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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 376 - (show annotations) (download)
Sun Mar 1 12:00:59 2009 UTC (5 years, 6 months ago) by ph10
File size: 160904 byte(s)
Bug fixes: forgot to commit these individually.

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

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