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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 427 - (show annotations) (download)
Fri Aug 28 09:55:54 2009 UTC (4 years, 11 months ago) by ph10
File size: 168654 byte(s)
Add new PCRE_PARTIAL_HARD option.

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8 was out-of-date, and there was no check on the pcre_compile() error code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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12