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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 399 - (show annotations) (download)
Sat Mar 21 12:34:15 2009 UTC (5 years, 6 months ago) by ph10
File size: 163599 byte(s)
Further fix to auto-callout with conditional groups whose condition is an 
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 whose
46 condition was an assertion, PCRE could crash during matching, both with
47 pcre_exec() and pcre_dfa_exec().
48
49 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
50 used for matching.
51
52 11. Unicode property support in character classes was not working for
53 characters (bytes) greater than 127 when not in UTF-8 mode.
54
55 12. Added the -M command line option to pcretest.
56
57 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
58
59 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
60
61 16. Added comments and documentation about mis-use of no_arg in the C++
62 wrapper.
63
64 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
65 from Martin Jerabek that uses macro names for all relevant character and
66 string constants.
67
68 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
69 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
70 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
71 these, but not everybody uses configure.
72
73 19. A conditional group that had only one branch was not being correctly
74 recognized as an item that could match an empty string. This meant that an
75 enclosing group might also not be so recognized, causing infinite looping
76 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
77 with the subject "ab", where knowledge that the repeated group can match
78 nothing is needed in order to break the loop.
79
80 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
81 exec(), but without supplying a callout function, matching went wrong.
82
83
84 Version 7.8 05-Sep-08
85 ---------------------
86
87 1. Replaced UCP searching code with optimized version as implemented for Ad
88 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
89 stage table and inline lookup instead of a function, giving speed ups of 2
90 to 5 times on some simple patterns that I tested. Permission was given to
91 distribute the MultiStage2.py script that generates the tables (it's not in
92 the tarball, but is in the Subversion repository).
93
94 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
95 scripts.
96
97 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
98 a group with a zero qualifier. The result of the study could be incorrect,
99 or the function might crash, depending on the pattern.
100
101 4. Caseless matching was not working for non-ASCII characters in back
102 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
103 It now works when Unicode Property Support is available.
104
105 5. In pcretest, an escape such as \x{de} in the data was always generating
106 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
107 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
108 truncation.
109
110 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
111
112 7. Added two (int) casts to pcregrep when printing the difference of two
113 pointers, in case they are 64-bit values.
114
115 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
116 test 2 if it fails.
117
118 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
119 and a #define of that name to empty if it is not externally set. This is to
120 allow users of MSVC to set it if necessary.
121
122 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
123 the convenience functions in the pcre_get.c source file.
124
125 11. An option change at the start of a pattern that had top-level alternatives
126 could cause overwriting and/or a crash. This command provoked a crash in
127 some environments:
128
129 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
130
131 This potential security problem was recorded as CVE-2008-2371.
132
133 12. For a pattern where the match had to start at the beginning or immediately
134 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
135 pcre_dfa_exec() could read past the end of the passed subject if there was
136 no match. To help with detecting such bugs (e.g. with valgrind), I modified
137 pcretest so that it places the subject at the end of its malloc-ed buffer.
138
139 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
140 exec() might read past the end of the data buffer in UTF-8 mode.
141
142 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
143 the data contained the byte 0x85 as part of a UTF-8 character within its
144 first line. This applied both to normal and DFA matching.
145
146 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
147 /^[^d]*?$/8 failed to match "abc".
148
149 16. Added a missing copyright notice to pcrecpp_internal.h.
150
151 17. Make it more clear in the documentation that values returned from
152 pcre_exec() in ovector are byte offsets, not character counts.
153
154 18. Tidied a few places to stop certain compilers from issuing warnings.
155
156 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
157 supplied by Stefan Weber. I made a further small update for 7.8 because
158 there is a change of source arrangements: the pcre_searchfuncs.c module is
159 replaced by pcre_ucd.c.
160
161
162 Version 7.7 07-May-08
163 ---------------------
164
165 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
166 a string to a long long, pretend we don't even have a long long." This is
167 done by checking for the strtoq, strtoll, and _strtoi64 functions.
168
169 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
170 pre-7.6 versions, which defined a global no_arg variable instead of putting
171 it in the RE class. (See also #8 below.)
172
173 3. Remove a line of dead code, identified by coverity and reported by Nuno
174 Lopes.
175
176 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
177
178 (1) The include/exclude patterns were being applied to the whole pathnames
179 of files, instead of just to the final components.
180
181 (2) If there was more than one level of directory, the subdirectories were
182 skipped unless they satisfied the include/exclude conditions. This is
183 inconsistent with GNU grep (and could even be seen as contrary to the
184 pcregrep specification - which I improved to make it absolutely clear).
185 The action now is always to scan all levels of directory, and just
186 apply the include/exclude patterns to regular files.
187
188 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
189 --exclude_dir in the tests to avoid scanning .svn directories.
190
191 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
192 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
193 doesn't support NULs in patterns.
194
195 7. Added some missing "const"s to declarations of static tables in
196 pcre_compile.c and pcre_dfa_exec.c.
197
198 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
199 caused by fix #2 above. (Subsequently also a second patch to fix the
200 first patch. And a third patch - this was a messy problem.)
201
202 9. Applied Craig's patch to remove the use of push_back().
203
204 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
205 matching function regexec().
206
207 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
208 which, however, unlike Perl's \g{...}, are subroutine calls, not back
209 references. PCRE supports relative numbers with this syntax (I don't think
210 Oniguruma does).
211
212 12. Previously, a group with a zero repeat such as (...){0} was completely
213 omitted from the compiled regex. However, this means that if the group
214 was called as a subroutine from elsewhere in the pattern, things went wrong
215 (an internal error was given). Such groups are now left in the compiled
216 pattern, with a new opcode that causes them to be skipped at execution
217 time.
218
219 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
220 to the way PCRE behaves:
221
222 (a) A lone ] character is dis-allowed (Perl treats it as data).
223
224 (b) A back reference to an unmatched subpattern matches an empty string
225 (Perl fails the current match path).
226
227 (c) A data ] in a character class must be notated as \] because if the
228 first data character in a class is ], it defines an empty class. (In
229 Perl it is not possible to have an empty class.) The empty class []
230 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
231 The negative empty class [^] matches any one character, independently
232 of the DOTALL setting.
233
234 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
235 non-existent subpattern following a character class starting with ']' and
236 containing () gave an internal compiling error instead of "reference to
237 non-existent subpattern". Fortunately, when the pattern did exist, the
238 compiled code was correct. (When scanning forwards to check for the
239 existencd of the subpattern, it was treating the data ']' as terminating
240 the class, so got the count wrong. When actually compiling, the reference
241 was subsequently set up correctly.)
242
243 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
244 it was being rejected as not supported by pcre_dfa_exec(), even though
245 other assertions are supported. I have made pcre_dfa_exec() support
246 (*FAIL).
247
248 16. The implementation of 13c above involved the invention of a new opcode,
249 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
250 cannot be changed at match time, I realized I could make a small
251 improvement to matching performance by compiling OP_ALLANY instead of
252 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
253 on the OP_ANY path.
254
255 17. Compiling pcretest on Windows with readline support failed without the
256 following two fixes: (1) Make the unistd.h include conditional on
257 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
258
259 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
260 ncurses library to be included for pcretest when ReadLine support is
261 requested, but also to allow for it to be overridden. This patch came from
262 Daniel Bergström.
263
264 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
265 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
266 any errors with the current Unicode tables. Thanks to Peter Kankowski for
267 spotting this.
268
269
270 Version 7.6 28-Jan-08
271 ---------------------
272
273 1. A character class containing a very large number of characters with
274 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
275 overflow.
276
277 2. Patch to cut out the "long long" test in pcrecpp_unittest when
278 HAVE_LONG_LONG is not defined.
279
280 3. Applied Christian Ehrlicher's patch to update the CMake build files to
281 bring them up to date and include new features. This patch includes:
282
283 - Fixed PH's badly added libz and libbz2 support.
284 - Fixed a problem with static linking.
285 - Added pcredemo. [But later removed - see 7 below.]
286 - Fixed dftables problem and added an option.
287 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
288 HAVE_LONG_LONG.
289 - Added readline support for pcretest.
290 - Added an listing of the option settings after cmake has run.
291
292 4. A user submitted a patch to Makefile that makes it easy to create
293 "pcre.dll" under mingw when using Configure/Make. I added stuff to
294 Makefile.am that cause it to include this special target, without
295 affecting anything else. Note that the same mingw target plus all
296 the other distribution libraries and programs are now supported
297 when configuring with CMake (see 6 below) instead of with
298 Configure/Make.
299
300 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
301 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
302 exported in the Windows port". It has not yet been confirmed that the patch
303 solves the problem, but it does no harm.
304
305 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
306 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
307 with CMake, and also correct the comment about stack recursion.
308
309 7. Remove the automatic building of pcredemo from the ./configure system and
310 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
311 of a program that users should build themselves after PCRE is installed, so
312 building it automatically is not really right. What is more, it gave
313 trouble in some build environments.
314
315 8. Further tidies to CMakeLists.txt from Sheri and Christian.
316
317
318 Version 7.5 10-Jan-08
319 ---------------------
320
321 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
322 values in parens when parsing an RE using the C++ wrapper."
323
324 2. Negative specials like \S did not work in character classes in UTF-8 mode.
325 Characters greater than 255 were excluded from the class instead of being
326 included.
327
328 3. The same bug as (2) above applied to negated POSIX classes such as
329 [:^space:].
330
331 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
332 defined or documented. It seems to have been a typo for PCRE_STATIC, so
333 I have changed it.
334
335 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
336 first named subpattern) and a construct such as (?&a) would reference the
337 first named subpattern whose name started with "a" (in other words, the
338 length check was missing). Both these problems are fixed. "Subpattern name
339 expected" is now given for (?&) (a zero-length name), and this patch also
340 makes it give the same error for \k'' (previously it complained that that
341 was a reference to a non-existent subpattern).
342
343 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
344 this is right because (?- can be followed by option settings as well as by
345 digits. I have, however, made the messages clearer.
346
347 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
348 than the number used in the conditional) now cause a compile-time error.
349 This is actually not compatible with Perl, which accepts such patterns, but
350 treats the conditional as always being FALSE (as PCRE used to), but it
351 seems to me that giving a diagnostic is better.
352
353 8. Change "alphameric" to the more common word "alphanumeric" in comments
354 and messages.
355
356 9. Fix two occurrences of "backslash" in comments that should have been
357 "backspace".
358
359 10. Remove two redundant lines of code that can never be obeyed (their function
360 was moved elsewhere).
361
362 11. The program that makes PCRE's Unicode character property table had a bug
363 which caused it to generate incorrect table entries for sequences of
364 characters that have the same character type, but are in different scripts.
365 It amalgamated them into a single range, with the script of the first of
366 them. In other words, some characters were in the wrong script. There were
367 thirteen such cases, affecting characters in the following ranges:
368
369 U+002b0 - U+002c1
370 U+0060c - U+0060d
371 U+0061e - U+00612
372 U+0064b - U+0065e
373 U+0074d - U+0076d
374 U+01800 - U+01805
375 U+01d00 - U+01d77
376 U+01d9b - U+01dbf
377 U+0200b - U+0200f
378 U+030fc - U+030fe
379 U+03260 - U+0327f
380 U+0fb46 - U+0fbb1
381 U+10450 - U+1049d
382
383 12. The -o option (show only the matching part of a line) for pcregrep was not
384 compatible with GNU grep in that, if there was more than one match in a
385 line, it showed only the first of them. It now behaves in the same way as
386 GNU grep.
387
388 13. If the -o and -v options were combined for pcregrep, it printed a blank
389 line for every non-matching line. GNU grep prints nothing, and pcregrep now
390 does the same. The return code can be used to tell if there were any
391 non-matching lines.
392
393 14. Added --file-offsets and --line-offsets to pcregrep.
394
395 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
396 infinitely looping recursion. The bug was that positive lookaheads were not
397 being skipped when checking for a possible empty match (negative lookaheads
398 and both kinds of lookbehind were skipped).
399
400 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
401 inclusion of <windows.h> to before rather than after the definition of
402 INVALID_FILE_ATTRIBUTES (patch from David Byron).
403
404 17. Specifying a possessive quantifier with a specific limit for a Unicode
405 character property caused pcre_compile() to compile bad code, which led at
406 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
407 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
408 caused the error; without that there was no problem.
409
410 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
411
412 19. Added --enable-pcretest-libreadline.
413
414 20. In pcrecpp.cc, the variable 'count' was incremented twice in
415 RE::GlobalReplace(). As a result, the number of replacements returned was
416 double what it should be. I removed one of the increments, but Craig sent a
417 later patch that removed the other one (the right fix) and added unit tests
418 that check the return values (which was not done before).
419
420 21. Several CMake things:
421
422 (1) Arranged that, when cmake is used on Unix, the libraries end up with
423 the names libpcre and libpcreposix, not just pcre and pcreposix.
424
425 (2) The above change means that pcretest and pcregrep are now correctly
426 linked with the newly-built libraries, not previously installed ones.
427
428 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
429
430 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
431 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
432 UTF-8 newline character). The key issue is that the pattern starts .*;
433 this means that the match must be either at the beginning, or after a
434 newline. The bug was in the code for advancing after a failed match and
435 checking that the new position followed a newline. It was not taking
436 account of UTF-8 characters correctly.
437
438 23. PCRE was behaving differently from Perl in the way it recognized POSIX
439 character classes. PCRE was not treating the sequence [:...:] as a
440 character class unless the ... were all letters. Perl, however, seems to
441 allow any characters between [: and :], though of course it rejects as
442 unknown any "names" that contain non-letters, because all the known class
443 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
444 for example, whereas PCRE did not - it did not recognize a POSIX character
445 class. This seemed a bit dangerous, so the code has been changed to be
446 closer to Perl. The behaviour is not identical to Perl, because PCRE will
447 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
448 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
449 Perl does, and where it didn't before.
450
451 24. Rewrite so as to remove the single use of %n from pcregrep because in some
452 Windows environments %n is disabled by default.
453
454
455 Version 7.4 21-Sep-07
456 ---------------------
457
458 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
459 means that a class such as [\s] counted as "explicit reference to CR or
460 LF". That isn't really right - the whole point of the change was to try to
461 help when there was an actual mention of one of the two characters. So now
462 the change happens only if \r or \n (or a literal CR or LF) character is
463 encountered.
464
465 2. The 32-bit options word was also used for 6 internal flags, but the numbers
466 of both had grown to the point where there were only 3 bits left.
467 Fortunately, there was spare space in the data structure, and so I have
468 moved the internal flags into a new 16-bit field to free up more option
469 bits.
470
471 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
472 but did not set the internal JCHANGED flag - either of these is enough to
473 control the way the "get" function works - but the PCRE_INFO_JCHANGED
474 facility is supposed to tell if (?J) was ever used, so now (?J) at the
475 start sets both bits.
476
477 4. Added options (at build time, compile time, exec time) to change \R from
478 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
479
480 5. doc/pcresyntax.html was missing from the distribution.
481
482 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
483 compatibility, even though it is no longer used.
484
485 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
486 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
487 windows.h file is present (where different names are used). [This was
488 reversed later after testing - see 16 below.]
489
490 8. Changed all #include <config.h> to #include "config.h". There were also
491 some further <pcre.h> cases that I changed to "pcre.h".
492
493 9. When pcregrep was used with the --colour option, it missed the line ending
494 sequence off the lines that it output.
495
496 10. It was pointed out to me that arrays of string pointers cause lots of
497 relocations when a shared library is dynamically loaded. A technique of
498 using a single long string with a table of offsets can drastically reduce
499 these. I have refactored PCRE in four places to do this. The result is
500 dramatic:
501
502 Originally: 290
503 After changing UCP table: 187
504 After changing error message table: 43
505 After changing table of "verbs" 36
506 After changing table of Posix names 22
507
508 Thanks to the folks working on Gregex for glib for this insight.
509
510 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
511 unicode-properties was also set.
512
513 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
514
515 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
516 checked only for CRLF.
517
518 14. Added casts to pcretest.c to avoid compiler warnings.
519
520 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
521
522 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
523 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
524 entirely. This removes changes made in 7 above.
525
526 17. The CMake files have been updated, and there is now more information about
527 building with CMake in the NON-UNIX-USE document.
528
529
530 Version 7.3 28-Aug-07
531 ---------------------
532
533 1. In the rejigging of the build system that eventually resulted in 7.1, the
534 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
535 brackets there is not right, since it causes compilers to look for an
536 installed pcre.h, not the version that is in the source that is being
537 compiled (which of course may be different). I have changed it back to:
538
539 #include "pcre.h"
540
541 I have a vague recollection that the change was concerned with compiling in
542 different directories, but in the new build system, that is taken care of
543 by the VPATH setting the Makefile.
544
545 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
546 when the subject happened to end in the byte 0x85 (e.g. if the last
547 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
548 characters but of course it shouldn't be taken as a newline when it is part
549 of another character. The bug was that, for an unlimited repeat of . in
550 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
551 characters when looking for a newline.
552
553 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
554
555 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
556 in debug output.
557
558 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
559 long printing in the pcrecpp unittest when running under MinGW.
560
561 6. ESC_K was left out of the EBCDIC table.
562
563 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
564 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
565 limit also applies to "virtual nesting" when a pattern is recursive, and in
566 this case 1000 isn't so big. I have been able to remove this limit at the
567 expense of backing off one optimization in certain circumstances. Normally,
568 when pcre_exec() would call its internal match() function recursively and
569 immediately return the result unconditionally, it uses a "tail recursion"
570 feature to save stack. However, when a subpattern that can match an empty
571 string has an unlimited repetition quantifier, it no longer makes this
572 optimization. That gives it a stack frame in which to save the data for
573 checking that an empty string has been matched. Previously this was taken
574 from the 1000-entry workspace that had been reserved. So now there is no
575 explicit limit, but more stack is used.
576
577 8. Applied Daniel's patches to solve problems with the import/export magic
578 syntax that is required for Windows, and which was going wrong for the
579 pcreposix and pcrecpp parts of the library. These were overlooked when this
580 problem was solved for the main library.
581
582 9. There were some crude static tests to avoid integer overflow when computing
583 the size of patterns that contain repeated groups with explicit upper
584 limits. As the maximum quantifier is 65535, the maximum group length was
585 set at 30,000 so that the product of these two numbers did not overflow a
586 32-bit integer. However, it turns out that people want to use groups that
587 are longer than 30,000 bytes (though not repeat them that many times).
588 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
589 made it possible to implement the integer overflow checks in a much more
590 dynamic way, which I have now done. The artificial limitation on group
591 length has been removed - we now have only the limit on the total length of
592 the compiled pattern, which depends on the LINK_SIZE setting.
593
594 10. Fixed a bug in the documentation for get/copy named substring when
595 duplicate names are permitted. If none of the named substrings are set, the
596 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
597 empty string.
598
599 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
600 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
601 because the ] is interpreted as the first data character and the
602 terminating ] is not found. PCRE has been made compatible with Perl in this
603 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
604 cause memory overwriting.
605
606 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
607 string has been matched (to stop an infinite loop). It was not recognizing
608 a conditional subpattern that could match an empty string if that
609 subpattern was within another subpattern. For example, it looped when
610 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
611 condition was not nested. This bug has been fixed.
612
613 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
614 past the start of the subject in the presence of bytes with the top bit
615 set, for example "\x8aBCD".
616
617 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
618 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
619
620 14. Optimized (?!) to (*FAIL).
621
622 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
623 This restricts code points to be within the range 0 to 0x10FFFF, excluding
624 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
625 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
626 does: it's just the validity check that is more restrictive.
627
628 16. Inserted checks for integer overflows during escape sequence (backslash)
629 processing, and also fixed erroneous offset values for syntax errors during
630 backslash processing.
631
632 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
633 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
634
635 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
636 caused an overrun.
637
638 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
639 something other than just ASCII characters) inside a group that had an
640 unlimited repeat caused a loop at compile time (while checking to see
641 whether the group could match an empty string).
642
643 20. Debugging a pattern containing \p or \P could cause a crash. For example,
644 [\P{Any}] did so. (Error in the code for printing property names.)
645
646 21. An orphan \E inside a character class could cause a crash.
647
648 22. A repeated capturing bracket such as (A)? could cause a wild memory
649 reference during compilation.
650
651 23. There are several functions in pcre_compile() that scan along a compiled
652 expression for various reasons (e.g. to see if it's fixed length for look
653 behind). There were bugs in these functions when a repeated \p or \P was
654 present in the pattern. These operators have additional parameters compared
655 with \d, etc, and these were not being taken into account when moving along
656 the compiled data. Specifically:
657
658 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
659 length.
660
661 (b) An item such as \pL+ within a repeated group could cause crashes or
662 loops.
663
664 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
665 "reference to non-existent subpattern" error.
666
667 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
668
669 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
670 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
671
672 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
673
674 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
675 character were causing crashes (broken optimization).
676
677 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
678 \p or \P) caused a compile-time loop.
679
680 28. More problems have arisen in unanchored patterns when CRLF is a valid line
681 break. For example, the unstudied pattern [\r\n]A does not match the string
682 "\r\nA" because change 7.0/46 below moves the current point on by two
683 characters after failing to match at the start. However, the pattern \nA
684 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
685 the same is true. There doesn't seem any very clean way out of this, but
686 what I have chosen to do makes the common cases work: PCRE now takes note
687 of whether there can be an explicit match for \r or \n anywhere in the
688 pattern, and if so, 7.0/46 no longer applies. As part of this change,
689 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
690 pattern has explicit CR or LF references.
691
692 29. Added (*CR) etc for changing newline setting at start of pattern.
693
694
695 Version 7.2 19-Jun-07
696 ---------------------
697
698 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
699 which is apparently normally available under Windows.
700
701 2. Re-jig the pcregrep tests with different newline settings in an attempt
702 to make them independent of the local environment's newline setting.
703
704 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
705
706 4. Some of the "internals" tests were previously cut out when the link size
707 was not 2, because the output contained actual offsets. The recent new
708 "Z" feature of pcretest means that these can be cut out, making the tests
709 usable with all link sizes.
710
711 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
712 stack recursion. This gives a massive performance boost under BSD, but just
713 a small improvement under Linux. However, it saves one field in the frame
714 in all cases.
715
716 6. Added more features from the forthcoming Perl 5.10:
717
718 (a) (?-n) (where n is a string of digits) is a relative subroutine or
719 recursion call. It refers to the nth most recently opened parentheses.
720
721 (b) (?+n) is also a relative subroutine call; it refers to the nth next
722 to be opened parentheses.
723
724 (c) Conditions that refer to capturing parentheses can be specified
725 relatively, for example, (?(-2)... or (?(+3)...
726
727 (d) \K resets the start of the current match so that everything before
728 is not part of it.
729
730 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
731
732 (f) \g{name} is another synonym - part of Perl 5.10's unification of
733 reference syntax.
734
735 (g) (?| introduces a group in which the numbering of parentheses in each
736 alternative starts with the same number.
737
738 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
739
740 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
741 PCRE_INFO_JCHANGED.
742
743 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
744 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
745 for detecting groups that can match an empty string.
746
747 9. A pattern with a very large number of alternatives (more than several
748 hundred) was running out of internal workspace during the pre-compile
749 phase, where pcre_compile() figures out how much memory will be needed. A
750 bit of new cunning has reduced the workspace needed for groups with
751 alternatives. The 1000-alternative test pattern now uses 12 bytes of
752 workspace instead of running out of the 4096 that are available.
753
754 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
755
756 11. Applied patch from Google to remove an optimization that didn't quite work.
757 The report of the bug said:
758
759 pcrecpp::RE("a*").FullMatch("aaa") matches, while
760 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
761 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
762
763 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
764 it matched the wrong number of bytes.
765
766
767 Version 7.1 24-Apr-07
768 ---------------------
769
770 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
771 that is more "standard", making use of automake and other Autotools. There
772 is some re-arrangement of the files and adjustment of comments consequent
773 on this.
774
775 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
776 for recursive directory scanning broke on some systems because the files
777 are not scanned in any specific order and on different systems the order
778 was different. A call to "sort" has been inserted into RunGrepTest for the
779 approprate test as a short-term fix. In the longer term there may be an
780 alternative.
781
782 3. I had an email from Eric Raymond about problems translating some of PCRE's
783 man pages to HTML (despite the fact that I distribute HTML pages, some
784 people do their own conversions for various reasons). The problems
785 concerned the use of low-level troff macros .br and .in. I have therefore
786 removed all such uses from the man pages (some were redundant, some could
787 be replaced by .nf/.fi pairs). The 132html script that I use to generate
788 HTML has been updated to handle .nf/.fi and to complain if it encounters
789 .br or .in.
790
791 4. Updated comments in configure.ac that get placed in config.h.in and also
792 arranged for config.h to be included in the distribution, with the name
793 config.h.generic, for the benefit of those who have to compile without
794 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
795
796 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
797 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
798 makevp.bat for the current PCRE, using the additional files
799 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
800
801 6. A Windows user reported a minor discrepancy with test 2, which turned out
802 to be caused by a trailing space on an input line that had got lost in his
803 copy. The trailing space was an accident, so I've just removed it.
804
805 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
806 that is needed.
807
808 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
809 as "const" (a) because they are and (b) because it helps the PHP
810 maintainers who have recently made a script to detect big data structures
811 in the php code that should be moved to the .rodata section. I remembered
812 to update Builducptable as well, so it won't revert if ucptable.h is ever
813 re-created.
814
815 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
816 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
817 order to be able to cut out the UTF-8 tables in the latter when UTF-8
818 support is not required. This saves 1.5-2K of code, which is important in
819 some applications.
820
821 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
822 so as not to refer to the tables, even though these functions will never be
823 called when UTF-8 support is disabled. Otherwise there are problems with a
824 shared library.
825
826 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
827
828 (a) It was defining its arguments as char * instead of void *.
829
830 (b) It was assuming that all moves were upwards in memory; this was true
831 a long time ago when I wrote it, but is no longer the case.
832
833 The emulated memove() is provided for those environments that have neither
834 memmove() nor bcopy(). I didn't think anyone used it these days, but that
835 is clearly not the case, as these two bugs were recently reported.
836
837 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
838 and Detrail to create the HTML documentation, the .txt form of the man
839 pages, and it removes trailing spaces from listed files. It also creates
840 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
841 case, it wraps all the #defines with #ifndefs. This script should be run
842 before "make dist".
843
844 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
845 with Unicode property support.
846
847 (a) For a maximizing quantifier, if the two different cases of the
848 character were of different lengths in their UTF-8 codings (there are
849 some cases like this - I found 11), and the matching function had to
850 back up over a mixture of the two cases, it incorrectly assumed they
851 were both the same length.
852
853 (b) When PCRE was configured to use the heap rather than the stack for
854 recursion during matching, it was not correctly preserving the data for
855 the other case of a UTF-8 character when checking ahead for a match
856 while processing a minimizing repeat. If the check also involved
857 matching a wide character, but failed, corruption could cause an
858 erroneous result when trying to check for a repeat of the original
859 character.
860
861 13. Some tidying changes to the testing mechanism:
862
863 (a) The RunTest script now detects the internal link size and whether there
864 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
865 values substituted by "configure". (The RunGrepTest script already did
866 this for UTF-8.) The configure.ac script no longer substitutes the
867 relevant variables.
868
869 (b) The debugging options /B and /D in pcretest show the compiled bytecode
870 with length and offset values. This means that the output is different
871 for different internal link sizes. Test 2 is skipped for link sizes
872 other than 2 because of this, bypassing the problem. Unfortunately,
873 there was also a test in test 3 (the locale tests) that used /B and
874 failed for link sizes other than 2. Rather than cut the whole test out,
875 I have added a new /Z option to pcretest that replaces the length and
876 offset values with spaces. This is now used to make test 3 independent
877 of link size. (Test 2 will be tidied up later.)
878
879 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
880 segmentation fault instead of returning the appropriate error message.
881
882 15. In multiline mode when the newline sequence was set to "any", the pattern
883 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
884 This doesn't seem right; it now treats the CRLF combination as the line
885 ending, and so does not match in that case. It's only a pattern such as ^$
886 that would hit this one: something like ^ABC$ would have failed after \r
887 and then tried again after \r\n.
888
889 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
890 in an attempt to make files that differ only in their line terminators
891 compare equal. This works on Linux.
892
893 17. Under certain error circumstances pcregrep might try to free random memory
894 as it exited. This is now fixed, thanks to valgrind.
895
896 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
897 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
898 was because its rules for how to advance for /g after matching an empty
899 string at the end of a line did not allow for this case. They now check for
900 it specially.
901
902 20. pcretest is supposed to handle patterns and data of any length, by
903 extending its buffers when necessary. It was getting this wrong when the
904 buffer for a data line had to be extended.
905
906 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
907 CRLF as a newline sequence.
908
909 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
910 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
911 I have nevertheless tidied it up.
912
913 23. Added some casts to kill warnings from HP-UX ia64 compiler.
914
915 24. Added a man page for pcre-config.
916
917
918 Version 7.0 19-Dec-06
919 ---------------------
920
921 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
922 moving to gcc 4.1.1.
923
924 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
925 sys/time.h, which is documented as needed for this function. It doesn't
926 seem to matter on Linux, but it showed up on some releases of OS X.
927
928 3. It seems that there are systems where bytes whose values are greater than
929 127 match isprint() in the "C" locale. The "C" locale should be the
930 default when a C program starts up. In most systems, only ASCII printing
931 characters match isprint(). This difference caused the output from pcretest
932 to vary, making some of the tests fail. I have changed pcretest so that:
933
934 (a) When it is outputting text in the compiled version of a pattern, bytes
935 other than 32-126 are always shown as hex escapes.
936
937 (b) When it is outputting text that is a matched part of a subject string,
938 it does the same, unless a different locale has been set for the match
939 (using the /L modifier). In this case, it uses isprint() to decide.
940
941 4. Fixed a major bug that caused incorrect computation of the amount of memory
942 required for a compiled pattern when options that changed within the
943 pattern affected the logic of the preliminary scan that determines the
944 length. The relevant options are -x, and -i in UTF-8 mode. The result was
945 that the computed length was too small. The symptoms of this bug were
946 either the PCRE error "internal error: code overflow" from pcre_compile(),
947 or a glibc crash with a message such as "pcretest: free(): invalid next
948 size (fast)". Examples of patterns that provoked this bug (shown in
949 pcretest format) are:
950
951 /(?-x: )/x
952 /(?x)(?-x: \s*#\s*)/
953 /((?i)[\x{c0}])/8
954 /(?i:[\x{c0}])/8
955
956 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
957 is now done differently.
958
959 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
960 wrapper classes; (b) implement a new function in the C++ scanner that is
961 more efficient than the old way of doing things because it avoids levels of
962 recursion in the regex matching; (c) add a paragraph to the documentation
963 for the FullMatch() function.
964
965 6. The escape sequence \n was being treated as whatever was defined as
966 "newline". Not only was this contrary to the documentation, which states
967 that \n is character 10 (hex 0A), but it also went horribly wrong when
968 "newline" was defined as CRLF. This has been fixed.
969
970 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
971 was being set to -1 for the "end of line" case (supposedly a value that no
972 character can have). Though this value is never used (the check for end of
973 line is "zero bytes in current character"), it caused compiler complaints.
974 I've changed it to 0xffffffff.
975
976 8. In pcre_version.c, the version string was being built by a sequence of
977 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
978 string (as it is for production releases) called a macro with an empty
979 argument. The C standard says the result of this is undefined. The gcc
980 compiler treats it as an empty string (which was what was wanted) but it is
981 reported that Visual C gives an error. The source has been hacked around to
982 avoid this problem.
983
984 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
985 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
986 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
987 of them did).
988
989 10. Originally, pcretest opened its input and output without "b"; then I was
990 told that "b" was needed in some environments, so it was added for release
991 5.0 to both the input and output. (It makes no difference on Unix-like
992 systems.) Later I was told that it is wrong for the input on Windows. I've
993 now abstracted the modes into two macros, to make it easier to fiddle with
994 them, and removed "b" from the input mode under Windows.
995
996 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
997
998 12. Added -help and --help to pcretest as an official way of being reminded
999 of the options.
1000
1001 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1002 and pcrecpp.cc because they annoy compilers at high warning levels.
1003
1004 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1005
1006 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1007 scripts are not C programs :-) and which was not noticed because it works
1008 on Linux.
1009
1010 16. pcretest is supposed to handle any length of pattern and data line (as one
1011 line or as a continued sequence of lines) by extending its input buffer if
1012 necessary. This feature was broken for very long pattern lines, leading to
1013 a string of junk being passed to pcre_compile() if the pattern was longer
1014 than about 50K.
1015
1016 17. I have done a major re-factoring of the way pcre_compile() computes the
1017 amount of memory needed for a compiled pattern. Previously, there was code
1018 that made a preliminary scan of the pattern in order to do this. That was
1019 OK when PCRE was new, but as the facilities have expanded, it has become
1020 harder and harder to keep it in step with the real compile phase, and there
1021 have been a number of bugs (see for example, 4 above). I have now found a
1022 cunning way of running the real compile function in a "fake" mode that
1023 enables it to compute how much memory it would need, while actually only
1024 ever using a few hundred bytes of working memory and without too many
1025 tests of the mode. This should make future maintenance and development
1026 easier. A side effect of this work is that the limit of 200 on the nesting
1027 depth of parentheses has been removed (though this was never a serious
1028 limitation, I suspect). However, there is a downside: pcre_compile() now
1029 runs more slowly than before (30% or more, depending on the pattern). I
1030 hope this isn't a big issue. There is no effect on runtime performance.
1031
1032 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1033 newline (only possible for the last line of a file) and it was a
1034 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1035
1036 19. Added additional timing features to pcretest. (1) The -tm option now times
1037 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1038 separate command line item, by a number that specifies the number of
1039 repeats to use when timing. The default is 50000; this gives better
1040 precision, but takes uncomfortably long for very large patterns.
1041
1042 20. Extended pcre_study() to be more clever in cases where a branch of a
1043 subpattern has no definite first character. For example, (a*|b*)[cd] would
1044 previously give no result from pcre_study(). Now it recognizes that the
1045 first character must be a, b, c, or d.
1046
1047 21. There was an incorrect error "recursive call could loop indefinitely" if
1048 a subpattern (or the entire pattern) that was being tested for matching an
1049 empty string contained only one non-empty item after a nested subpattern.
1050 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1051 incorrectly, because the \d was being skipped in the check.
1052
1053 22. The pcretest program now has a new pattern option /B and a command line
1054 option -b, which is equivalent to adding /B to every pattern. This causes
1055 it to show the compiled bytecode, without the additional information that
1056 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1057 is the same as /B/I).
1058
1059 23. A new optimization is now able automatically to treat some sequences such
1060 as a*b as a*+b. More specifically, if something simple (such as a character
1061 or a simple class like \d) has an unlimited quantifier, and is followed by
1062 something that cannot possibly match the quantified thing, the quantifier
1063 is automatically "possessified".
1064
1065 24. A recursive reference to a subpattern whose number was greater than 39
1066 went wrong under certain circumstances in UTF-8 mode. This bug could also
1067 have affected the operation of pcre_study().
1068
1069 25. Realized that a little bit of performance could be had by replacing
1070 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1071
1072 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1073
1074 27. Possessive quantifiers such as a++ were previously implemented by turning
1075 them into atomic groups such as ($>a+). Now they have their own opcodes,
1076 which improves performance. This includes the automatically created ones
1077 from 23 above.
1078
1079 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1080 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1081 the first matched character to be a colon. This applied both to named and
1082 numbered groups.
1083
1084 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1085
1086 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1087 building PCRE on Windows easier, so I have included it in the distribution.
1088
1089 31. There is now a check in pcretest against a ridiculously large number being
1090 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1091 loop, the loop is abandoned.
1092
1093 32. Forward references to subpatterns in conditions such as (?(2)...) where
1094 subpattern 2 is defined later cause pcre_compile() to search forwards in
1095 the pattern for the relevant set of parentheses. This search went wrong
1096 when there were unescaped parentheses in a character class, parentheses
1097 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1098
1099 33. "Subroutine" calls and backreferences were previously restricted to
1100 referencing subpatterns earlier in the regex. This restriction has now
1101 been removed.
1102
1103 34. Added a number of extra features that are going to be in Perl 5.10. On the
1104 whole, these are just syntactic alternatives for features that PCRE had
1105 previously implemented using the Python syntax or my own invention. The
1106 other formats are all retained for compatibility.
1107
1108 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1109 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1110 also .NET compatible.
1111
1112 (b) A recursion or subroutine call to a named group can now be defined as
1113 (?&name) as well as (?P>name).
1114
1115 (c) A backreference to a named group can now be defined as \k<name> or
1116 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1117 5.10, are also .NET compatible.
1118
1119 (d) A conditional reference to a named group can now use the syntax
1120 (?(<name>) or (?('name') as well as (?(name).
1121
1122 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1123 groups (named and numbered) that are never evaluated inline, but can be
1124 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1125 is always false. There may be only one alternative in such a group.
1126
1127 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1128 as the simple (?(R). The condition is true only if the most recent
1129 recursion is that of the given number or name. It does not search out
1130 through the entire recursion stack.
1131
1132 (g) The escape \gN or \g{N} has been added, where N is a positive or
1133 negative number, specifying an absolute or relative reference.
1134
1135 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1136 some "unreachable code" warnings.
1137
1138 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1139 things, this adds five new scripts.
1140
1141 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1142 There were also incompatibilities regarding the handling of \Q..\E inside
1143 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1144 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1145
1146 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1147 matches an empty string, and forcibly breaks the loop. There were bugs in
1148 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1149 against aaaa the result was just "a" rather than "aaaa", for example. Two
1150 separate and independent bugs (that affected different cases) have been
1151 fixed.
1152
1153 39. Refactored the code to abolish the use of different opcodes for small
1154 capturing bracket numbers. This is a tidy that I avoided doing when I
1155 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1156 The new approach is not only tidier, it makes it possible to reduce the
1157 memory needed to fix the previous bug (38).
1158
1159 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1160 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1161 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1162 mode.
1163
1164 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1165 report.
1166
1167 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1168 copy construction and assignment in the C++ wrapper.
1169
1170 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1171 couple of bugs that could have given wrong results in the "--newline=crlf"
1172 case.
1173
1174 44. Added a number of casts and did some reorganization of signed/unsigned int
1175 variables following suggestions from Dair Grant. Also renamed the variable
1176 "this" as "item" because it is a C++ keyword.
1177
1178 45. Arranged for dftables to add
1179
1180 #include "pcre_internal.h"
1181
1182 to pcre_chartables.c because without it, gcc 4.x may remove the array
1183 definition from the final binary if PCRE is built into a static library and
1184 dead code stripping is activated.
1185
1186 46. For an unanchored pattern, if a match attempt fails at the start of a
1187 newline sequence, and the newline setting is CRLF or ANY, and the next two
1188 characters are CRLF, advance by two characters instead of one.
1189
1190
1191 Version 6.7 04-Jul-06
1192 ---------------------
1193
1194 1. In order to handle tests when input lines are enormously long, pcretest has
1195 been re-factored so that it automatically extends its buffers when
1196 necessary. The code is crude, but this _is_ just a test program. The
1197 default size has been increased from 32K to 50K.
1198
1199 2. The code in pcre_study() was using the value of the re argument before
1200 testing it for NULL. (Of course, in any sensible call of the function, it
1201 won't be NULL.)
1202
1203 3. The memmove() emulation function in pcre_internal.h, which is used on
1204 systems that lack both memmove() and bcopy() - that is, hardly ever -
1205 was missing a "static" storage class specifier.
1206
1207 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1208 containing an extended class (one that cannot be represented by a bitmap
1209 because it contains high-valued characters or Unicode property items, e.g.
1210 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1211 pattern, but PCRE should not loop if you do not (it no longer does).
1212 [Detail: two cases were found: (a) a repeated subpattern containing an
1213 extended class; (b) a recursive reference to a subpattern that followed a
1214 previous extended class. It wasn't skipping over the extended class
1215 correctly when UTF-8 mode was not set.]
1216
1217 5. A negated single-character class was not being recognized as fixed-length
1218 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1219 compile error "lookbehind assertion is not fixed length".
1220
1221 6. The RunPerlTest auxiliary script was showing an unexpected difference
1222 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1223 write a Perl script that can interpret lines of an input file either as
1224 byte characters or as UTF-8, which is what "perltest" was being required to
1225 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1226 can't do is switch easily at run time between having the "use utf8;" pragma
1227 or not. In the end, I fudged it by using the RunPerlTest script to insert
1228 "use utf8;" explicitly for the UTF-8 tests.
1229
1230 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1231 the end of the subject string, contrary to the documentation and to what
1232 Perl does. This was true of both matching functions. Now it matches only at
1233 the start of the subject and immediately after *internal* newlines.
1234
1235 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1236 a pointer to an int instead of a pointer to an unsigned long int. This
1237 caused problems on 64-bit systems.
1238
1239 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1240 instance of the 'standard' template library not being so standard".
1241
1242 10. There was no check on the number of named subpatterns nor the maximum
1243 length of a subpattern name. The product of these values is used to compute
1244 the size of the memory block for a compiled pattern. By supplying a very
1245 long subpattern name and a large number of named subpatterns, the size
1246 computation could be caused to overflow. This is now prevented by limiting
1247 the length of names to 32 characters, and the number of named subpatterns
1248 to 10,000.
1249
1250 11. Subpatterns that are repeated with specific counts have to be replicated in
1251 the compiled pattern. The size of memory for this was computed from the
1252 length of the subpattern and the repeat count. The latter is limited to
1253 65535, but there was no limit on the former, meaning that integer overflow
1254 could in principle occur. The compiled length of a repeated subpattern is
1255 now limited to 30,000 bytes in order to prevent this.
1256
1257 12. Added the optional facility to have named substrings with the same name.
1258
1259 13. Added the ability to use a named substring as a condition, using the
1260 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1261 are numbers (not recommended). Forward references are permitted.
1262
1263 14. Added forward references in named backreferences (if you see what I mean).
1264
1265 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1266 pattern could run off the end of the subject. For example, the pattern
1267 "(?s)(.{1,5})"8 did this with the subject "ab".
1268
1269 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1270 PCRE_CASELESS was set when matching characters that were quantified with ?
1271 or *.
1272
1273 17. A character class other than a single negated character that had a minimum
1274 but no maximum quantifier - for example [ab]{6,} - was not handled
1275 correctly by pce_dfa_exec(). It would match only one character.
1276
1277 18. A valid (though odd) pattern that looked like a POSIX character
1278 class but used an invalid character after [ (for example [[,abc,]]) caused
1279 pcre_compile() to give the error "Failed: internal error: code overflow" or
1280 in some cases to crash with a glibc free() error. This could even happen if
1281 the pattern terminated after [[ but there just happened to be a sequence of
1282 letters, a binary zero, and a closing ] in the memory that followed.
1283
1284 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1285 over the years. Originally (before any Unicode support), just the bottom 8
1286 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1287 output from "man perlunicode" includes this:
1288
1289 The regular expression compiler produces polymorphic opcodes. That
1290 is, the pattern adapts to the data and automatically switches to
1291 the Unicode character scheme when presented with Unicode data--or
1292 instead uses a traditional byte scheme when presented with byte
1293 data.
1294
1295 Sadly, a wide octal escape does not cause a switch, and in a string with
1296 no other multibyte characters, these octal escapes are treated as before.
1297 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1298 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1299 Unicode string.
1300
1301 I have not perpetrated such confusion in PCRE. Up till now, it took just
1302 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1303 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1304 translate to the appropriate multibyte character.
1305
1306 29. Applied some refactoring to reduce the number of warnings from Microsoft
1307 and Borland compilers. This has included removing the fudge introduced
1308 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1309 a warning about an unused variable.
1310
1311 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1312 characters since release 4.0, because Perl (from release 5.004) does not.
1313 [Or at least, is documented not to: some releases seem to be in conflict
1314 with the documentation.] However, when a pattern was studied with
1315 pcre_study() and all its branches started with \s, PCRE still included VT
1316 as a possible starting character. Of course, this did no harm; it just
1317 caused an unnecessary match attempt.
1318
1319 22. Removed a now-redundant internal flag bit that recorded the fact that case
1320 dependency changed within the pattern. This was once needed for "required
1321 byte" processing, but is no longer used. This recovers a now-scarce options
1322 bit. Also moved the least significant internal flag bit to the most-
1323 significant bit of the word, which was not previously used (hangover from
1324 the days when it was an int rather than a uint) to free up another bit for
1325 the future.
1326
1327 23. Added support for CRLF line endings as well as CR and LF. As well as the
1328 default being selectable at build time, it can now be changed at runtime
1329 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1330 specify that it is scanning data with non-default line endings.
1331
1332 24. Changed the definition of CXXLINK to make it agree with the definition of
1333 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1334
1335 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1336 recursions. This makes a big different to stack usage for some patterns.
1337
1338 26. If a subpattern containing a named recursion or subroutine reference such
1339 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1340 the space required for the compiled pattern went wrong and gave too small a
1341 value. Depending on the environment, this could lead to "Failed: internal
1342 error: code overflow at offset 49" or "glibc detected double free or
1343 corruption" errors.
1344
1345 27. Applied patches from Google (a) to support the new newline modes and (b) to
1346 advance over multibyte UTF-8 characters in GlobalReplace.
1347
1348 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1349 difference for some implementation of PCRE in some Windows version.
1350
1351 29. Added some extra testing facilities to pcretest:
1352
1353 \q<number> in a data line sets the "match limit" value
1354 \Q<number> in a data line sets the "match recursion limt" value
1355 -S <number> sets the stack size, where <number> is in megabytes
1356
1357 The -S option isn't available for Windows.
1358
1359
1360 Version 6.6 06-Feb-06
1361 ---------------------
1362
1363 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1364 in pcreposix.h. I have copied the definition from pcre.h.
1365
1366 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1367 because pcre.h is no longer a built file.
1368
1369 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1370 not normally included in the compiled code.
1371
1372
1373 Version 6.5 01-Feb-06
1374 ---------------------
1375
1376 1. When using the partial match feature with pcre_dfa_exec(), it was not
1377 anchoring the second and subsequent partial matches at the new starting
1378 point. This could lead to incorrect results. For example, with the pattern
1379 /1234/, partially matching against "123" and then "a4" gave a match.
1380
1381 2. Changes to pcregrep:
1382
1383 (a) All non-match returns from pcre_exec() were being treated as failures
1384 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1385 error message is output. Some extra information is given for the
1386 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1387 probably the only errors that are likely to be caused by users (by
1388 specifying a regex that has nested indefinite repeats, for instance).
1389 If there are more than 20 of these errors, pcregrep is abandoned.
1390
1391 (b) A binary zero was treated as data while matching, but terminated the
1392 output line if it was written out. This has been fixed: binary zeroes
1393 are now no different to any other data bytes.
1394
1395 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1396 used to set a locale for matching. The --locale=xxxx long option has
1397 been added (no short equivalent) to specify a locale explicitly on the
1398 pcregrep command, overriding the environment variables.
1399
1400 (d) When -B was used with -n, some line numbers in the output were one less
1401 than they should have been.
1402
1403 (e) Added the -o (--only-matching) option.
1404
1405 (f) If -A or -C was used with -c (count only), some lines of context were
1406 accidentally printed for the final match.
1407
1408 (g) Added the -H (--with-filename) option.
1409
1410 (h) The combination of options -rh failed to suppress file names for files
1411 that were found from directory arguments.
1412
1413 (i) Added the -D (--devices) and -d (--directories) options.
1414
1415 (j) Added the -F (--fixed-strings) option.
1416
1417 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1418
1419 (l) Added the --colo(u)r option.
1420
1421 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1422 is not present by default.
1423
1424 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1425 items such as (?R) or (?1), when the recursion could match a number of
1426 alternatives. If it matched one of the alternatives, but subsequently,
1427 outside the recursion, there was a failure, the code tried to back up into
1428 the recursion. However, because of the way PCRE is implemented, this is not
1429 possible, and the result was an incorrect result from the match.
1430
1431 In order to prevent this happening, the specification of recursion has
1432 been changed so that all such subpatterns are automatically treated as
1433 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1434
1435 4. I had overlooked the fact that, in some locales, there are characters for
1436 which isalpha() is true but neither isupper() nor islower() are true. In
1437 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1438 and ordfeminine) are like this. This affected the treatment of \w and \W
1439 when they appeared in character classes, but not when they appeared outside
1440 a character class. The bit map for "word" characters is now created
1441 separately from the results of isalnum() instead of just taking it from the
1442 upper, lower, and digit maps. (Plus the underscore character, of course.)
1443
1444 5. The above bug also affected the handling of POSIX character classes such as
1445 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1446 permanent tables. Instead, the bit maps for such a class were previously
1447 created as the appropriate unions of the upper, lower, and digit bitmaps.
1448 Now they are created by subtraction from the [[:word:]] class, which has
1449 its own bitmap.
1450
1451 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1452 It is created by subtracting the vertical space characters (\x09, \x0a,
1453 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1454 subtraction was done in the overall bitmap for a character class, meaning
1455 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1456 be recognized. This bug has been fixed.
1457
1458 7. Patches from the folks at Google:
1459
1460 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1461 real life, but is still worth protecting against".
1462
1463 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1464 regular expressions".
1465
1466 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1467 have it.
1468
1469 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1470 "configure" and the latter not, in order to fix a problem somebody had
1471 with compiling the Arg class on HP-UX.
1472
1473 (e) Improve the error-handling of the C++ wrapper a little bit.
1474
1475 (f) New tests for checking recursion limiting.
1476
1477 8. The pcre_memmove() function, which is used only if the environment does not
1478 have a standard memmove() function (and is therefore rarely compiled),
1479 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1480 returning a result (though PCRE never actually uses the result).
1481
1482 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1483 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1484 returned instead of calling malloc() with an overflowing number that would
1485 most likely cause subsequent chaos.
1486
1487 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1488
1489 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1490 with this option is matched, the nmatch and pmatch options of regexec() are
1491 ignored.
1492
1493 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1494 provided in case anyone wants to the the POSIX interface with UTF-8
1495 strings.
1496
1497 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1498 C++ linking (needed for some HP-UX environments).
1499
1500 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1501 (unused parameter) and in the pcre_printint() function (omitted "default"
1502 switch label when the default is to do nothing).
1503
1504 15. Added some code to make it possible, when PCRE is compiled as a C++
1505 library, to replace subject pointers for pcre_exec() with a smart pointer
1506 class, thus making it possible to process discontinuous strings.
1507
1508 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1509 much the same function. They were added by different people who were trying
1510 to make PCRE easy to compile on non-Unix systems. It has been suggested
1511 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1512 for compiling on Windows systems. I have therefore replaced it with
1513 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1514 defaults to "extern" for C or "extern C" for C++, which works fine on
1515 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1516 SCOPE with something explicit in config.h. In addition:
1517
1518 (a) pcreposix.h still had just "extern" instead of either of these macros;
1519 I have replaced it with PCRE_DATA_SCOPE.
1520
1521 (b) Functions such as _pcre_xclass(), which are internal to the library,
1522 but external in the C sense, all had PCRE_EXPORT in their definitions.
1523 This is apparently wrong for the Windows case, so I have removed it.
1524 (It makes no difference on Unix-like systems.)
1525
1526 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1527 of recursive calls to match(). This is different to MATCH_LIMIT because
1528 that limits the total number of calls to match(), not all of which increase
1529 the depth of recursion. Limiting the recursion depth limits the amount of
1530 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1531 when PCRE is compiled, and changed at run time. A patch from Google adds
1532 this functionality to the C++ interface.
1533
1534 18. Changes to the handling of Unicode character properties:
1535
1536 (a) Updated the table to Unicode 4.1.0.
1537
1538 (b) Recognize characters that are not in the table as "Cn" (undefined).
1539
1540 (c) I revised the way the table is implemented to a much improved format
1541 which includes recognition of ranges. It now supports the ranges that
1542 are defined in UnicodeData.txt, and it also amalgamates other
1543 characters into ranges. This has reduced the number of entries in the
1544 table from around 16,000 to around 3,000, thus reducing its size
1545 considerably. I realized I did not need to use a tree structure after
1546 all - a binary chop search is just as efficient. Having reduced the
1547 number of entries, I extended their size from 6 bytes to 8 bytes to
1548 allow for more data.
1549
1550 (d) Added support for Unicode script names via properties such as \p{Han}.
1551
1552 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1553 matching that character.
1554
1555 20. When matching a repeated Unicode property with a minimum greater than zero,
1556 (for example \pL{2,}), PCRE could look past the end of the subject if it
1557 reached it while seeking the minimum number of characters. This could
1558 happen only if some of the characters were more than one byte long, because
1559 there is a check for at least the minimum number of bytes.
1560
1561 21. Refactored the implementation of \p and \P so as to be more general, to
1562 allow for more different types of property in future. This has changed the
1563 compiled form incompatibly. Anybody with saved compiled patterns that use
1564 \p or \P will have to recompile them.
1565
1566 22. Added "Any" and "L&" to the supported property types.
1567
1568 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1569 but give a compile time error if the value is greater than 0xff.
1570
1571 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1572 accidentally not being installed or uninstalled.
1573
1574 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1575 made were to insert the current release number. This seemed silly, because
1576 it made things harder for people building PCRE on systems that don't run
1577 "configure". I have turned pcre.h into a distributed file, no longer built
1578 by "configure", with the version identification directly included. There is
1579 no longer a pcre.h.in file.
1580
1581 However, this change necessitated a change to the pcre-config script as
1582 well. It is built from pcre-config.in, and one of the substitutions was the
1583 release number. I have updated configure.ac so that ./configure now finds
1584 the release number by grepping pcre.h.
1585
1586 26. Added the ability to run the tests under valgrind.
1587
1588
1589 Version 6.4 05-Sep-05
1590 ---------------------
1591
1592 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1593 "--" to be printed when multiple files were scanned, even when none of the
1594 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1595 consider it to be a bug, and have restored the previous behaviour.
1596
1597 2. A couple of code tidies to get rid of compiler warnings.
1598
1599 3. The pcretest program used to cheat by referring to symbols in the library
1600 whose names begin with _pcre_. These are internal symbols that are not
1601 really supposed to be visible externally, and in some environments it is
1602 possible to suppress them. The cheating is now confined to including
1603 certain files from the library's source, which is a bit cleaner.
1604
1605 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1606 file's purpose clearer.
1607
1608 5. Reorganized pcre_ucp_findchar().
1609
1610
1611 Version 6.3 15-Aug-05
1612 ---------------------
1613
1614 1. The file libpcre.pc.in did not have general read permission in the tarball.
1615
1616 2. There were some problems when building without C++ support:
1617
1618 (a) If C++ support was not built, "make install" and "make test" still
1619 tried to test it.
1620
1621 (b) There were problems when the value of CXX was explicitly set. Some
1622 changes have been made to try to fix these, and ...
1623
1624 (c) --disable-cpp can now be used to explicitly disable C++ support.
1625
1626 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1627 backslash in a target when C++ was disabled. This confuses some
1628 versions of "make", apparently. Using an intermediate variable solves
1629 this. (Same for CPP_LOBJ.)
1630
1631 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1632 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1633 necessary on certain architectures.
1634
1635 4. Added a setting of -export-symbols-regex to the link command to remove
1636 those symbols that are exported in the C sense, but actually are local
1637 within the library, and not documented. Their names all begin with
1638 "_pcre_". This is not a perfect job, because (a) we have to except some
1639 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1640 available (and never for static libraries). I have made a note to try to
1641 find a way round (a) in the future.
1642
1643
1644 Version 6.2 01-Aug-05
1645 ---------------------
1646
1647 1. There was no test for integer overflow of quantifier values. A construction
1648 such as {1111111111111111} would give undefined results. What is worse, if
1649 a minimum quantifier for a parenthesized subpattern overflowed and became
1650 negative, the calculation of the memory size went wrong. This could have
1651 led to memory overwriting.
1652
1653 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1654
1655 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1656 operating environments where this matters.
1657
1658 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1659 PCRE options from within the C++ wrapper.
1660
1661 5. Named capturing subpatterns were not being correctly counted when a pattern
1662 was compiled. This caused two problems: (a) If there were more than 100
1663 such subpatterns, the calculation of the memory needed for the whole
1664 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1665 back references of the form \12, where the number was greater than 9, were
1666 not recognized as back references, even though there were sufficient
1667 previous subpatterns.
1668
1669 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1670 versions of gcc, e.g. 2.95.4.
1671
1672
1673 Version 6.1 21-Jun-05
1674 ---------------------
1675
1676 1. There was one reference to the variable "posix" in pcretest.c that was not
1677 surrounded by "#if !defined NOPOSIX".
1678
1679 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1680 the cross-check on the old pcre_info() function, for the benefit of the
1681 cut-down version of PCRE that is currently imported into Exim.
1682
1683 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1684 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1685 patterns that start (?i) but I don't think that matters. The use of (?i) is
1686 just an example; this all applies to the other options as well.
1687
1688 4. Since libtool seems to echo the compile commands it is issuing, the output
1689 from "make" can be reduced a bit by putting "@" in front of each libtool
1690 compile command.
1691
1692 5. Patch from the folks at Google for configure.in to be a bit more thorough
1693 in checking for a suitable C++ installation before trying to compile the
1694 C++ stuff. This should fix a reported problem when a compiler was present,
1695 but no suitable headers.
1696
1697 6. The man pages all had just "PCRE" as their title. I have changed them to
1698 be the relevant file name. I have also arranged that these names are
1699 retained in the file doc/pcre.txt, which is a concatenation in text format
1700 of all the man pages except the little individual ones for each function.
1701
1702 7. The NON-UNIX-USE file had not been updated for the different set of source
1703 files that come with release 6. I also added a few comments about the C++
1704 wrapper.
1705
1706
1707 Version 6.0 07-Jun-05
1708 ---------------------
1709
1710 1. Some minor internal re-organization to help with my DFA experiments.
1711
1712 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1713 didn't matter for the library itself when fully configured, but did matter
1714 when compiling without UCP support, or within Exim, where the ucp files are
1715 not imported.
1716
1717 3. Refactoring of the library code to split up the various functions into
1718 different source modules. The addition of the new DFA matching code (see
1719 below) to a single monolithic source would have made it really too
1720 unwieldy, quite apart from causing all the code to be include in a
1721 statically linked application, when only some functions are used. This is
1722 relevant even without the DFA addition now that patterns can be compiled in
1723 one application and matched in another.
1724
1725 The downside of splitting up is that there have to be some external
1726 functions and data tables that are used internally in different modules of
1727 the library but which are not part of the API. These have all had their
1728 names changed to start with "_pcre_" so that they are unlikely to clash
1729 with other external names.
1730
1731 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1732 a different (DFA) algorithm. Although it is slower than the original
1733 function, it does have some advantages for certain types of matching
1734 problem.
1735
1736 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1737 including restarting after a partial match.
1738
1739 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1740 defined when compiling for Windows was sent to me. I have put it into the
1741 code, though I have no means of testing or verifying it.
1742
1743 7. Added the pcre_refcount() auxiliary function.
1744
1745 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1746 match before or at the first newline in the subject string. In pcretest,
1747 the /f option on a pattern can be used to set this.
1748
1749 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1750 would behave wrongly. This has been present in PCRE since release 4.0.
1751
1752 10. A number of changes to the pcregrep command:
1753
1754 (a) Refactored how -x works; insert ^(...)$ instead of setting
1755 PCRE_ANCHORED and checking the length, in preparation for adding
1756 something similar for -w.
1757
1758 (b) Added the -w (match as a word) option.
1759
1760 (c) Refactored the way lines are read and buffered so as to have more
1761 than one at a time available.
1762
1763 (d) Implemented a pcregrep test script.
1764
1765 (e) Added the -M (multiline match) option. This allows patterns to match
1766 over several lines of the subject. The buffering ensures that at least
1767 8K, or the rest of the document (whichever is the shorter) is available
1768 for matching (and similarly the previous 8K for lookbehind assertions).
1769
1770 (f) Changed the --help output so that it now says
1771
1772 -w, --word-regex(p)
1773
1774 instead of two lines, one with "regex" and the other with "regexp"
1775 because that confused at least one person since the short forms are the
1776 same. (This required a bit of code, as the output is generated
1777 automatically from a table. It wasn't just a text change.)
1778
1779 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1780 option but starts with a hyphen. Could be a pattern or a path name
1781 starting with a hyphen, for instance.
1782
1783 (h) "-" can be given as a file name to represent stdin.
1784
1785 (i) When file names are being printed, "(standard input)" is used for
1786 the standard input, for compatibility with GNU grep. Previously
1787 "<stdin>" was used.
1788
1789 (j) The option --label=xxx can be used to supply a name to be used for
1790 stdin when file names are being printed. There is no short form.
1791
1792 (k) Re-factored the options decoding logic because we are going to add
1793 two more options that take data. Such options can now be given in four
1794 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1795
1796 (l) Added the -A, -B, and -C options for requesting that lines of context
1797 around matches be printed.
1798
1799 (m) Added the -L option to print the names of files that do not contain
1800 any matching lines, that is, the complement of -l.
1801
1802 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1803 continue to scan other files.
1804
1805 (o) The -s option was incorrectly implemented. For compatibility with other
1806 greps, it now suppresses the error message for a non-existent or non-
1807 accessible file (but not the return code). There is a new option called
1808 -q that suppresses the output of matching lines, which was what -s was
1809 previously doing.
1810
1811 (p) Added --include and --exclude options to specify files for inclusion
1812 and exclusion when recursing.
1813
1814 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1815 Hopefully, it now does.
1816
1817 12. Missing cast in pcre_study().
1818
1819 13. Added an "uninstall" target to the makefile.
1820
1821 14. Replaced "extern" in the function prototypes in Makefile.in with
1822 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1823 world, but is set differently for Windows.
1824
1825 15. Added a second compiling function called pcre_compile2(). The only
1826 difference is that it has an extra argument, which is a pointer to an
1827 integer error code. When there is a compile-time failure, this is set
1828 non-zero, in addition to the error test pointer being set to point to an
1829 error message. The new argument may be NULL if no error number is required
1830 (but then you may as well call pcre_compile(), which is now just a
1831 wrapper). This facility is provided because some applications need a
1832 numeric error indication, but it has also enabled me to tidy up the way
1833 compile-time errors are handled in the POSIX wrapper.
1834
1835 16. Added VPATH=.libs to the makefile; this should help when building with one
1836 prefix path and installing with another. (Or so I'm told by someone who
1837 knows more about this stuff than I do.)
1838
1839 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1840 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1841 match everything, including newlines. This is not POSIX-compatible, but
1842 somebody wanted the feature. From pcretest it can be activated by using
1843 both the P and the s flags.
1844
1845 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1846
1847 19. libpcre.pc was being incorrectly installed as executable.
1848
1849 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1850 it now also looks for '\r' so that it will work unmodified on Windows.
1851
1852 21. Added Google's contributed C++ wrapper to the distribution.
1853
1854 22. Added some untidy missing memory free() calls in pcretest, to keep
1855 Electric Fence happy when testing.
1856
1857
1858
1859 Version 5.0 13-Sep-04
1860 ---------------------
1861
1862 1. Internal change: literal characters are no longer packed up into items
1863 containing multiple characters in a single byte-string. Each character
1864 is now matched using a separate opcode. However, there may be more than one
1865 byte in the character in UTF-8 mode.
1866
1867 2. The pcre_callout_block structure has two new fields: pattern_position and
1868 next_item_length. These contain the offset in the pattern to the next match
1869 item, and its length, respectively.
1870
1871 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1872 insertion of callouts before each pattern item. Added the /C option to
1873 pcretest to make use of this.
1874
1875 4. On the advice of a Windows user, the lines
1876
1877 #if defined(_WIN32) || defined(WIN32)
1878 _setmode( _fileno( stdout ), 0x8000 );
1879 #endif /* defined(_WIN32) || defined(WIN32) */
1880
1881 have been added to the source of pcretest. This apparently does useful
1882 magic in relation to line terminators.
1883
1884 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1885 for the benefit of those environments where the "b" makes a difference.
1886
1887 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1888 to know about it. I have put a hack into configure.in that adds in code
1889 to set GCC=yes if CC=icc. This seems to end up at a point in the
1890 generated configure script that is early enough to affect the setting of
1891 compiler options, which is what is needed, but I have no means of testing
1892 whether it really works. (The user who reported this had patched the
1893 generated configure script, which of course I cannot do.)
1894
1895 LATER: After change 22 below (new libtool files), the configure script
1896 seems to know about icc (and also ecc). Therefore, I have commented out
1897 this hack in configure.in.
1898
1899 7. Added support for pkg-config (2 patches were sent in).
1900
1901 8. Negated POSIX character classes that used a combination of internal tables
1902 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1903 [[:^ascii]]. Typically, they would match almost any characters. The other
1904 POSIX classes were not broken in this way.
1905
1906 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1907 to find the match, as PCRE was deluded into thinking that the match had to
1908 start at the start point or following a newline. The same bug applied to
1909 patterns with negative forward assertions or any backward assertions
1910 preceding ".*" at the start, unless the pattern required a fixed first
1911 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1912
1913 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1914 starting at the last subject character, bytes beyond the end of the subject
1915 string were read.
1916
1917 11. Renamed the variable "class" as "classbits" to make life easier for C++
1918 users. (Previously there was a macro definition, but it apparently wasn't
1919 enough.)
1920
1921 12. Added the new field "tables" to the extra data so that tables can be passed
1922 in at exec time, or the internal tables can be re-selected. This allows
1923 a compiled regex to be saved and re-used at a later time by a different
1924 program that might have everything at different addresses.
1925
1926 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1927 -R library as well as a -L library.
1928
1929 14. The debugging options of pcretest (-d on the command line or D on a
1930 pattern) showed incorrect output for anything following an extended class
1931 that contained multibyte characters and which was followed by a quantifier.
1932
1933 15. Added optional support for general category Unicode character properties
1934 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1935 support. It adds about 90K to the size of the library. The meanings of the
1936 inbuilt class escapes such as \d and \s have NOT been changed.
1937
1938 16. Updated pcredemo.c to include calls to free() to release the memory for the
1939 compiled pattern.
1940
1941 17. The generated file chartables.c was being created in the source directory
1942 instead of in the building directory. This caused the build to fail if the
1943 source directory was different from the building directory, and was
1944 read-only.
1945
1946 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1947 file. No doubt somebody will tell me if they don't make sense... Also added
1948 Dan Mooney's comments about building on OpenVMS.
1949
1950 19. Added support for partial matching via the PCRE_PARTIAL option for
1951 pcre_exec() and the \P data escape in pcretest.
1952
1953 20. Extended pcretest with 3 new pattern features:
1954
1955 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1956 write the compiled pattern to the file whose name is "rest-of-line".
1957 This is a straight binary dump of the data, with the saved pointer to
1958 the character tables forced to be NULL. The study data, if any, is
1959 written too. After writing, pcretest reads a new pattern.
1960
1961 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1962 compiled pattern from the given file. There must not be any
1963 occurrences of "<" in the file name (pretty unlikely); if there are,
1964 pcretest will instead treat the initial "<" as a pattern delimiter.
1965 After reading in the pattern, pcretest goes on to read data lines as
1966 usual.
1967
1968 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1969 and 16-bit fields in a compiled pattern, to simulate a pattern that
1970 was compiled on a host of opposite endianness.
1971
1972 21. The pcre-exec() function can now cope with patterns that were compiled on
1973 hosts of opposite endianness, with this restriction:
1974
1975 As for any compiled expression that is saved and used later, the tables
1976 pointer field cannot be preserved; the extra_data field in the arguments
1977 to pcre_exec() should be used to pass in a tables address if a value
1978 other than the default internal tables were used at compile time.
1979
1980 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1981 now diagnosed as an error. Previously, most of the time, a negative number
1982 would have been treated as zero, but if in addition "ovector" was passed as
1983 NULL, a crash could occur.
1984
1985 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1986 new versions from the libtool 1.5 distribution (the last one is a copy of
1987 a file called libtool.m4). This seems to have fixed the need to patch
1988 "configure" to support Darwin 1.3 (which I used to do). However, I still
1989 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1990 workstation).
1991
1992 24. Changed the PCRE licence to be the more standard "BSD" licence.
1993
1994
1995 Version 4.5 01-Dec-03
1996 ---------------------
1997
1998 1. There has been some re-arrangement of the code for the match() function so
1999 that it can be compiled in a version that does not call itself recursively.
2000 Instead, it keeps those local variables that need separate instances for
2001 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2002 needs to "recurse". Keeping track of where control must go is done by means
2003 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2004 hide most of the details from the main code, and operates only if
2005 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2006 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2007 operating.
2008
2009 To make it easier for callers to provide specially tailored get/free
2010 functions for this usage, two new functions, pcre_stack_malloc, and
2011 pcre_stack_free, are used. They are always called in strict stacking order,
2012 and the size of block requested is always the same.
2013
2014 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2015 PCRE has been compiled to use the stack or the heap for recursion. The
2016 -C option of pcretest uses this to show which version is compiled.
2017
2018 A new data escape \S, is added to pcretest; it causes the amounts of store
2019 obtained and freed by both kinds of malloc/free at match time to be added
2020 to the output.
2021
2022 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2023 what's available on my current Linux desktop machine.
2024
2025 3. When matching a UTF-8 string, the test for a valid string at the start has
2026 been extended. If start_offset is not zero, PCRE now checks that it points
2027 to a byte that is the start of a UTF-8 character. If not, it returns
2028 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2029 this is necessary because there may be backward assertions in the pattern.
2030 When matching the same subject several times, it may save resources to use
2031 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2032
2033 4. The code for checking the validity of UTF-8 strings has been tightened so
2034 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2035 containing "overlong sequences".
2036
2037 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2038 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2039 should have been "&", but it just so happened that all the cases this let
2040 through by mistake were picked up later in the function.
2041
2042 6. I had used a variable called "isblank" - this is a C99 function, causing
2043 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2044
2045 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2046 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2047 so the version is shown; (c) stop "make test" echoing "./RunTest".
2048
2049 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2050
2051 9. The prototype for memmove() for systems that don't have it was using
2052 size_t, but the inclusion of the header that defines size_t was later. I've
2053 moved the #includes for the C headers earlier to avoid this.
2054
2055 10. Added some adjustments to the code to make it easier to compiler on certain
2056 special systems:
2057
2058 (a) Some "const" qualifiers were missing.
2059 (b) Added the macro EXPORT before all exported functions; by default this
2060 is defined to be empty.
2061 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2062 that it reads its output file name as an argument instead of writing
2063 to the standard output and assuming this can be redirected.
2064
2065 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2066 class containing characters with values greater than 255, PCRE compilation
2067 went into a loop.
2068
2069 12. A recursive reference to a subpattern that was within another subpattern
2070 that had a minimum quantifier of zero caused PCRE to crash. For example,
2071 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2072 recursion. If the recursively-called subpattern itself had a zero repeat,
2073 that was OK.
2074
2075 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2076 buffer into which it was copied (for escape processing) was still set at
2077 1024, so long lines caused crashes.
2078
2079 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2080 "internal error: code overflow...". This applied to any character class
2081 that was followed by a possessive quantifier.
2082
2083 15. Modified the Makefile to add libpcre.la as a prerequisite for
2084 libpcreposix.la because I was told this is needed for a parallel build to
2085 work.
2086
2087 16. If a pattern that contained .* following optional items at the start was
2088 studied, the wrong optimizing data was generated, leading to matching
2089 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2090 matching string must start with a or b or c. The correct conclusion for
2091 this pattern is that a match can start with any character.
2092
2093
2094 Version 4.4 13-Aug-03
2095 ---------------------
2096
2097 1. In UTF-8 mode, a character class containing characters with values between
2098 127 and 255 was not handled correctly if the compiled pattern was studied.
2099 In fixing this, I have also improved the studying algorithm for such
2100 classes (slightly).
2101
2102 2. Three internal functions had redundant arguments passed to them. Removal
2103 might give a very teeny performance improvement.
2104
2105 3. Documentation bug: the value of the capture_top field in a callout is *one
2106 more than* the number of the hightest numbered captured substring.
2107
2108 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2109 in incorrectly linking with a previously installed version. They now link
2110 explicitly with libpcre.la.
2111
2112 5. configure.in no longer needs to recognize Cygwin specially.
2113
2114 6. A problem in pcre.in for Windows platforms is fixed.
2115
2116 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2117 pcretest, it used to include the size of the study block as part of its
2118 output. Unfortunately, the structure contains a field that has a different
2119 size on different hardware architectures. This meant that the tests that
2120 showed this size failed. As the block is currently always of a fixed size,
2121 this information isn't actually particularly useful in pcretest output, so
2122 I have just removed it.
2123
2124 8. Three pre-processor statements accidentally did not start in column 1.
2125 Sadly, there are *still* compilers around that complain, even though
2126 standard C has not required this for well over a decade. Sigh.
2127
2128 9. In pcretest, the code for checking callouts passed small integers in the
2129 callout_data field, which is a void * field. However, some picky compilers
2130 complained about the casts involved for this on 64-bit systems. Now
2131 pcretest passes the address of the small integer instead, which should get
2132 rid of the warnings.
2133
2134 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2135 both compile and run time, and gives an error if an invalid UTF-8 sequence
2136 is found. There is a option for disabling this check in cases where the
2137 string is known to be correct and/or the maximum performance is wanted.
2138
2139 11. In response to a bug report, I changed one line in Makefile.in from
2140
2141 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2142 to
2143 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2144
2145 to look similar to other lines, but I have no way of telling whether this
2146 is the right thing to do, as I do not use Windows. No doubt I'll get told
2147 if it's wrong...
2148
2149
2150 Version 4.3 21-May-03
2151 ---------------------
2152
2153 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2154 Makefile.
2155
2156 2. Some refactoring to improve the quality of the code:
2157
2158 (i) The utf8_table... variables are now declared "const".
2159
2160 (ii) The code for \cx, which used the "case flipping" table to upper case
2161 lower case letters, now just substracts 32. This is ASCII-specific,
2162 but the whole concept of \cx is ASCII-specific, so it seems
2163 reasonable.
2164
2165 (iii) PCRE was using its character types table to recognize decimal and
2166 hexadecimal digits in the pattern. This is silly, because it handles
2167 only 0-9, a-f, and A-F, but the character types table is locale-
2168 specific, which means strange things might happen. A private
2169 table is now used for this - though it costs 256 bytes, a table is
2170 much faster than multiple explicit tests. Of course, the standard
2171 character types table is still used for matching digits in subject
2172 strings against \d.
2173
2174 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2175 ending in _t are). So I've renamed it as ESC_tee.
2176
2177 3. The first argument for regexec() in the POSIX wrapper should have been
2178 defined as "const".
2179
2180 4. Changed pcretest to use malloc() for its buffers so that they can be
2181 Electric Fenced for debugging.
2182
2183 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2184 to read one or more bytes before the start of the subject string. Often this
2185 had no effect on PCRE's behaviour, but in some circumstances it could
2186 provoke a segmentation fault.
2187
2188 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2189 to try to read one or more bytes before the start of the subject string.
2190
2191 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2192 UTF-8 support could misbehave in various ways if the subject string
2193 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2194 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2195 back over UTF-8 characters.)
2196
2197
2198 Version 4.2 14-Apr-03
2199 ---------------------
2200
2201 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2202
2203 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2204 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2205 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2206 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2207 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2208 and BUILD_EXEEXT
2209 Note: automatic setting of the BUILD variables is not yet working
2210 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2211 compile-time but not at link-time
2212 [LINK]: use for linking executables only
2213 make different versions for Windows and non-Windows
2214 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2215 libraries
2216 [LINK_FOR_BUILD]: new variable
2217 [OBJEXT]: use throughout
2218 [EXEEXT]: use throughout
2219 <winshared>: new target
2220 <wininstall>: new target
2221 <dftables.o>: use native compiler
2222 <dftables>: use native linker
2223 <install>: handle Windows platform correctly
2224 <clean>: ditto
2225 <check>: ditto
2226 copy DLL to top builddir before testing
2227
2228 As part of these changes, -no-undefined was removed again. This was reported
2229 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2230 in any case.
2231
2232 3. Some tidies to get rid of compiler warnings:
2233
2234 . In the match_data structure, match_limit was an unsigned long int, whereas
2235 match_call_count was an int. I've made them both unsigned long ints.
2236
2237 . In pcretest the fact that a const uschar * doesn't automatically cast to
2238 a void * provoked a warning.
2239
2240 . Turning on some more compiler warnings threw up some "shadow" variables
2241 and a few more missing casts.
2242
2243 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2244 option, a class that contained a single character with a value between 128
2245 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2246
2247 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2248 option, a class that contained several characters, but with at least one
2249 whose value was between 128 and 255 caused PCRE to crash.
2250
2251
2252 Version 4.1 12-Mar-03
2253 ---------------------
2254
2255 1. Compiling with gcc -pedantic found a couple of places where casts were
2256 needed, and a string in dftables.c that was longer than standard compilers are
2257 required to support.
2258
2259 2. Compiling with Sun's compiler found a few more places where the code could
2260 be tidied up in order to avoid warnings.
2261
2262 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2263 first of these names is deprecated in the latest Autoconf in favour of the name
2264 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2265 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2266 analogy I have changed it to CFLAGS_FOR_BUILD.
2267
2268 4. Added -no-undefined to the linking command in the Makefile, because this is
2269 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2270 linking step for the pcreposix library.
2271
2272 5. PCRE was failing to diagnose the case of two named groups with the same
2273 name.
2274
2275 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2276 literal character that is needed in the subject for a match, and scans along to
2277 ensure that it is present before embarking on the full matching process. This
2278 saves time in cases of nested unlimited repeats that are never going to match.
2279 Problem: the scan can take a lot of time if the subject is very long (e.g.
2280 megabytes), thus penalizing straightforward matches. It is now done only if the
2281 amount of subject to be scanned is less than 1000 bytes.
2282
2283 7. A lesser problem with the same optimization is that it was recording the
2284 first character of an anchored pattern as "needed", thus provoking a search
2285 right along the subject, even when the first match of the pattern was going to
2286 fail. The "needed" character is now not set for anchored patterns, unless it
2287 follows something in the pattern that is of non-fixed length. Thus, it still
2288 fulfils its original purpose of finding quick non-matches in cases of nested
2289 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2290
2291
2292 Version 4.0 17-Feb-03
2293 ---------------------
2294
2295 1. If a comment in an extended regex that started immediately after a meta-item
2296 extended to the end of string, PCRE compiled incorrect data. This could lead to
2297 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2298
2299 2. Moved to autoconf 2.53 and libtool 1.4.2.
2300
2301 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2302 the special perltest8 script is no longer needed - all the tests can be run
2303 from a single perltest script.
2304
2305 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2306 by \s. It has now been removed in PCRE. This means it isn't recognized as
2307 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2308 class [:space:] *does* include VT, thereby creating a mess.
2309
2310 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2311 space and tab.
2312
2313 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2314 its new features into the main test script, reducing the number of scripts.
2315
2316 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2317 were backward compatible, and made the (?i) apply to the whole pattern, as if
2318 /i were given. Now it behaves more logically, and applies the option setting
2319 only to what follows. PCRE has been changed to follow suit. However, if it
2320 finds options settings right at the start of the pattern, it extracts them into
2321 the global options, as before. Thus, they show up in the info data.
2322
2323 8. Added support for the \Q...\E escape sequence. Characters in between are
2324 treated as literals. This is slightly different from Perl in that $ and @ are
2325 also handled as literals inside the quotes. In Perl, they will cause variable
2326 interpolation. Note the following examples:
2327
2328 Pattern PCRE matches Perl matches
2329
2330 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2331 \Qabc\$xyz\E abc\$xyz abc\$xyz
2332 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2333
2334 For compatibility with Perl, \Q...\E sequences are recognized inside character
2335 classes as well as outside them.
2336
2337 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2338 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2339 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2340 signed/unsigned warnings.
2341
2342 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2343 option for pcretest, so I've replaced it by a simple function that does just
2344 that job.
2345
2346 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2347 "pcregrep -".
2348
2349 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2350 Java package. This provides some syntactic sugar for simple cases of what my
2351 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2352 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2353 item, you can use this simplified notation. Note that only makes sense with
2354 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2355 greediness, whatever the setting of the PCRE_UNGREEDY option.
2356
2357 13. A change of greediness default within a pattern was not taking effect at
2358 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2359 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2360 was abstracted outside.
2361
2362 14. PCRE now supports the \G assertion. It is true when the current matching
2363 position is at the start point of the match. This differs from \A when the
2364 starting offset is non-zero. Used with the /g option of pcretest (or similar
2365 code), it works in the same way as it does for Perl's /g option. If all
2366 alternatives of a regex begin with \G, the expression is anchored to the start
2367 match position, and the "anchored" flag is set in the compiled expression.
2368
2369 15. Some bugs concerning the handling of certain option changes within patterns
2370 have been fixed. These applied to options other than (?ims). For example,
2371 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2372 been the other way round. Some of this was related to change 7 above.
2373
2374 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2375 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2376 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2377 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2378
2379 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2380 mode. Unlike ".", it always matches newline, whatever the setting of
2381 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2382 assertions. Perl allows it, but it doesn't (in general) work because it can't
2383 calculate the length of the lookbehind. At least, that's the case for Perl
2384 5.8.0 - I've been told they are going to document that it doesn't work in
2385 future.
2386
2387 18. Added an error diagnosis for escapes that PCRE does not support: these are
2388 \L, \l, \N, \P, \p, \U, \u, and \X.
2389
2390 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2391 reading past the end of the pattern in cases such as /[abcd/.
2392
2393 20. PCRE was getting more memory than necessary for patterns with classes that
2394 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2395
2396 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2397 compiling PCRE for use with Virtual Pascal.
2398
2399 22. Small fix to the Makefile to make it work properly if the build is done
2400 outside the source tree.
2401
2402 23. Added a new extension: a condition to go with recursion. If a conditional
2403 subpattern starts with (?(R) the "true" branch is used if recursion has
2404 happened, whereas the "false" branch is used only at the top level.
2405
2406 24. When there was a very long string of literal characters (over 255 bytes
2407 without UTF support, over 250 bytes with UTF support), the computation of how
2408 much memory was required could be incorrect, leading to segfaults or other
2409 strange effects.
2410
2411 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2412 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2413 there was a subsequent back reference to those brackets. This meant that, for
2414 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2415 possible to check for precisely this case. All we can do is abandon the
2416 optimization if .* occurs inside capturing brackets when there are any back
2417 references whatsoever. (See below for a better fix that came later.)
2418
2419 26. The handling of the optimization for finding the first character of a
2420 non-anchored pattern, and for finding a character that is required later in the
2421 match were failing in some cases. This didn't break the matching; it just
2422 failed to optimize when it could. The way this is done has been re-implemented.
2423
2424 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2425
2426 28. Added a new feature that provides some of the functionality that Perl
2427 provides with (?{...}). The facility is termed a "callout". The way it is done
2428 in PCRE is for the caller to provide an optional function, by setting
2429 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2430 global variable. By default it is unset, which disables all calling out. To get
2431 the function called, the regex must include (?C) at appropriate points. This
2432 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2433 This provides a means of identifying different callout points. When PCRE
2434 reaches such a point in the regex, if pcre_callout has been set, the external
2435 function is called. It is provided with data in a structure called
2436 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2437 matching continues; if it returns a non-zero value, the match at the current
2438 point fails. However, backtracking will occur if possible. [This was changed
2439 later and other features added - see item 49 below.]
2440
2441 29. pcretest is upgraded to test the callout functionality. It provides a
2442 callout function that displays information. By default, it shows the start of
2443 the match and the current position in the text. There are some new data escapes
2444 to vary what happens:
2445
2446 \C+ in addition, show current contents of captured substrings
2447 \C- do not supply a callout function
2448 \C!n return 1 when callout number n is reached
2449 \C!n!m return 1 when callout number n is reached for the mth time
2450
2451 30. If pcregrep was called with the -l option and just a single file name, it
2452 output "<stdin>" if a match was found, instead of the file name.
2453
2454 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2455 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2456 pcre_exec(). This saves a malloc/free per call. The default value of
2457 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2458 when configuring.
2459
2460 32. The default maximum size of a compiled pattern is 64K. There have been a
2461 few cases of people hitting this limit. The code now uses macros to handle the
2462 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2463 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2464 configuring. Tests 2 and 5 work only with 2-byte links because they output
2465 debugging information about compiled patterns.
2466
2467 33. Internal code re-arrangements:
2468
2469 (a) Moved the debugging function for printing out a compiled regex into
2470 its own source file (printint.c) and used #include to pull it into
2471 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2472 separate copies.
2473
2474 (b) Defined the list of op-code names for debugging as a macro in
2475 internal.h so that it is next to the definition of the opcodes.
2476
2477 (c) Defined a table of op-code lengths for simpler skipping along compiled
2478 code. This is again a macro in internal.h so that it is next to the
2479 definition of the opcodes.
2480
2481 34. Added support for recursive calls to individual subpatterns, along the
2482 lines of Robin Houston's patch (but implemented somewhat differently).
2483
2484 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2485 allow it to read and process whole directories in Win32. This code was
2486 contributed by Lionel Fourquaux; it has not been tested by me.
2487
2488 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2489 used to name a group. Names consist of alphanumerics and underscores, and must
2490 be unique. Back references use the syntax (?P=name) and recursive calls use
2491 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2492 numbers. The function pcre_fullinfo() can be used after compilation to extract
2493 a name/number map. There are three relevant calls:
2494
2495 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2496 PCRE_INFO_NAMECOUNT yields the number of entries
2497 PCRE_INFO_NAMETABLE yields a pointer to the map.
2498
2499 The map is a vector of fixed-size entries. The size of each entry depends on
2500 the length of the longest name used. The first two bytes of each entry are the
2501 group number, most significant byte first. There follows the corresponding
2502 name, zero terminated. The names are in alphabetical order.
2503
2504 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2505 case instead of 255. Making it the same both with and without UTF-8 support
2506 means that the same test output works with both.
2507
2508 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2509 calling malloc() with a zero argument.
2510
2511 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2512 optimization. I've improved things by keeping a bitmap of backreferences with
2513 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2514 fact referenced, the optimization can be applied. It is unlikely that a
2515 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2516 the match to follow \n) will appear inside brackets with a number greater than
2517 31, but if it does, any back reference > 31 suppresses the optimization.
2518
2519 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2520 of disabling numbered capturing parentheses. Any opening parenthesis that is
2521 not followed by ? behaves as if it were followed by ?: but named parentheses
2522 can still be used for capturing (and they will acquire numbers in the usual
2523 way).
2524
2525 41. Redesigned the return codes from the match() function into yes/no/error so
2526 that errors can be passed back from deep inside the nested calls. A malloc
2527 failure while inside a recursive subpattern call now causes the
2528 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2529
2530 42. It is now possible to set a limit on the number of times the match()
2531 function is called in a call to pcre_exec(). This facility makes it possible to
2532 limit the amount of recursion and backtracking, though not in a directly
2533 obvious way, because the match() function is used in a number of different
2534 circumstances. The count starts from zero for each position in the subject
2535 string (for non-anchored patterns). The default limit is, for compatibility, a
2536 large number, namely 10 000 000. You can change this in two ways:
2537
2538 (a) When configuring PCRE before making, you can use --with-match-limit=n
2539 to set a default value for the compiled library.
2540
2541 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2542 a different value is set. See 45 below.
2543
2544 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2545
2546 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2547 of things that can be changed at compile time. The first argument specifies
2548 what is wanted and the second points to where the information is to be placed.
2549 The current list of available information is:
2550
2551 PCRE_CONFIG_UTF8
2552
2553 The output is an integer that is set to one if UTF-8 support is available;
2554 otherwise it is set to zero.
2555
2556 PCRE_CONFIG_NEWLINE
2557
2558 The output is an integer that it set to the value of the code that is used for
2559 newline. It is either LF (10) or CR (13).
2560
2561 PCRE_CONFIG_LINK_SIZE
2562
2563 The output is an integer that contains the number of bytes used for internal
2564 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2565
2566 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2567
2568 The output is an integer that contains the threshold above which the POSIX
2569 interface uses malloc() for output vectors. See item 31 above.
2570
2571 PCRE_CONFIG_MATCH_LIMIT
2572
2573 The output is an unsigned integer that contains the default limit of the number
2574 of match() calls in a pcre_exec() execution. See 42 above.
2575
2576 44. pcretest has been upgraded by the addition of the -C option. This causes it
2577 to extract all the available output from the new pcre_config() function, and to
2578 output it. The program then exits immediately.
2579
2580 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2581 order to support additional features. One way would have been to define
2582 pcre_exec2() (for example) with extra arguments, but this would not have been
2583 extensible, and would also have required all calls to the original function to
2584 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2585 is used for passing in "extra" data from pcre_study().
2586
2587 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2588 contains the following fields:
2589
2590 flags a bitmap indicating which of the following fields are set
2591 study_data opaque data from pcre_study()
2592 match_limit a way of specifying a limit on match() calls for a specific
2593 call to pcre_exec()
2594 callout_data data for callouts (see 49 below)
2595
2596 The flag bits are also defined in pcre.h, and are
2597
2598 PCRE_EXTRA_STUDY_DATA
2599 PCRE_EXTRA_MATCH_LIMIT
2600 PCRE_EXTRA_CALLOUT_DATA
2601
2602 The pcre_study() function now returns one of these new pcre_extra blocks, with
2603 the actual study data pointed to by the study_data field, and the
2604 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2605 before. That is, this change is entirely upwards-compatible and requires no
2606 change to existing code.
2607
2608 If you want to pass in additional data to pcre_exec(), you can either place it
2609 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2610 block.
2611
2612 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2613 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2614 times with different match limits, until it finds the minimum value needed for
2615 pcre_exec() to complete. The value is then output. This can be instructive; for
2616 most simple matches the number is quite small, but for pathological cases it
2617 gets very large very quickly.
2618
2619 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2620 returns the size of the data block pointed to by the study_data field in a
2621 pcre_extra block, that is, the value that was passed as the argument to
2622 pcre_malloc() when PCRE was getting memory in which to place the information
2623 created by pcre_study(). The fourth argument should point to a size_t variable.
2624 pcretest has been extended so that this information is shown after a successful
2625 pcre_study() call when information about the compiled regex is being displayed.
2626
2627 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2628 because what follows is always an absolute path. (Later: it turns out that this
2629 is more than cosmetic for MinGW, because it doesn't like empty path
2630 components.)
2631
2632 49. Some changes have been made to the callout feature (see 28 above):
2633
2634 (i) A callout function now has three choices for what it returns:
2635
2636 0 => success, carry on matching
2637 > 0 => failure at this point, but backtrack if possible
2638 < 0 => serious error, return this value from pcre_exec()
2639
2640 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2641 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2642 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2643 use by callout functions. It will never be used by PCRE itself.
2644
2645 (ii) The pcre_extra structure (see 45 above) has a void * field called
2646 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2647 pcre_callout_block structure has a field of the same name. The contents of
2648 the field passed in the pcre_extra structure are passed to the callout
2649 function in the corresponding field in the callout block. This makes it
2650 easier to use the same callout-containing regex from multiple threads. For
2651 testing, the pcretest program has a new data escape
2652
2653 \C*n pass the number n (may be negative) as callout_data
2654
2655 If the callout function in pcretest receives a non-zero value as
2656 callout_data, it returns that value.
2657
2658 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2659 there were some redundant $(CFLAGS) in commands that are now specified as
2660 $(LINK), which already includes $(CFLAGS).
2661
2662 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2663 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2664 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2665 one-byte characters throughout. Note that case-insensitive matching applies
2666 only to characters whose values are less than 256. PCRE doesn't support the
2667 notion of cases for higher-valued characters.
2668
2669 (i) A character class whose characters are all within 0-255 is handled as
2670 a bit map, and the map is inverted for negative classes. Previously, a
2671 character > 255 always failed to match such a class; however it should
2672 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2673
2674 (ii) A negated character class with a single character < 255 is coded as
2675 "not this character" (OP_NOT). This wasn't working properly when the test
2676 character was multibyte, either singly or repeated.
2677
2678 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2679 mode, for example: \x{100}{2,3}.
2680
2681 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2682 singly or repeated) now correctly test multibyte characters. However,
2683 PCRE doesn't recognize any characters with values greater than 255 as
2684 digits, spaces, or word characters. Such characters always match \D, \S,
2685 and \W, and never match \d, \s, or \w.
2686
2687 (v) Classes may now contain characters and character ranges with values
2688 greater than 255. For example: [ab\x{100}-\x{400}].
2689
2690 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2691 PCRE in UTF-8 mode.
2692
2693 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2694 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2695 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2696 value.)
2697
2698 53. The single man page has become too large. I have therefore split it up into
2699 a number of separate man pages. These also give rise to individual HTML pages;
2700 these are now put in a separate directory, and there is an index.html page that
2701 lists them all. Some hyperlinking between the pages has been installed.
2702
2703 54. Added convenience functions for handling named capturing parentheses.
2704
2705 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2706 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2707 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2708 are faulted.
2709
2710 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2711 calling configure. These values are used when compiling the dftables.c program
2712 which is run to generate the source of the default character tables. They
2713 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2714 you will need to set these values.
2715
2716 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2717
2718
2719 Version 3.9 02-Jan-02
2720 ---------------------
2721
2722 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2723
2724 2. If --disable-static was given, the building process failed when trying to
2725 build pcretest and pcregrep. (For some reason it was using libtool to compile
2726 them, which is not right, as they aren't part of the library.)
2727
2728
2729 Version 3.8 18-Dec-01
2730 ---------------------
2731
2732 1. The experimental UTF-8 code was completely screwed up. It was packing the
2733 bytes in the wrong order. How dumb can you get?
2734
2735
2736 Version 3.7 29-Oct-01
2737 ---------------------
2738
2739 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2740 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2741 this didn't happen under Solaris 8, where I normally test things.
2742
2743 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2744 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2745 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2746 this fix an hour or so after the initial 3.7 release.)
2747
2748
2749 Version 3.6 23-Oct-01
2750 ---------------------
2751
2752 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2753 offsets passed as NULL with zero offset count.
2754
2755 2. The config.guess and config.sub files had not been updated when I moved to
2756 the latest autoconf.
2757
2758
2759 Version 3.5 15-Aug-01
2760 ---------------------
2761
2762 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2763 had been forgotten.
2764
2765 2. By using declared but undefined structures, we can avoid using "void"
2766 definitions in pcre.h while keeping the internal definitions of the structures
2767 private.
2768
2769 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2770 user point of view, this means that both static and shared libraries are built
2771 by default, but this can be individually controlled. More of the work of
2772 handling this static/shared cases is now inside libtool instead of PCRE's make
2773 file.
2774
2775 4. The pcretest utility is now installed along with pcregrep because it is
2776 useful for users (to test regexs) and by doing this, it automatically gets
2777 relinked by libtool. The documentation has been turned into a man page, so
2778 there are now .1, .txt, and .html versions in /doc.
2779
2780 5. Upgrades to pcregrep:
2781 (i) Added long-form option names like gnu grep.
2782 (ii) Added --help to list all options with an explanatory phrase.
2783 (iii) Added -r, --recursive to recurse into sub-directories.
2784 (iv) Added -f, --file to read patterns from a file.
2785
2786 6. pcre_exec() was referring to its "code" argument before testing that
2787 argument for NULL (and giving an error if it was NULL).
2788
2789 7. Upgraded Makefile.in to allow for compiling in a different directory from
2790 the source directory.
2791
2792 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2793 options bits, the pointer it was passed was to an int instead of to an unsigned
2794 long int. This mattered only on 64-bit systems.
2795
2796 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2797 generated) instead of pcre.in, which it its source. Also made the same change
2798 in several of the .c files.
2799
2800 10. A new release of gcc defines printf() as a macro, which broke pcretest
2801 because it had an ifdef in the middle of a string argument for printf(). Fixed
2802 by using separate calls to printf().
2803
2804 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2805 script, to force use of CR or LF instead of \n in the source. On non-Unix
2806 systems, the value can be set in config.h.
2807
2808 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2809 absolute limit. Changed the text of the error message to make this clear, and
2810 likewise updated the man page.
2811
2812 13. The limit of 99 on the number of capturing subpatterns has been removed.
2813 The new limit is 65535, which I hope will not be a "real" limit.
2814
2815
2816 Version 3.4 22-Aug-00
2817 ---------------------
2818
2819 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2820
2821 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2822
2823
2824 Version 3.3 01-Aug-00
2825 ---------------------
2826
2827 1. If an octal character was given, but the value was greater than \377, it
2828 was not getting masked to the least significant bits, as documented. This could
2829 lead to crashes in some systems.
2830
2831 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2832 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2833
2834 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2835 These just pass their arguments on to (pcre_free)(), but they are provided
2836 because some uses of PCRE bind it to non-C systems that can call its functions,
2837 but cannot call free() or pcre_free() directly.
2838
2839 4. Add "make test" as a synonym for "make check". Corrected some comments in
2840 the Makefile.
2841
2842 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2843 Makefile.
2844
2845 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2846 command called pgrep for grepping around the active processes.
2847
2848 7. Added the beginnings of support for UTF-8 character strings.
2849
2850 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2851 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2852 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2853 out for the ar command.)
2854
2855
2856 Version 3.2 12-May-00
2857 ---------------------
2858
2859 This is purely a bug fixing release.
2860
2861 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2862 of ZA. This was just one example of several cases that could provoke this bug,
2863 which was introduced by change 9 of version 2.00. The code for breaking
2864 infinite loops after an iteration that matches an empty string was't working
2865 correctly.
2866
2867 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2868 when matched against abbab (for example). After matching an empty string, it
2869 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2870 caused it to match further down the string than it should.
2871
2872 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2873 was there because it doesn't seem to be needed, and it causes trouble on some
2874 systems, as it is not a Standard C header. It has been removed.
2875
2876 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2877 were reported on the Macintosh. The changes were from
2878
2879 while ((c = *(++ptr)) != 0 && c != '\n');
2880 to
2881 while ((c = *(++ptr)) != 0 && c != '\n') ;
2882
2883 Totally extraordinary, but if that's what it takes...
2884
2885 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2886 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2887 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2888 assumes the way PCRE uses memmove() (always moving upwards).
2889
2890 6. PCRE is being used in one environment where strchr() is not available. There
2891 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2892 faster code anyway.
2893
2894
2895 Version 3.1 09-Feb-00
2896 ---------------------
2897
2898 The only change in this release is the fixing of some bugs in Makefile.in for
2899 the "install" target:
2900
2901 (1) It was failing to install pcreposix.h.
2902
2903 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2904
2905
2906 Version 3.0 01-Feb-00
2907 ---------------------
2908
2909 1. Add support for the /+ modifier to perltest (to output $` like it does in
2910 pcretest).
2911
2912 2. Add support for the /g modifier to perltest.
2913
2914 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2915 matches null strings.
2916
2917 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2918 pattern. Perl treats empty patterns specially - it reuses the most recent
2919 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2920 effect.
2921
2922 5. The POSIX interface was broken in that it was just handing over the POSIX
2923 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2924 required a bigger vector, with some working space on the end. This means that
2925 the POSIX wrapper now has to get and free some memory, and copy the results.
2926
2927 6. Added some simple autoconf support, placing the test data and the
2928 documentation in separate directories, re-organizing some of the
2929 information files, and making it build pcre-config (a GNU standard). Also added
2930 libtool support for building PCRE as a shared library, which is now the
2931 default.
2932
2933 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2934 09 are not valid octal constants. Single digits will be used for minor values
2935 less than 10.
2936
2937 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2938 existing programs that set these in the POSIX interface can use PCRE without
2939 modification.
2940
2941 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2942 return all that pcre_info() returns, plus additional data. The pcre_info()
2943 function is retained for compatibility, but is considered to be obsolete.
2944
2945 10. Added experimental recursion feature (?R) to handle one common case that
2946 Perl 5.6 will be able to do with (?p{...}).
2947
2948 11. Added support for POSIX character classes like [:alpha:], which Perl is
2949 adopting.
2950
2951
2952 Version 2.08 31-Aug-99
2953 ----------------------
2954
2955 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2956 trying to match at the startoffset position, but instead was moving forward to
2957 the next newline as if a previous match had failed.
2958
2959 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2960 and could get into a loop if a null string was matched other than at the start
2961 of the subject.
2962
2963 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2964 be distinguished at compile time, and for completeness also added PCRE_DATE.
2965
2966 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2967 in GnuWin32 environments.
2968
2969
2970 Version 2.07 29-Jul-99
2971 ----------------------
2972
2973 1. The documentation is now supplied in plain text form and HTML as well as in
2974 the form of man page sources.
2975
2976 2. C++ compilers don't like assigning (void *) values to other pointer types.
2977 In particular this affects malloc(). Although there is no problem in Standard
2978 C, I've put in casts to keep C++ compilers happy.
2979
2980 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2981 should be (const char *).
2982
2983 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2984 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2985 However, I haven't made this a standard facility. The documentation doesn't
2986 mention it, and the Makefile doesn't support it.
2987
2988 5. The Makefile now contains an "install" target, with editable destinations at
2989 the top of the file. The pcretest program is not installed.
2990
2991 6. pgrep -V now gives the PCRE version number and date.
2992
2993 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2994 causing the entire string to be ignored, instead of just the last character.
2995
2996 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2997 non-matching string, it can take a very, very long time, even for strings of
2998 quite modest length, because of the nested recursion. PCRE now does better in
2999 some of these cases. It does this by remembering the last required literal
3000 character in the pattern, and pre-searching the subject to ensure it is present
3001 before running the real match. In other words, it applies a heuristic to detect
3002 some types of certain failure quickly, and in the above example, if presented
3003 with a string that has no trailing " it gives "no match" very quickly.
3004
3005 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3006 other alternatives are tried instead.
3007
3008
3009 Version 2.06 09-Jun-99
3010 ----------------------
3011
3012 1. Change pcretest's output for amount of store used to show just the code
3013 space, because the remainder (the data block) varies in size between 32-bit and
3014 64-bit systems.
3015
3016 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3017 start matching at. This allows lookbehinds to work when searching for multiple
3018 occurrences in a string.
3019
3020 3. Added additional options to pcretest for testing multiple occurrences:
3021
3022 /+ outputs the rest of the string that follows a match
3023 /g loops for multiple occurrences, using the new startoffset argument
3024 /G loops for multiple occurrences by passing an incremented pointer
3025
3026 4. PCRE wasn't doing the "first character" optimization for patterns starting
3027 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3028 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3029 the letter 'x'. On long subject strings, this gives a significant speed-up.
3030
3031
3032 Version 2.05 21-Apr-99
3033 ----------------------
3034
3035 1. Changed the type of magic_number from int to long int so that it works
3036 properly on 16-bit systems.
3037
3038 2. Fixed a bug which caused patterns starting with .* not to work correctly
3039 when the subject string contained newline characters. PCRE was assuming
3040 anchoring for such patterns in all cases, which is not correct because .* will
3041 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3042 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3043 must be retried after every newline in the subject.
3044
3045
3046 Version 2.04 18-Feb-99
3047 ----------------------
3048
3049 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3050 computation of the store needed to hold the pattern was incorrect (too large).
3051 If such patterns were nested a few deep, this could multiply and become a real
3052 problem.
3053
3054 2. Added /M option to pcretest to show the memory requirement of a specific
3055 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3056
3057 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3058 compiled in such a way that the backtracking after subsequent failure was
3059 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3060 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3061
3062
3063 Version 2.03 02-Feb-99
3064 ----------------------
3065
3066 1. Fixed typo and small mistake in man page.
3067
3068 2. Added 4th condition (GPL supersedes if conflict) and created separate
3069 LICENCE file containing the conditions.
3070
3071 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3072 Perl, that is the internal \ allows the delimiter to be included in the
3073 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3074 the final delimiter, add \ to the end of the pattern (to test the error).
3075
3076 4. Added the convenience functions for extracting substrings after a successful
3077 match. Updated pcretest to make it able to test these functions.
3078
3079
3080 Version 2.02 14-Jan-99
3081 ----------------------
3082
3083 1. Initialized the working variables associated with each extraction so that
3084 their saving and restoring doesn't refer to uninitialized store.
3085
3086 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3087 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3088 fix the problem.
3089
3090 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3091 calls, and wasn't printing the correct value for compiling calls. Increased the
3092 default value of LOOPREPEAT, and the number of significant figures in the
3093 times.
3094
3095 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3096
3097 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3098 a building problem on Windows NT with a FAT file system.
3099
3100
3101 Version 2.01 21-Oct-98
3102 ----------------------
3103
3104 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3105 to character tables built by pcre_maketables() in the current locale. If NULL
3106 is passed, the default tables are used.
3107
3108
3109 Version 2.00 24-Sep-98
3110 ----------------------
3111
3112 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3113 it any more.
3114
3115 2. Allow quantification of (?>) groups, and make it work correctly.
3116
3117 3. The first character computation wasn't working for (?>) groups.
3118
3119 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3120 end of the subject) and add 5.005's \z, which really does match only at the
3121 very end of the subject.
3122
3123 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3124
3125 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3126 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3127 localized options. All options to pcre_study() were also removed.
3128
3129 7. Add other new features from 5.005:
3130
3131 $(?<= positive lookbehind
3132 $(?<! negative lookbehind
3133 (?imsx-imsx) added the unsetting capability
3134 such a setting is global if at outer level; local otherwise
3135 (?imsx-imsx:) non-capturing groups with option setting
3136 (?(cond)re|re) conditional pattern matching
3137
3138 A backreference to itself in a repeated group matches the previous
3139 captured string.
3140
3141 8. General tidying up of studying (both automatic and via "study")
3142 consequential on the addition of new assertions.
3143
3144 9. As in 5.005, unlimited repeated groups that could match an empty substring
3145 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3146 runtime if any iteration does actually match an empty substring.
3147
3148 10. Include the RunTest script in the distribution.
3149
3150 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3151 discrepancies, some of which were old and were also with respect to 5.004. They
3152 have now been fixed.
3153
3154
3155 Version 1.09 28-Apr-98
3156 ----------------------
3157
3158 1. A negated single character class followed by a quantifier with a minimum
3159 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3160 program crashes, or just wrong answers. This did not apply to negated classes
3161 containing more than one character, or to minima other than one.
3162
3163
3164 Version 1.08 27-Mar-98
3165 ----------------------
3166
3167 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3168
3169 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3170 latter must appear before anything that relies on it in the pattern.
3171
3172
3173 Version 1.07 16-Feb-98
3174 ----------------------
3175
3176 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3177 repeat of a potentially empty string).
3178
3179
3180 Version 1.06 23-Jan-98
3181 ----------------------
3182
3183 1. Added Markus Oberhumer's little patches for C++.
3184
3185 2. Literal strings longer than 255 characters were broken.
3186
3187
3188 Version 1.05 23-Dec-97
3189 ----------------------
3190
3191 1. Negated character classes containing more than one character were failing if
3192 PCRE_CASELESS was set at run time.
3193
3194
3195 Version 1.04 19-Dec-97
3196 ----------------------
3197
3198 1. Corrected the man page, where some "const" qualifiers had been omitted.
3199
3200 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3201 input syntax.
3202
3203 3. Fixed memory leak which occurred when a regex with back references was
3204 matched with an offsets vector that wasn't big enough. The temporary memory
3205 that is used in this case wasn't being freed if the match failed.
3206
3207 4. Tidied pcretest to ensure it frees memory that it gets.
3208
3209 5. Temporary memory was being obtained in the case where the passed offsets
3210 vector was exactly big enough.
3211
3212 6. Corrected definition of offsetof() from change 5 below.
3213
3214 7. I had screwed up change 6 below and broken the rules for the use of
3215 setjmp(). Now fixed.
3216
3217
3218 Version 1.03 18-Dec-97
3219 ----------------------
3220
3221 1. A erroneous regex with a missing opening parenthesis was correctly
3222 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3223 on some systems.
3224
3225 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3226 it was reported that one broken compiler failed on the former because "code" is
3227 also an independent variable.
3228
3229 3. The erroneous regex a[]b caused an array overrun reference.
3230
3231 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3232 fail on data ending with that character. (It was going on too far, and checking
3233 the next character, typically a binary zero.) This was specific to the
3234 optimized code for single-character negative classes.
3235
3236 5. Added a contributed patch from the TIN world which does the following:
3237
3238 + Add an undef for memmove, in case the the system defines a macro for it.
3239
3240 + Add a definition of offsetof(), in case there isn't one. (I don't know
3241 the reason behind this - offsetof() is part of the ANSI standard - but
3242 it does no harm).
3243
3244 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3245 most of the places where whitespace preceded '#'. I have given up and
3246 allowed the remaining 2 cases to be at the margin.
3247
3248 + Rename some variables in pcre to eliminate shadowing. This seems very
3249 pedantic, but does no harm, of course.
3250
3251 6. Moved the call to setjmp() into its own function, to get rid of warnings
3252 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3253
3254 7. Constructs such as \d{8,} were compiling into the equivalent of
3255 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3256 outcome, but in this particular case used more store than had been allocated,
3257 which caused the bug to be discovered because it threw up an internal error.
3258
3259 8. The debugging code in both pcre and pcretest for outputting the compiled
3260 form of a regex was going wrong in the case of back references followed by
3261 curly-bracketed repeats.
3262
3263
3264 Version 1.02 12-Dec-97
3265 ----------------------
3266
3267 1. Typos in pcre.3 and comments in the source fixed.
3268
3269 2. Applied a contributed patch to get rid of places where it used to remove
3270 'const' from variables, and fixed some signed/unsigned and uninitialized
3271 variable warnings.
3272
3273 3. Added the "runtest" target to Makefile.
3274
3275 4. Set default compiler flag to -O2 rather than just -O.
3276
3277
3278 Version 1.01 19-Nov-97
3279 ----------------------
3280
3281 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3282 like /([ab]*)*/, that is, for classes with more than one character in them.
3283
3284 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3285 as /((?>a*))*/ (a PCRE_EXTRA facility).
3286
3287
3288 Version 1.00 18-Nov-97
3289 ----------------------
3290
3291 1. Added compile-time macros to support systems such as SunOS4 which don't have
3292 memmove() or strerror() but have other things that can be used instead.
3293
3294 2. Arranged that "make clean" removes the executables.
3295
3296
3297 Version 0.99 27-Oct-97
3298 ----------------------
3299
3300 1. Fixed bug in code for optimizing classes with only one character. It was
3301 initializing a 32-byte map regardless, which could cause it to run off the end
3302 of the memory it had got.
3303
3304 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3305
3306
3307 Version 0.98 22-Oct-97
3308 ----------------------
3309
3310 1. Fixed bug in code for handling temporary memory usage when there are more
3311 back references than supplied space in the ovector. This could cause segfaults.
3312
3313
3314 Version 0.97 21-Oct-97
3315 ----------------------
3316
3317 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3318
3319 2. Optimized negated single characters not to use a bit map.
3320
3321 3. Brought error texts together as macro definitions; clarified some of them;
3322 fixed one that was wrong - it said "range out of order" when it meant "invalid
3323 escape sequence".
3324
3325 4. Changed some char * arguments to const char *.
3326
3327 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3328
3329 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3330 pcretest.
3331
3332
3333 Version 0.96 16-Oct-97
3334 ----------------------
3335
3336 1. Added a simple "pgrep" utility to the distribution.
3337
3338 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3339 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3340 where "ddd" means "one or more decimal digits".
3341
3342 3. Fixed serious bug. If a pattern had a back reference, but the call to
3343 pcre_exec() didn't supply a large enough ovector to record the related
3344 identifying subpattern, the match always failed. PCRE now remembers the number
3345 of the largest back reference, and gets some temporary memory in which to save
3346 the offsets during matching if necessary, in order to ensure that
3347 backreferences always work.
3348
3349 4. Increased the compatibility with Perl in a number of ways:
3350
3351 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3352 to request this handling. The option can be set at compile or exec time.
3353
3354 (b) $ matches before a terminating newline by default; an option
3355 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3356 mode). The option can be set at compile or exec time.
3357
3358 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3359 the same as Perl's. If the decimal number it represents is less than 10
3360 or there aren't that many previous left capturing parentheses, an octal
3361 escape is read. Inside a character class, it's always an octal escape,
3362 even if it is a single digit.
3363
3364 (d) An escaped but undefined alphabetic character is taken as a literal,
3365 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3366 escapes.
3367
3368 (e) {0} is now permitted. (The previous item is removed from the compiled
3369 pattern).
3370
3371 5. Changed all the names of code files so that the basic parts are no longer
3372 than 10 characters, and abolished the teeny "globals.c" file.
3373
3374 6. Changed the handling of character classes; they are now done with a 32-byte
3375 bit map always.
3376
3377 7. Added the -d and /D options to pcretest to make it possible to look at the
3378 internals of compilation without having to recompile pcre.
3379
3380
3381 Version 0.95 23-Sep-97
3382 ----------------------
3383
3384 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3385 \x20 at the start of a run of normal characters. These were being treated as
3386 real characters, instead of the source characters being re-checked.
3387
3388
3389 Version 0.94 18-Sep-97
3390 ----------------------
3391
3392 1. The functions are now thread-safe, with the caveat that the global variables
3393 containing pointers to malloc() and free() or alternative functions are the
3394 same for all threads.
3395
3396 2. Get pcre_study() to generate a bitmap of initial characters for non-
3397 anchored patterns when this is possible, and use it if passed to pcre_exec().
3398
3399
3400 Version 0.93 15-Sep-97
3401 ----------------------
3402
3403 1. /(b)|(:+)/ was computing an incorrect first character.
3404
3405 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3406 but not actually doing anything yet.
3407
3408 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3409 as Perl does (e.g. [-az] or [az-]).
3410
3411 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3412 all possible positions.
3413
3414 5. Split up into different modules to avoid including unneeded functions in a
3415 compiled binary. However, compile and exec are still in one module. The "study"
3416 function is split off.
3417
3418 6. The character tables are now in a separate module whose source is generated
3419 by an auxiliary program - but can then be edited by hand if required. There are
3420 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3421 toupper() in the code.
3422
3423 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3424 make them global. Abolish the function for setting them, as the caller can now
3425 set them directly.
3426
3427
3428 Version 0.92 11-Sep-97
3429 ----------------------
3430
3431 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3432 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3433
3434 2. Caseless matching was not working in character classes if the characters in
3435 the pattern were in upper case.
3436
3437 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3438
3439 4. Make PCRE_ANCHORED public and accept as a compile option.
3440
3441 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3442 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3443 pass them.
3444
3445 6. Give an error if bad option bits passed at compile or run time.
3446
3447 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3448 pcretest to cause it to pass that flag.
3449
3450 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3451 options, and the first character, if set.
3452
3453 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3454
3455
3456 Version 0.91 10-Sep-97
3457 ----------------------
3458
3459 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3460 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3461
3462 2. PCRE was looping on encountering an indefinitely repeated back reference to
3463 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3464 Perl does - treats the match as successful.
3465
3466 ****

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