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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 392 - (show annotations) (download)
Tue Mar 17 21:30:30 2009 UTC (5 years ago) by ph10
File size: 162635 byte(s)
Update after detrailing for a test release.

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

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