 2. The pcre_callout_block structure has two new fields: pattern_position and
2002 next_item_length. These contain the offset in the pattern to the next match
2003 item, and its length, respectively.
2004
2005 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2006 insertion of callouts before each pattern item. Added the /C option to
2007 pcretest to make use of this.
2008
2009 4. On the advice of a Windows user, the lines
2010
2011 #if defined(_WIN32) || defined(WIN32)
2012 _setmode( _fileno( stdout ), 0x8000 );
2013 #endif /* defined(_WIN32) || defined(WIN32) */
2014
2015 have been added to the source of pcretest. This apparently does useful
2016 magic in relation to line terminators.
2017
2018 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2019 for the benefit of those environments where the "b" makes a difference.
2020
2021 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2022 to know about it. I have put a hack into configure.in that adds in code
2023 to set GCC=yes if CC=icc. This seems to end up at a point in the
2024 generated configure script that is early enough to affect the setting of
2025 compiler options, which is what is needed, but I have no means of testing
2026 whether it really works. (The user who reported this had patched the
2027 generated configure script, which of course I cannot do.)
2028
2029 LATER: After change 22 below (new libtool files), the configure script
2030 seems to know about icc (and also ecc). Therefore, I have commented out
2031 this hack in configure.in.
2032
2033 7. Added support for pkg-config (2 patches were sent in).
2034
2035 8. Negated POSIX character classes that used a combination of internal tables
2036 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2037 [[:^ascii]]. Typically, they would match almost any characters. The other
2038 POSIX classes were not broken in this way.
2039
2040 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2041 to find the match, as PCRE was deluded into thinking that the match had to
2042 start at the start point or following a newline. The same bug applied to
2043 patterns with negative forward assertions or any backward assertions
2044 preceding ".*" at the start, unless the pattern required a fixed first
2045 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2046
2047 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2048 starting at the last subject character, bytes beyond the end of the subject
2049 string were read.
2050
2051 11. Renamed the variable "class" as "classbits" to make life easier for C++
2052 users. (Previously there was a macro definition, but it apparently wasn't
2053 enough.)
2054
2055 12. Added the new field "tables" to the extra data so that tables can be passed
2056 in at exec time, or the internal tables can be re-selected. This allows
2057 a compiled regex to be saved and re-used at a later time by a different
2058 program that might have everything at different addresses.
2059
2060 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2061 -R library as well as a -L library.
2062
2063 14. The debugging options of pcretest (-d on the command line or D on a
2064 pattern) showed incorrect output for anything following an extended class
2065 that contained multibyte characters and which was followed by a quantifier.
2066
2067 15. Added optional support for general category Unicode character properties
2068 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2069 support. It adds about 90K to the size of the library. The meanings of the
2070 inbuilt class escapes such as \d and \s have NOT been changed.
2071
2072 16. Updated pcredemo.c to include calls to free() to release the memory for the
2073 compiled pattern.
2074
2075 17. The generated file chartables.c was being created in the source directory
2076 instead of in the building directory. This caused the build to fail if the
2077 source directory was different from the building directory, and was
2078 read-only.
2079
2080 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2081 file. No doubt somebody will tell me if they don't make sense... Also added
2082 Dan Mooney's comments about building on OpenVMS.
2083
2084 19. Added support for partial matching via the PCRE_PARTIAL option for
2085 pcre_exec() and the \P data escape in pcretest.
2086
2087 20. Extended pcretest with 3 new pattern features:
2088
2089 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2090 write the compiled pattern to the file whose name is "rest-of-line".
2091 This is a straight binary dump of the data, with the saved pointer to
2092 the character tables forced to be NULL. The study data, if any, is
2093 written too. After writing, pcretest reads a new pattern.
2094
2095 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2096 compiled pattern from the given file. There must not be any
2097 occurrences of "<" in the file name (pretty unlikely); if there are,
2098 pcretest will instead treat the initial "<" as a pattern delimiter.
2099 After reading in the pattern, pcretest goes on to read data lines as
2100 usual.
2101
2102 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2103 and 16-bit fields in a compiled pattern, to simulate a pattern that
2104 was compiled on a host of opposite endianness.
2105
2106 21. The pcre-exec() function can now cope with patterns that were compiled on
2107 hosts of opposite endianness, with this restriction:
2108
2109 As for any compiled expression that is saved and used later, the tables
2110 pointer field cannot be preserved; the extra_data field in the arguments
2111 to pcre_exec() should be used to pass in a tables address if a value
2112 other than the default internal tables were used at compile time.
2113
2114 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2115 now diagnosed as an error. Previously, most of the time, a negative number
2116 would have been treated as zero, but if in addition "ovector" was passed as
2117 NULL, a crash could occur.
2118
2119 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2120 new versions from the libtool 1.5 distribution (the last one is a copy of
2121 a file called libtool.m4). This seems to have fixed the need to patch
2122 "configure" to support Darwin 1.3 (which I used to do). However, I still
2123 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2124 workstation).
2125
2126 24. Changed the PCRE licence to be the more standard "BSD" licence.
2127
2128
2129 Version 4.5 01-Dec-03
2130 ---------------------
2131
2132 1. There has been some re-arrangement of the code for the match() function so
2133 that it can be compiled in a version that does not call itself recursively.
2134 Instead, it keeps those local variables that need separate instances for
2135 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2136 needs to "recurse". Keeping track of where control must go is done by means
2137 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2138 hide most of the details from the main code, and operates only if
2139 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2140 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2141 operating.
2142
2143 To make it easier for callers to provide specially tailored get/free
2144 functions for this usage, two new functions, pcre_stack_malloc, and
2145 pcre_stack_free, are used. They are always called in strict stacking order,
2146 and the size of block requested is always the same.
2147
2148 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2149 PCRE has been compiled to use the stack or the heap for recursion. The
2150 -C option of pcretest uses this to show which version is compiled.
2151
2152 A new data escape \S, is added to pcretest; it causes the amounts of store
2153 obtained and freed by both kinds of malloc/free at match time to be added
2154 to the output.
2155
2156 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2157 what's available on my current Linux desktop machine.
2158
2159 3. When matching a UTF-8 string, the test for a valid string at the start has
2160 been extended. If start_offset is not zero, PCRE now checks that it points
2161 to a byte that is the start of a UTF-8 character. If not, it returns
2162 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2163 this is necessary because there may be backward assertions in the pattern.
2164 When matching the same subject several times, it may save resources to use
2165 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2166
2167 4. The code for checking the validity of UTF-8 strings has been tightened so
2168 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2169 containing "overlong sequences".
2170
2171 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2172 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2173 should have been "&", but it just so happened that all the cases this let
2174 through by mistake were picked up later in the function.
2175
2176 6. I had used a variable called "isblank" - this is a C99 function, causing
2177 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2178
2179 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2180 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2181 so the version is shown; (c) stop "make test" echoing "./RunTest".
2182
2183 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2184
2185 9. The prototype for memmove() for systems that don't have it was using
2186 size_t, but the inclusion of the header that defines size_t was later. I've
2187 moved the #includes for the C headers earlier to avoid this.
2188
2189 10. Added some adjustments to the code to make it easier to compiler on certain
2190 special systems:
2191
2192 (a) Some "const" qualifiers were missing.
2193 (b) Added the macro EXPORT before all exported functions; by default this
2194 is defined to be empty.
2195 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2196 that it reads its output file name as an argument instead of writing
2197 to the standard output and assuming this can be redirected.
2198
2199 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2200 class containing characters with values greater than 255, PCRE compilation
2201 went into a loop.
2202
2203 12. A recursive reference to a subpattern that was within another subpattern
2204 that had a minimum quantifier of zero caused PCRE to crash. For example,
2205 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2206 recursion. If the recursively-called subpattern itself had a zero repeat,
2207 that was OK.
2208
2209 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2210 buffer into which it was copied (for escape processing) was still set at
2211 1024, so long lines caused crashes.
2212
2213 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2214 "internal error: code overflow...". This applied to any character class
2215 that was followed by a possessive quantifier.
2216
2217 15. Modified the Makefile to add libpcre.la as a prerequisite for
2218 libpcreposix.la because I was told this is needed for a parallel build to
2219 work.
2220
2221 16. If a pattern that contained .* following optional items at the start was
2222 studied, the wrong optimizing data was generated, leading to matching
2223 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2224 matching string must start with a or b or c. The correct conclusion for
2225 this pattern is that a match can start with any character.
2226
2227
2228 Version 4.4 13-Aug-03
2229 ---------------------
2230
2231 1. In UTF-8 mode, a character class containing characters with values between
2232 127 and 255 was not handled correctly if the compiled pattern was studied.
2233 In fixing this, I have also improved the studying algorithm for such
2234 classes (slightly).
2235
2236 2. Three internal functions had redundant arguments passed to them. Removal
2237 might give a very teeny performance improvement.
2238
2239 3. Documentation bug: the value of the capture_top field in a callout is *one
2240 more than* the number of the hightest numbered captured substring.
2241
2242 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2243 in incorrectly linking with a previously installed version. They now link
2244 explicitly with libpcre.la.
2245
2246 5. configure.in no longer needs to recognize Cygwin specially.
2247
2248 6. A problem in pcre.in for Windows platforms is fixed.
2249
2250 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2251 pcretest, it used to include the size of the study block as part of its
2252 output. Unfortunately, the structure contains a field that has a different
2253 size on different hardware architectures. This meant that the tests that
2254 showed this size failed. As the block is currently always of a fixed size,
2255 this information isn't actually particularly useful in pcretest output, so
2256 I have just removed it.
2257
2258 8. Three pre-processor statements accidentally did not start in column 1.
2259 Sadly, there are *still* compilers around that complain, even though
2260 standard C has not required this for well over a decade. Sigh.
2261
2262 9. In pcretest, the code for checking callouts passed small integers in the
2263 callout_data field, which is a void * field. However, some picky compilers
2264 complained about the casts involved for this on 64-bit systems. Now
2265 pcretest passes the address of the small integer instead, which should get
2266 rid of the warnings.
2267
2268 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2269 both compile and run time, and gives an error if an invalid UTF-8 sequence
2270 is found. There is a option for disabling this check in cases where the
2271 string is known to be correct and/or the maximum performance is wanted.
2272
2273 11. In response to a bug report, I changed one line in Makefile.in from
2274
2275 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2276 to
2277 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2278
2279 to look similar to other lines, but I have no way of telling whether this
2280 is the right thing to do, as I do not use Windows. No doubt I'll get told
2281 if it's wrong...
2282
2283
2284 Version 4.3 21-May-03
2285 ---------------------
2286
2287 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2288 Makefile.
2289
2290 2. Some refactoring to improve the quality of the code:
2291
2292 (i) The utf8_table... variables are now declared "const".
2293
2294 (ii) The code for \cx, which used the "case flipping" table to upper case
2295 lower case letters, now just substracts 32. This is ASCII-specific,
2296 but the whole concept of \cx is ASCII-specific, so it seems
2297 reasonable.
2298
2299 (iii) PCRE was using its character types table to recognize decimal and
2300 hexadecimal digits in the pattern. This is silly, because it handles
2301 only 0-9, a-f, and A-F, but the character types table is locale-
2302 specific, which means strange things might happen. A private
2303 table is now used for this - though it costs 256 bytes, a table is
2304 much faster than multiple explicit tests. Of course, the standard
2305 character types table is still used for matching digits in subject
2306 strings against \d.
2307
2308 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2309 ending in _t are). So I've renamed it as ESC_tee.
2310
2311 3. The first argument for regexec() in the POSIX wrapper should have been
2312 defined as "const".
2313
2314 4. Changed pcretest to use malloc() for its buffers so that they can be
2315 Electric Fenced for debugging.
2316
2317 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2318 to read one or more bytes before the start of the subject string. Often this
2319 had no effect on PCRE's behaviour, but in some circumstances it could
2320 provoke a segmentation fault.
2321
2322 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2323 to try to read one or more bytes before the start of the subject string.
2324
2325 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2326 UTF-8 support could misbehave in various ways if the subject string
2327 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2328 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2329 back over UTF-8 characters.)
2330
2331
2332 Version 4.2 14-Apr-03
2333 ---------------------
2334
2335 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2336
2337 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2338 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2339 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2340 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2341 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2342 and BUILD_EXEEXT
2343 Note: automatic setting of the BUILD variables is not yet working
2344 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2345 compile-time but not at link-time
2346 [LINK]: use for linking executables only
2347 make different versions for Windows and non-Windows
2348 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2349 libraries
2350 [LINK_FOR_BUILD]: new variable
2351 [OBJEXT]: use throughout
2352 [EXEEXT]: use throughout
2353 <winshared>: new target
2354 <wininstall>: new target
2355 <dftables.o>: use native compiler
2356 <dftables>: use native linker
2357 <install>: handle Windows platform correctly
2358 <clean>: ditto
2359 <check>: ditto
2360 copy DLL to top builddir before testing
2361
2362 As part of these changes, -no-undefined was removed again. This was reported
2363 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2364 in any case.
2365
2366 3. Some tidies to get rid of compiler warnings:
2367
2368 . In the match_data structure, match_limit was an unsigned long int, whereas
2369 match_call_count was an int. I've made them both unsigned long ints.
2370
2371 . In pcretest the fact that a const uschar * doesn't automatically cast to
2372 a void * provoked a warning.
2373
2374 . Turning on some more compiler warnings threw up some "shadow" variables
2375 and a few more missing casts.
2376
2377 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2378 option, a class that contained a single character with a value between 128
2379 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2380
2381 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2382 option, a class that contained several characters, but with at least one
2383 whose value was between 128 and 255 caused PCRE to crash.
2384
2385
2386 Version 4.1 12-Mar-03
2387 ---------------------
2388
2389 1. Compiling with gcc -pedantic found a couple of places where casts were
2390 needed, and a string in dftables.c that was longer than standard compilers are
2391 required to support.
2392
2393 2. Compiling with Sun's compiler found a few more places where the code could
2394 be tidied up in order to avoid warnings.
2395
2396 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2397 first of these names is deprecated in the latest Autoconf in favour of the name
2398 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2399 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2400 analogy I have changed it to CFLAGS_FOR_BUILD.
2401
2402 4. Added -no-undefined to the linking command in the Makefile, because this is
2403 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2404 linking step for the pcreposix library.
2405
2406 5. PCRE was failing to diagnose the case of two named groups with the same
2407 name.
2408
2409 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2410 literal character that is needed in the subject for a match, and scans along to
2411 ensure that it is present before embarking on the full matching process. This
2412 saves time in cases of nested unlimited repeats that are never going to match.
2413 Problem: the scan can take a lot of time if the subject is very long (e.g.
2414 megabytes), thus penalizing straightforward matches. It is now done only if the
2415 amount of subject to be scanned is less than 1000 bytes.
2416
2417 7. A lesser problem with the same optimization is that it was recording the
2418 first character of an anchored pattern as "needed", thus provoking a search
2419 right along the subject, even when the first match of the pattern was going to
2420 fail. The "needed" character is now not set for anchored patterns, unless it
2421 follows something in the pattern that is of non-fixed length. Thus, it still
2422 fulfils its original purpose of finding quick non-matches in cases of nested
2423 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2424
2425
2426 Version 4.0 17-Feb-03
2427 ---------------------
2428
2429 1. If a comment in an extended regex that started immediately after a meta-item
2430 extended to the end of string, PCRE compiled incorrect data. This could lead to
2431 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2432
2433 2. Moved to autoconf 2.53 and libtool 1.4.2.
2434
2435 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2436 the special perltest8 script is no longer needed - all the tests can be run
2437 from a single perltest script.
2438
2439 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2440 by \s. It has now been removed in PCRE. This means it isn't recognized as
2441 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2442 class [:space:] *does* include VT, thereby creating a mess.
2443
2444 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2445 space and tab.
2446
2447 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2448 its new features into the main test script, reducing the number of scripts.
2449
2450 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2451 were backward compatible, and made the (?i) apply to the whole pattern, as if
2452 /i were given. Now it behaves more logically, and applies the option setting
2453 only to what follows. PCRE has been changed to follow suit. However, if it
2454 finds options settings right at the start of the pattern, it extracts them into
2455 the global options, as before. Thus, they show up in the info data.
2456
2457 8. Added support for the \Q...\E escape sequence. Characters in between are
2458 treated as literals. This is slightly different from Perl in that $ and @ are
2459 also handled as literals inside the quotes. In Perl, they will cause variable
2460 interpolation. Note the following examples:
2461
2462 Pattern PCRE matches Perl matches
2463
2464 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2465 \Qabc\$xyz\E abc\$xyz abc\$xyz
2466 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2467
2468 For compatibility with Perl, \Q...\E sequences are recognized inside character
2469 classes as well as outside them.
2470
2471 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2472 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2473 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2474 signed/unsigned warnings.
2475
2476 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2477 option for pcretest, so I've replaced it by a simple function that does just
2478 that job.
2479
2480 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2481 "pcregrep -".
2482
2483 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2484 Java package. This provides some syntactic sugar for simple cases of what my
2485 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2486 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2487 item, you can use this simplified notation. Note that only makes sense with
2488 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2489 greediness, whatever the setting of the PCRE_UNGREEDY option.
2490
2491 13. A change of greediness default within a pattern was not taking effect at
2492 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2493 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2494 was abstracted outside.
2495
2496 14. PCRE now supports the \G assertion. It is true when the current matching
2497 position is at the start point of the match. This differs from \A when the
2498 starting offset is non-zero. Used with the /g option of pcretest (or similar
2499 code), it works in the same way as it does for Perl's /g option. If all
2500 alternatives of a regex begin with \G, the expression is anchored to the start
2501 match position, and the "anchored" flag is set in the compiled expression.
2502
2503 15. Some bugs concerning the handling of certain option changes within patterns
2504 have been fixed. These applied to options other than (?ims). For example,
2505 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2506 been the other way round. Some of this was related to change 7 above.
2507
2508 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2509 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2510 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2511 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2512
2513 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2514 mode. Unlike ".", it always matches newline, whatever the setting of
2515 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2516 assertions. Perl allows it, but it doesn't (in general) work because it can't
2517 calculate the length of the lookbehind. At least, that's the case for Perl
2518 5.8.0 - I've been told they are going to document that it doesn't work in
2519 future.
2520
2521 18. Added an error diagnosis for escapes that PCRE does not support: these are
2522 \L, \l, \N, \P, \p, \U, \u, and \X.
2523
2524 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2525 reading past the end of the pattern in cases such as /[abcd/.
2526
2527 20. PCRE was getting more memory than necessary for patterns with classes that
2528 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2529
2530 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2531 compiling PCRE for use with Virtual Pascal.
2532
2533 22. Small fix to the Makefile to make it work properly if the build is done
2534 outside the source tree.
2535
2536 23. Added a new extension: a condition to go with recursion. If a conditional
2537 subpattern starts with (?(R) the "true" branch is used if recursion has
2538 happened, whereas the "false" branch is used only at the top level.
2539
2540 24. When there was a very long string of literal characters (over 255 bytes
2541 without UTF support, over 250 bytes with UTF support), the computation of how
2542 much memory was required could be incorrect, leading to segfaults or other
2543 strange effects.
2544
2545 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2546 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2547 there was a subsequent back reference to those brackets. This meant that, for
2548 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2549 possible to check for precisely this case. All we can do is abandon the
2550 optimization if .* occurs inside capturing brackets when there are any back
2551 references whatsoever. (See below for a better fix that came later.)
2552
2553 26. The handling of the optimization for finding the first character of a
2554 non-anchored pattern, and for finding a character that is required later in the
2555 match were failing in some cases. This didn't break the matching; it just
2556 failed to optimize when it could. The way this is done has been re-implemented.
2557
2558 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2559
2560 28. Added a new feature that provides some of the functionality that Perl
2561 provides with (?{...}). The facility is termed a "callout". The way it is done
2562 in PCRE is for the caller to provide an optional function, by setting
2563 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2564 global variable. By default it is unset, which disables all calling out. To get
2565 the function called, the regex must include (?C) at appropriate points. This
2566 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2567 This provides a means of identifying different callout points. When PCRE
2568 reaches such a point in the regex, if pcre_callout has been set, the external
2569 function is called. It is provided with data in a structure called
2570 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2571 matching continues; if it returns a non-zero value, the match at the current
2572 point fails. However, backtracking will occur if possible. [This was changed
2573 later and other features added - see item 49 below.]
2574
2575 29. pcretest is upgraded to test the callout functionality. It provides a
2576 callout function that displays information. By default, it shows the start of
2577 the match and the current position in the text. There are some new data escapes
2578 to vary what happens:
2579
2580 \C+ in addition, show current contents of captured substrings
2581 \C- do not supply a callout function
2582 \C!n return 1 when callout number n is reached
2583 \C!n!m return 1 when callout number n is reached for the mth time
2584
2585 30. If pcregrep was called with the -l option and just a single file name, it
2586 output "<stdin>" if a match was found, instead of the file name.
2587
2588 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2589 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2590 pcre_exec(). This saves a malloc/free per call. The default value of
2591 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2592 when configuring.
2593
2594 32. The default maximum size of a compiled pattern is 64K. There have been a
2595 few cases of people hitting this limit. The code now uses macros to handle the
2596 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2597 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2598 configuring. Tests 2 and 5 work only with 2-byte links because they output
2599 debugging information about compiled patterns.
2600
2601 33. Internal code re-arrangements:
2602
2603 (a) Moved the debugging function for printing out a compiled regex into
2604 its own source file (printint.c) and used #include to pull it into
2605 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2606 separate copies.
2607
2608 (b) Defined the list of op-code names for debugging as a macro in
2609 internal.h so that it is next to the definition of the opcodes.
2610
2611 (c) Defined a table of op-code lengths for simpler skipping along compiled
2612 code. This is again a macro in internal.h so that it is next to the
2613 definition of the opcodes.
2614
2615 34. Added support for recursive calls to individual subpatterns, along the
2616 lines of Robin Houston's patch (but implemented somewhat differently).
2617
2618 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2619 allow it to read and process whole directories in Win32. This code was
2620 contributed by Lionel Fourquaux; it has not been tested by me.
2621
2622 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2623 used to name a group. Names consist of alphanumerics and underscores, and must
2624 be unique. Back references use the syntax (?P=name) and recursive calls use
2625 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2626 numbers. The function pcre_fullinfo() can be used after compilation to extract
2627 a name/number map. There are three relevant calls:
2628
2629 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2630 PCRE_INFO_NAMECOUNT yields the number of entries
2631 PCRE_INFO_NAMETABLE yields a pointer to the map.
2632
2633 The map is a vector of fixed-size entries. The size of each entry depends on
2634 the length of the longest name used. The first two bytes of each entry are the
2635 group number, most significant byte first. There follows the corresponding
2636 name, zero terminated. The names are in alphabetical order.
2637
2638 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2639 case instead of 255. Making it the same both with and without UTF-8 support
2640 means that the same test output works with both.
2641
2642 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2643 calling malloc() with a zero argument.
2644
2645 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2646 optimization. I've improved things by keeping a bitmap of backreferences with
2647 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2648 fact referenced, the optimization can be applied. It is unlikely that a
2649 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2650 the match to follow \n) will appear inside brackets with a number greater than
2651 31, but if it does, any back reference > 31 suppresses the optimization.
2652
2653 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2654 of disabling numbered capturing parentheses. Any opening parenthesis that is
2655 not followed by ? behaves as if it were followed by ?: but named parentheses
2656 can still be used for capturing (and they will acquire numbers in the usual
2657 way).
2658
2659 41. Redesigned the return codes from the match() function into yes/no/error so
2660 that errors can be passed back from deep inside the nested calls. A malloc
2661 failure while inside a recursive subpattern call now causes the
2662 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2663
2664 42. It is now possible to set a limit on the number of times the match()
2665 function is called in a call to pcre_exec(). This facility makes it possible to
2666 limit the amount of recursion and backtracking, though not in a directly
2667 obvious way, because the match() function is used in a number of different
2668 circumstances. The count starts from zero for each position in the subject
2669 string (for non-anchored patterns). The default limit is, for compatibility, a
2670 large number, namely 10 000 000. You can change this in two ways:
2671
2672 (a) When configuring PCRE before making, you can use --with-match-limit=n
2673 to set a default value for the compiled library.
2674
2675 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2676 a different value is set. See 45 below.
2677
2678 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2679
2680 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2681 of things that can be changed at compile time. The first argument specifies
2682 what is wanted and the second points to where the information is to be placed.
2683 The current list of available information is:
2684
2685 PCRE_CONFIG_UTF8
2686
2687 The output is an integer that is set to one if UTF-8 support is available;
2688 otherwise it is set to zero.
2689
2690 PCRE_CONFIG_NEWLINE
2691
2692 The output is an integer that it set to the value of the code that is used for
2693 newline. It is either LF (10) or CR (13).
2694
2695 PCRE_CONFIG_LINK_SIZE
2696
2697 The output is an integer that contains the number of bytes used for internal
2698 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2699
2700 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2701
2702 The output is an integer that contains the threshold above which the POSIX
2703 interface uses malloc() for output vectors. See item 31 above.
2704
2705 PCRE_CONFIG_MATCH_LIMIT
2706
2707 The output is an unsigned integer that contains the default limit of the number
2708 of match() calls in a pcre_exec() execution. See 42 above.
2709
2710 44. pcretest has been upgraded by the addition of the -C option. This causes it
2711 to extract all the available output from the new pcre_config() function, and to
2712 output it. The program then exits immediately.
2713
2714 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2715 order to support additional features. One way would have been to define
2716 pcre_exec2() (for example) with extra arguments, but this would not have been
2717 extensible, and would also have required all calls to the original function to
2718 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2719 is used for passing in "extra" data from pcre_study().
2720
2721 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2722 contains the following fields:
2723
2724 flags a bitmap indicating which of the following fields are set
2725 study_data opaque data from pcre_study()
2726 match_limit a way of specifying a limit on match() calls for a specific
2727 call to pcre_exec()
2728 callout_data data for callouts (see 49 below)
2729
2730 The flag bits are also defined in pcre.h, and are
2731
2732 PCRE_EXTRA_STUDY_DATA
2733 PCRE_EXTRA_MATCH_LIMIT
2734 PCRE_EXTRA_CALLOUT_DATA
2735
2736 The pcre_study() function now returns one of these new pcre_extra blocks, with
2737 the actual study data pointed to by the study_data field, and the
2738 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2739 before. That is, this change is entirely upwards-compatible and requires no
2740 change to existing code.
2741
2742 If you want to pass in additional data to pcre_exec(), you can either place it
2743 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2744 block.
2745
2746 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2747 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2748 times with different match limits, until it finds the minimum value needed for
2749 pcre_exec() to complete. The value is then output. This can be instructive; for
2750 most simple matches the number is quite small, but for pathological cases it
2751 gets very large very quickly.
2752
2753 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2754 returns the size of the data block pointed to by the study_data field in a
2755 pcre_extra block, that is, the value that was passed as the argument to
2756 pcre_malloc() when PCRE was getting memory in which to place the information
2757 created by pcre_study(). The fourth argument should point to a size_t variable.
2758 pcretest has been extended so that this information is shown after a successful
2759 pcre_study() call when information about the compiled regex is being displayed.
2760
2761 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2762 because what follows is always an absolute path. (Later: it turns out that this
2763 is more than cosmetic for MinGW, because it doesn't like empty path
2764 components.)
2765
2766 49. Some changes have been made to the callout feature (see 28 above):
2767
2768 (i) A callout function now has three choices for what it returns:
2769
2770 0 => success, carry on matching
2771 > 0 => failure at this point, but backtrack if possible
2772 < 0 => serious error, return this value from pcre_exec()
2773
2774 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2775 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2776 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2777 use by callout functions. It will never be used by PCRE itself.
2778
2779 (ii) The pcre_extra structure (see 45 above) has a void * field called
2780 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2781 pcre_callout_block structure has a field of the same name. The contents of
2782 the field passed in the pcre_extra structure are passed to the callout
2783 function in the corresponding field in the callout block. This makes it
2784 easier to use the same callout-containing regex from multiple threads. For
2785 testing, the pcretest program has a new data escape
2786
2787 \C*n pass the number n (may be negative) as callout_data
2788
2789 If the callout function in pcretest receives a non-zero value as
2790 callout_data, it returns that value.
2791
2792 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2793 there were some redundant $(CFLAGS) in commands that are now specified as
2794 $(LINK), which already includes $(CFLAGS).
2795
2796 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2797 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2798 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2799 one-byte characters throughout. Note that case-insensitive matching applies
2800 only to characters whose values are less than 256. PCRE doesn't support the
2801 notion of cases for higher-valued characters.
2802
2803 (i) A character class whose characters are all within 0-255 is handled as
2804 a bit map, and the map is inverted for negative classes. Previously, a
2805 character > 255 always failed to match such a class; however it should
2806 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2807
2808 (ii) A negated character class with a single character < 255 is coded as
2809 "not this character" (OP_NOT). This wasn't working properly when the test
2810 character was multibyte, either singly or repeated.
2811
2812 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2813 mode, for example: \x{100}{2,3}.
2814
2815 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2816 singly or repeated) now correctly test multibyte characters. However,
2817 PCRE doesn't recognize any characters with values greater than 255 as
2818 digits, spaces, or word characters. Such characters always match \D, \S,
2819 and \W, and never match \d, \s, or \w.
2820
2821 (v) Classes may now contain characters and character ranges with values
2822 greater than 255. For example: [ab\x{100}-\x{400}].
2823
2824 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2825 PCRE in UTF-8 mode.
2826
2827 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2828 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2829 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2830 value.)
2831
2832 53. The single man page has become too large. I have therefore split it up into
2833 a number of separate man pages. These also give rise to individual HTML pages;
2834 these are now put in a separate directory, and there is an index.html page that
2835 lists them all. Some hyperlinking between the pages has been installed.
2836
2837 54. Added convenience functions for handling named capturing parentheses.
2838
2839 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2840 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2841 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2842 are faulted.
2843
2844 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2845 calling configure. These values are used when compiling the dftables.c program
2846 which is run to generate the source of the default character tables. They
2847 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2848 you will need to set these values.
2849
2850 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2851
2852
2853 Version 3.9 02-Jan-02
2854 ---------------------
2855
2856 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2857
2858 2. If --disable-static was given, the building process failed when trying to
2859 build pcretest and pcregrep. (For some reason it was using libtool to compile
2860 them, which is not right, as they aren't part of the library.)
2861
2862
2863 Version 3.8 18-Dec-01
2864 ---------------------
2865
2866 1. The experimental UTF-8 code was completely screwed up. It was packing the
2867 bytes in the wrong order. How dumb can you get?
2868
2869
2870 Version 3.7 29-Oct-01
2871 ---------------------
2872
2873 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2874 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2875 this didn't happen under Solaris 8, where I normally test things.
2876
2877 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2878 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2879 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2880 this fix an hour or so after the initial 3.7 release.)
2881
2882
2883 Version 3.6 23-Oct-01
2884 ---------------------
2885
2886 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2887 offsets passed as NULL with zero offset count.
2888
2889 2. The config.guess and config.sub files had not been updated when I moved to
2890 the latest autoconf.
2891
2892
2893 Version 3.5 15-Aug-01
2894 ---------------------
2895
2896 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2897 had been forgotten.
2898
2899 2. By using declared but undefined structures, we can avoid using "void"
2900 definitions in pcre.h while keeping the internal definitions of the structures
2901 private.
2902
2903 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2904 user point of view, this means that both static and shared libraries are built
2905 by default, but this can be individually controlled. More of the work of
2906 handling this static/shared cases is now inside libtool instead of PCRE's make
2907 file.
2908
2909 4. The pcretest utility is now installed along with pcregrep because it is
2910 useful for users (to test regexs) and by doing this, it automatically gets
2911 relinked by libtool. The documentation has been turned into a man page, so
2912 there are now .1, .txt, and .html versions in /doc.
2913
2914 5. Upgrades to pcregrep:
2915 (i) Added long-form option names like gnu grep.
2916 (ii) Added --help to list all options with an explanatory phrase.
2917 (iii) Added -r, --recursive to recurse into sub-directories.
2918 (iv) Added -f, --file to read patterns from a file.
2919
2920 6. pcre_exec() was referring to its "code" argument before testing that
2921 argument for NULL (and giving an error if it was NULL).
2922
2923 7. Upgraded Makefile.in to allow for compiling in a different directory from
2924 the source directory.
2925
2926 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2927 options bits, the pointer it was passed was to an int instead of to an unsigned
2928 long int. This mattered only on 64-bit systems.
2929
2930 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2931 generated) instead of pcre.in, which it its source. Also made the same change
2932 in several of the .c files.
2933
2934 10. A new release of gcc defines printf() as a macro, which broke pcretest
2935 because it had an ifdef in the middle of a string argument for printf(). Fixed
2936 by using separate calls to printf().
2937
2938 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2939 script, to force use of CR or LF instead of \n in the source. On non-Unix
2940 systems, the value can be set in config.h.
2941
2942 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2943 absolute limit. Changed the text of the error message to make this clear, and
2944 likewise updated the man page.
2945
2946 13. The limit of 99 on the number of capturing subpatterns has been removed.
2947 The new limit is 65535, which I hope will not be a "real" limit.
2948
2949
2950 Version 3.4 22-Aug-00
2951 ---------------------
2952
2953 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2954
2955 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2956
2957
2958 Version 3.3 01-Aug-00
2959 ---------------------
2960
2961 1. If an octal character was given, but the value was greater than \377, it
2962 was not getting masked to the least significant bits, as documented. This could
2963 lead to crashes in some systems.
2964
2965 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2966 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2967
2968 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2969 These just pass their arguments on to (pcre_free)(), but they are provided
2970 because some uses of PCRE bind it to non-C systems that can call its functions,
2971 but cannot call free() or pcre_free() directly.
2972
2973 4. Add "make test" as a synonym for "make check". Corrected some comments in
2974 the Makefile.
2975
2976 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2977 Makefile.
2978
2979 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2980 command called pgrep for grepping around the active processes.
2981
2982 7. Added the beginnings of support for UTF-8 character strings.
2983
2984 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2985 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2986 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2987 out for the ar command.)
2988
2989
2990 Version 3.2 12-May-00
2991 ---------------------
2992
2993 This is purely a bug fixing release.
2994
2995 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2996 of ZA. This was just one example of several cases that could provoke this bug,
2997 which was introduced by change 9 of version 2.00. The code for breaking
2998 infinite loops after an iteration that matches an empty string was't working
2999 correctly.
3000
3001 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3002 when matched against abbab (for example). After matching an empty string, it
3003 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3004 caused it to match further down the string than it should.
3005
3006 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3007 was there because it doesn't seem to be needed, and it causes trouble on some
3008 systems, as it is not a Standard C header. It has been removed.
3009
3010 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3011 were reported on the Macintosh. The changes were from
3012
3013 while ((c = *(++ptr)) != 0 && c != '\n');
3014 to
3015 while ((c = *(++ptr)) != 0 && c != '\n') ;
3016
3017 Totally extraordinary, but if that's what it takes...
3018
3019 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3020 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3021 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3022 assumes the way PCRE uses memmove() (always moving upwards).
3023
3024 6. PCRE is being used in one environment where strchr() is not available. There
3025 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3026 faster code anyway.
3027
3028
3029 Version 3.1 09-Feb-00
3030 ---------------------
3031
3032 The only change in this release is the fixing of some bugs in Makefile.in for
3033 the "install" target:
3034
3035 (1) It was failing to install pcreposix.h.
3036
3037 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3038
3039
3040 Version 3.0 01-Feb-00
3041 ---------------------
3042
3043 1. Add support for the /+ modifier to perltest (to output $` like it does in
3044 pcretest).
3045
3046 2. Add support for the /g modifier to perltest.
3047
3048 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3049 matches null strings.
3050
3051 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3052 pattern. Perl treats empty patterns specially - it reuses the most recent
3053 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3054 effect.
3055
3056 5. The POSIX interface was broken in that it was just handing over the POSIX
3057 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3058 required a bigger vector, with some working space on the end. This means that
3059 the POSIX wrapper now has to get and free some memory, and copy the results.
3060
3061 6. Added some simple autoconf support, placing the test data and the
3062 documentation in separate directories, re-organizing some of the
3063 information files, and making it build pcre-config (a GNU standard). Also added
3064 libtool support for building PCRE as a shared library, which is now the
3065 default.
3066
3067 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3068 09 are not valid octal constants. Single digits will be used for minor values
3069 less than 10.
3070
3071 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3072 existing programs that set these in the POSIX interface can use PCRE without
3073 modification.
3074
3075 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3076 return all that pcre_info() returns, plus additional data. The pcre_info()
3077 function is retained for compatibility, but is considered to be obsolete.
3078
3079 10. Added experimental recursion feature (?R) to handle one common case that
3080 Perl 5.6 will be able to do with (?p{...}).
3081
3082 11. Added support for POSIX character classes like [:alpha:], which Perl is
3083 adopting.
3084
3085
3086 Version 2.08 31-Aug-99
3087 ----------------------
3088
3089 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3090 trying to match at the startoffset position, but instead was moving forward to
3091 the next newline as if a previous match had failed.
3092
3093 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3094 and could get into a loop if a null string was matched other than at the start
3095 of the subject.
3096
3097 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3098 be distinguished at compile time, and for completeness also added PCRE_DATE.
3099
3100 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3101 in GnuWin32 environments.
3102
3103
3104 Version 2.07 29-Jul-99
3105 ----------------------
3106
3107 1. The documentation is now supplied in plain text form and HTML as well as in
3108 the form of man page sources.
3109
3110 2. C++ compilers don't like assigning (void *) values to other pointer types.
3111 In particular this affects malloc(). Although there is no problem in Standard
3112 C, I've put in casts to keep C++ compilers happy.
3113
3114 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3115 should be (const char *).
3116
3117 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3118 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3119 However, I haven't made this a standard facility. The documentation doesn't
3120 mention it, and the Makefile doesn't support it.
3121
3122 5. The Makefile now contains an "install" target, with editable destinations at
3123 the top of the file. The pcretest program is not installed.
3124
3125 6. pgrep -V now gives the PCRE version number and date.
3126
3127 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3128 causing the entire string to be ignored, instead of just the last character.
3129
3130 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3131 non-matching string, it can take a very, very long time, even for strings of
3132 quite modest length, because of the nested recursion. PCRE now does better in
3133 some of these cases. It does this by remembering the last required literal
3134 character in the pattern, and pre-searching the subject to ensure it is present
3135 before running the real match. In other words, it applies a heuristic to detect
3136 some types of certain failure quickly, and in the above example, if presented
3137 with a string that has no trailing " it gives "no match" very quickly.
3138
3139 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3140 other alternatives are tried instead.
3141
3142
3143 Version 2.06 09-Jun-99
3144 ----------------------
3145
3146 1. Change pcretest's output for amount of store used to show just the code
3147 space, because the remainder (the data block) varies in size between 32-bit and
3148 64-bit systems.
3149
3150 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3151 start matching at. This allows lookbehinds to work when searching for multiple
3152 occurrences in a string.
3153
3154 3. Added additional options to pcretest for testing multiple occurrences:
3155
3156 /+ outputs the rest of the string that follows a match
3157 /g loops for multiple occurrences, using the new startoffset argument
3158 /G loops for multiple occurrences by passing an incremented pointer
3159
3160 4. PCRE wasn't doing the "first character" optimization for patterns starting
3161 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3162 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3163 the letter 'x'. On long subject strings, this gives a significant speed-up.
3164
3165
3166 Version 2.05 21-Apr-99
3167 ----------------------
3168
3169 1. Changed the type of magic_number from int to long int so that it works
3170 properly on 16-bit systems.
3171
3172 2. Fixed a bug which caused patterns starting with .* not to work correctly
3173 when the subject string contained newline characters. PCRE was assuming
3174 anchoring for such patterns in all cases, which is not correct because .* will
3175 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3176 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3177 must be retried after every newline in the subject.
3178
3179
3180 Version 2.04 18-Feb-99
3181 ----------------------
3182
3183 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3184 computation of the store needed to hold the pattern was incorrect (too large).
3185 If such patterns were nested a few deep, this could multiply and become a real
3186 problem.
3187
3188 2. Added /M option to pcretest to show the memory requirement of a specific
3189 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3190
3191 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3192 compiled in such a way that the backtracking after subsequent failure was
3193 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3194 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3195
3196
3197 Version 2.03 02-Feb-99
3198 ----------------------
3199
3200 1. Fixed typo and small mistake in man page.
3201
3202 2. Added 4th condition (GPL supersedes if conflict) and created separate
3203 LICENCE file containing the conditions.
3204
3205 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3206 Perl, that is the internal \ allows the delimiter to be included in the
3207 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3208 the final delimiter, add \ to the end of the pattern (to test the error).
3209
3210 4. Added the convenience functions for extracting substrings after a successful
3211 match. Updated pcretest to make it able to test these functions.
3212
3213
3214 Version 2.02 14-Jan-99
3215 ----------------------
3216
3217 1. Initialized the working variables associated with each extraction so that
3218 their saving and restoring doesn't refer to uninitialized store.
3219
3220 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3221 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3222 fix the problem.
3223
3224 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3225 calls, and wasn't printing the correct value for compiling calls. Increased the
3226 default value of LOOPREPEAT, and the number of significant figures in the
3227 times.
3228
3229 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3230
3231 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3232 a building problem on Windows NT with a FAT file system.
3233
3234
3235 Version 2.01 21-Oct-98
3236 ----------------------
3237
3238 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3239 to character tables built by pcre_maketables() in the current locale. If NULL
3240 is passed, the default tables are used.
3241
3242
3243 Version 2.00 24-Sep-98
3244 ----------------------
3245
3246 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3247 it any more.
3248
3249 2. Allow quantification of (?>) groups, and make it work correctly.
3250
3251 3. The first character computation wasn't working for (?>) groups.
3252
3253 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3254 end of the subject) and add 5.005's \z, which really does match only at the
3255 very end of the subject.
3256
3257 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3258
3259 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3260 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3261 localized options. All options to pcre_study() were also removed.
3262
3263 7. Add other new features from 5.005:
3264
3265 $(?<= positive lookbehind
3266 $(?<! negative lookbehind
3267 (?imsx-imsx) added the unsetting capability
3268 such a setting is global if at outer level; local otherwise
3269 (?imsx-imsx:) non-capturing groups with option setting
3270 (?(cond)re|re) conditional pattern matching
3271
3272 A backreference to itself in a repeated group matches the previous
3273 captured string.
3274
3275 8. General tidying up of studying (both automatic and via "study")
3276 consequential on the addition of new assertions.
3277
3278 9. As in 5.005, unlimited repeated groups that could match an empty substring
3279 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3280 runtime if any iteration does actually match an empty substring.
3281
3282 10. Include the RunTest script in the distribution.
3283
3284 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3285 discrepancies, some of which were old and were also with respect to 5.004. They
3286 have now been fixed.
3287
3288
3289 Version 1.09 28-Apr-98
3290 ----------------------
3291
3292 1. A negated single character class followed by a quantifier with a minimum
3293 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3294 program crashes, or just wrong answers. This did not apply to negated classes
3295 containing more than one character, or to minima other than one.
3296
3297
3298 Version 1.08 27-Mar-98
3299 ----------------------
3300
3301 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3302
3303 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3304 latter must appear before anything that relies on it in the pattern.
3305
3306
3307 Version 1.07 16-Feb-98
3308 ----------------------
3309
3310 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3311 repeat of a potentially empty string).
3312
3313
3314 Version 1.06 23-Jan-98
3315 ----------------------
3316
3317 1. Added Markus Oberhumer's little patches for C++.
3318
3319 2. Literal strings longer than 255 characters were broken.
3320
3321
3322 Version 1.05 23-Dec-97
3323 ----------------------
3324
3325 1. Negated character classes containing more than one character were failing if
3326 PCRE_CASELESS was set at run time.
3327
3328
3329 Version 1.04 19-Dec-97
3330 ----------------------
3331
3332 1. Corrected the man page, where some "const" qualifiers had been omitted.
3333
3334 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3335 input syntax.
3336
3337 3. Fixed memory leak which occurred when a regex with back references was
3338 matched with an offsets vector that wasn't big enough. The temporary memory
3339 that is used in this case wasn't being freed if the match failed.
3340
3341 4. Tidied pcretest to ensure it frees memory that it gets.
3342
3343 5. Temporary memory was being obtained in the case where the passed offsets
3344 vector was exactly big enough.
3345
3346 6. Corrected definition of offsetof() from change 5 below.
3347
3348 7. I had screwed up change 6 below and broken the rules for the use of
3349 setjmp(). Now fixed.
3350
3351
3352 Version 1.03 18-Dec-97
3353 ----------------------
3354
3355 1. A erroneous regex with a missing opening parenthesis was correctly
3356 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3357 on some systems.
3358
3359 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3360 it was reported that one broken compiler failed on the former because "code" is
3361 also an independent variable.
3362
3363 3. The erroneous regex a[]b caused an array overrun reference.
3364
3365 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3366 fail on data ending with that character. (It was going on too far, and checking
3367 the next character, typically a binary zero.) This was specific to the
3368 optimized code for single-character negative classes.
3369
3370 5. Added a contributed patch from the TIN world which does the following:
3371
3372 + Add an undef for memmove, in case the the system defines a macro for it.
3373
3374 + Add a definition of offsetof(), in case there isn't one. (I don't know
3375 the reason behind this - offsetof() is part of the ANSI standard - but
3376 it does no harm).
3377
3378 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3379 most of the places where whitespace preceded '#'. I have given up and
3380 allowed the remaining 2 cases to be at the margin.
3381
3382 + Rename some variables in pcre to eliminate shadowing. This seems very
3383 pedantic, but does no harm, of course.
3384
3385 6. Moved the call to setjmp() into its own function, to get rid of warnings
3386 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3387
3388 7. Constructs such as \d{8,} were compiling into the equivalent of
3389 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3390 outcome, but in this particular case used more store than had been allocated,
3391 which caused the bug to be discovered because it threw up an internal error.
3392
3393 8. The debugging code in both pcre and pcretest for outputting the compiled
3394 form of a regex was going wrong in the case of back references followed by
3395 curly-bracketed repeats.
3396
3397
3398 Version 1.02 12-Dec-97
3399 ----------------------
3400
3401 1. Typos in pcre.3 and comments in the source fixed.
3402
3403 2. Applied a contributed patch to get rid of places where it used to remove
3404 'const' from variables, and fixed some signed/unsigned and uninitialized
3405 variable warnings.
3406
3407 3. Added the "runtest" target to Makefile.
3408
3409 4. Set default compiler flag to -O2 rather than just -O.
3410
3411
3412 Version 1.01 19-Nov-97
3413 ----------------------
3414
3415 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3416 like /([ab]*)*/, that is, for classes with more than one character in them.
3417
3418 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3419 as /((?>a*))*/ (a PCRE_EXTRA facility).
3420
3421
3422 Version 1.00 18-Nov-97
3423 ----------------------
3424
3425 1. Added compile-time macros to support systems such as SunOS4 which don't have
3426 memmove() or strerror() but have other things that can be used instead.
3427
3428 2. Arranged that "make clean" removes the executables.
3429
3430
3431 Version 0.99 27-Oct-97
3432 ----------------------
3433
3434 1. Fixed bug in code for optimizing classes with only one character. It was
3435 initializing a 32-byte map regardless, which could cause it to run off the end
3436 of the memory it had got.
3437
3438 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3439
3440
3441 Version 0.98 22-Oct-97
3442 ----------------------
3443
3444 1. Fixed bug in code for handling temporary memory usage when there are more
3445 back references than supplied space in the ovector. This could cause segfaults.
3446
3447
3448 Version 0.97 21-Oct-97
3449 ----------------------
3450
3451 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3452
3453 2. Optimized negated single characters not to use a bit map.
3454
3455 3. Brought error texts together as macro definitions; clarified some of them;
3456 fixed one that was wrong - it said "range out of order" when it meant "invalid
3457 escape sequence".
3458
3459 4. Changed some char * arguments to const char *.
3460
3461 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3462
3463 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3464 pcretest.
3465
3466
3467 Version 0.96 16-Oct-97
3468 ----------------------
3469
3470 1. Added a simple "pgrep" utility to the distribution.
3471
3472 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3473 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3474 where "ddd" means "one or more decimal digits".
3475
3476 3. Fixed serious bug. If a pattern had a back reference, but the call to
3477 pcre_exec() didn't supply a large enough ovector to record the related
3478 identifying subpattern, the match always failed. PCRE now remembers the number
3479 of the largest back reference, and gets some temporary memory in which to save
3480 the offsets during matching if necessary, in order to ensure that
3481 backreferences always work.
3482
3483 4. Increased the compatibility with Perl in a number of ways:
3484
3485 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3486 to request this handling. The option can be set at compile or exec time.
3487
3488 (b) $ matches before a terminating newline by default; an option
3489 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3490 mode). The option can be set at compile or exec time.
3491
3492 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3493 the same as Perl's. If the decimal number it represents is less than 10
3494 or there aren't that many previous left capturing parentheses, an octal
3495 escape is read. Inside a character class, it's always an octal escape,
3496 even if it is a single digit.
3497
3498 (d) An escaped but undefined alphabetic character is taken as a literal,
3499 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3500 escapes.
3501
3502 (e) {0} is now permitted. (The previous item is removed from the compiled
3503 pattern).
3504
3505 5. Changed all the names of code files so that the basic parts are no longer
3506 than 10 characters, and abolished the teeny "globals.c" file.
3507
3508 6. Changed the handling of character classes; they are now done with a 32-byte
3509 bit map always.
3510
3511 7. Added the -d and /D options to pcretest to make it possible to look at the
3512 internals of compilation without having to recompile pcre.
3513
3514
3515 Version 0.95 23-Sep-97
3516 ----------------------
3517
3518 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3519 \x20 at the start of a run of normal characters. These were being treated as
3520 real characters, instead of the source characters being re-checked.
3521
3522
3523 Version 0.94 18-Sep-97
3524 ----------------------
3525
3526 1. The functions are now thread-safe, with the caveat that the global variables
3527 containing pointers to malloc() and free() or alternative functions are the
3528 same for all threads.
3529
3530 2. Get pcre_study() to generate a bitmap of initial characters for non-
3531 anchored patterns when this is possible, and use it if passed to pcre_exec().
3532
3533
3534 Version 0.93 15-Sep-97
3535 ----------------------
3536
3537 1. /(b)|(:+)/ was computing an incorrect first character.
3538
3539 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3540 but not actually doing anything yet.
3541
3542 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3543 as Perl does (e.g. [-az] or [az-]).
3544
3545 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3546 all possible positions.
3547
3548 5. Split up into different modules to avoid including unneeded functions in a
3549 compiled binary. However, compile and exec are still in one module. The "study"
3550 function is split off.
3551
3552 6. The character tables are now in a separate module whose source is generated
3553 by an auxiliary program - but can then be edited by hand if required. There are
3554 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3555 toupper() in the code.
3556
3557 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3558 make them global. Abolish the function for setting them, as the caller can now
3559 set them directly.
3560
3561
3562 Version 0.92 11-Sep-97
3563 ----------------------
3564
3565 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3566 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3567
3568 2. Caseless matching was not working in character classes if the characters in
3569 the pattern were in upper case.
3570
3571 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3572
3573 4. Make PCRE_ANCHORED public and accept as a compile option.
3574
3575 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3576 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3577 pass them.
3578
3579 6. Give an error if bad option bits passed at compile or run time.
3580
3581 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3582 pcretest to cause it to pass that flag.
3583
3584 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3585 options, and the first character, if set.
3586
3587 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3588
3589
3590 Version 0.91 10-Sep-97
3591 ----------------------
3592
3593 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3594 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3595
3596 2. PCRE was looping on encountering an indefinitely repeated back reference to
3597 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3598 Perl does - treats the match as successful.
3599
3600 ****

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5