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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 397 - (show annotations) (download)
Fri Mar 20 19:40:08 2009 UTC (5 years, 6 months ago) by ph10
File size: 163518 byte(s)
Fix non-callout function supplied for pcre_dfa_exec() bug.

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

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