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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 390 - (show annotations) (download)
Tue Mar 17 10:58:16 2009 UTC (5 years, 6 months ago) by ph10
File size: 162528 byte(s)
Comments about no_arg in the C++ wrapper.

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

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