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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 381 - (show annotations) (download)
Tue Mar 3 16:08:23 2009 UTC (5 years, 7 months ago) by ph10
File size: 162010 byte(s)
Fix bug with (?(?=.*b)b|^) thinking it must match at start of line; also fix 
bug causing a crash when auto-callout is used with a conditional assertion.

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

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