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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 379 - (show annotations) (download)
Mon Mar 2 20:30:05 2009 UTC (5 years, 8 months ago) by ph10
File size: 161600 byte(s)
Lock out empty string matches in pcregrep.

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

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