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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 378 - (show annotations) (download)
Sun Mar 1 14:13:34 2009 UTC (5 years, 6 months ago) by ph10
File size: 161322 byte(s)
Make pcregrep with --colour show all matches in a line in colour.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.9 xx-xxx-09
5 ---------------------
6
7 1. When building with support for bzlib/zlib (pcregrep) and/or readline
8 (pcretest), all targets were linked against these libraries. This included
9 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10 libraries. This caused unwanted dependencies to be created. This problem
11 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12 pcretest is linked with readline.
13
14 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
15 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16 moved outside it again, because FALSE and TRUE are already defined in AIX,
17 but BOOL is not.
18
19 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20 PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21
22 4. The pcregrep documentation said spaces were inserted as well as colons (or
23 hyphens) following file names and line numbers when outputting matching
24 lines. This is not true; no spaces are inserted. I have also clarified the
25 wording for the --colour (or --color) option.
26
27 5. In pcregrep, when --colour was used with -o, the list of matching strings
28 was not coloured; this is different to GNU grep, so I have changed it to be
29 the same.
30
31 6. When --colo(u)r was used in pcregrep, only the first matching substring in
32 each matching line was coloured. Now it goes on to look for further matches
33 of any of the test patterns, which is the same behaviour as GNU grep.
34
35
36 Version 7.8 05-Sep-08
37 ---------------------
38
39 1. Replaced UCP searching code with optimized version as implemented for Ad
40 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
41 stage table and inline lookup instead of a function, giving speed ups of 2
42 to 5 times on some simple patterns that I tested. Permission was given to
43 distribute the MultiStage2.py script that generates the tables (it's not in
44 the tarball, but is in the Subversion repository).
45
46 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
47 scripts.
48
49 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
50 a group with a zero qualifier. The result of the study could be incorrect,
51 or the function might crash, depending on the pattern.
52
53 4. Caseless matching was not working for non-ASCII characters in back
54 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
55 It now works when Unicode Property Support is available.
56
57 5. In pcretest, an escape such as \x{de} in the data was always generating
58 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
59 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
60 truncation.
61
62 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
63
64 7. Added two (int) casts to pcregrep when printing the difference of two
65 pointers, in case they are 64-bit values.
66
67 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
68 test 2 if it fails.
69
70 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
71 and a #define of that name to empty if it is not externally set. This is to
72 allow users of MSVC to set it if necessary.
73
74 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
75 the convenience functions in the pcre_get.c source file.
76
77 11. An option change at the start of a pattern that had top-level alternatives
78 could cause overwriting and/or a crash. This command provoked a crash in
79 some environments:
80
81 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
82
83 This potential security problem was recorded as CVE-2008-2371.
84
85 12. For a pattern where the match had to start at the beginning or immediately
86 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
87 pcre_dfa_exec() could read past the end of the passed subject if there was
88 no match. To help with detecting such bugs (e.g. with valgrind), I modified
89 pcretest so that it places the subject at the end of its malloc-ed buffer.
90
91 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
92 exec() might read past the end of the data buffer in UTF-8 mode.
93
94 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
95 the data contained the byte 0x85 as part of a UTF-8 character within its
96 first line. This applied both to normal and DFA matching.
97
98 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
99 /^[^d]*?$/8 failed to match "abc".
100
101 16. Added a missing copyright notice to pcrecpp_internal.h.
102
103 17. Make it more clear in the documentation that values returned from
104 pcre_exec() in ovector are byte offsets, not character counts.
105
106 18. Tidied a few places to stop certain compilers from issuing warnings.
107
108 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
109 supplied by Stefan Weber. I made a further small update for 7.8 because
110 there is a change of source arrangements: the pcre_searchfuncs.c module is
111 replaced by pcre_ucd.c.
112
113
114 Version 7.7 07-May-08
115 ---------------------
116
117 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
118 a string to a long long, pretend we don't even have a long long." This is
119 done by checking for the strtoq, strtoll, and _strtoi64 functions.
120
121 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
122 pre-7.6 versions, which defined a global no_arg variable instead of putting
123 it in the RE class. (See also #8 below.)
124
125 3. Remove a line of dead code, identified by coverity and reported by Nuno
126 Lopes.
127
128 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
129
130 (1) The include/exclude patterns were being applied to the whole pathnames
131 of files, instead of just to the final components.
132
133 (2) If there was more than one level of directory, the subdirectories were
134 skipped unless they satisfied the include/exclude conditions. This is
135 inconsistent with GNU grep (and could even be seen as contrary to the
136 pcregrep specification - which I improved to make it absolutely clear).
137 The action now is always to scan all levels of directory, and just
138 apply the include/exclude patterns to regular files.
139
140 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
141 --exclude_dir in the tests to avoid scanning .svn directories.
142
143 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
144 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
145 doesn't support NULs in patterns.
146
147 7. Added some missing "const"s to declarations of static tables in
148 pcre_compile.c and pcre_dfa_exec.c.
149
150 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
151 caused by fix #2 above. (Subsequently also a second patch to fix the
152 first patch. And a third patch - this was a messy problem.)
153
154 9. Applied Craig's patch to remove the use of push_back().
155
156 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
157 matching function regexec().
158
159 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
160 which, however, unlike Perl's \g{...}, are subroutine calls, not back
161 references. PCRE supports relative numbers with this syntax (I don't think
162 Oniguruma does).
163
164 12. Previously, a group with a zero repeat such as (...){0} was completely
165 omitted from the compiled regex. However, this means that if the group
166 was called as a subroutine from elsewhere in the pattern, things went wrong
167 (an internal error was given). Such groups are now left in the compiled
168 pattern, with a new opcode that causes them to be skipped at execution
169 time.
170
171 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
172 to the way PCRE behaves:
173
174 (a) A lone ] character is dis-allowed (Perl treats it as data).
175
176 (b) A back reference to an unmatched subpattern matches an empty string
177 (Perl fails the current match path).
178
179 (c) A data ] in a character class must be notated as \] because if the
180 first data character in a class is ], it defines an empty class. (In
181 Perl it is not possible to have an empty class.) The empty class []
182 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
183 The negative empty class [^] matches any one character, independently
184 of the DOTALL setting.
185
186 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
187 non-existent subpattern following a character class starting with ']' and
188 containing () gave an internal compiling error instead of "reference to
189 non-existent subpattern". Fortunately, when the pattern did exist, the
190 compiled code was correct. (When scanning forwards to check for the
191 existencd of the subpattern, it was treating the data ']' as terminating
192 the class, so got the count wrong. When actually compiling, the reference
193 was subsequently set up correctly.)
194
195 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
196 it was being rejected as not supported by pcre_dfa_exec(), even though
197 other assertions are supported. I have made pcre_dfa_exec() support
198 (*FAIL).
199
200 16. The implementation of 13c above involved the invention of a new opcode,
201 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
202 cannot be changed at match time, I realized I could make a small
203 improvement to matching performance by compiling OP_ALLANY instead of
204 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
205 on the OP_ANY path.
206
207 17. Compiling pcretest on Windows with readline support failed without the
208 following two fixes: (1) Make the unistd.h include conditional on
209 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
210
211 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
212 ncurses library to be included for pcretest when ReadLine support is
213 requested, but also to allow for it to be overridden. This patch came from
214 Daniel Bergström.
215
216 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
217 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
218 any errors with the current Unicode tables. Thanks to Peter Kankowski for
219 spotting this.
220
221
222 Version 7.6 28-Jan-08
223 ---------------------
224
225 1. A character class containing a very large number of characters with
226 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
227 overflow.
228
229 2. Patch to cut out the "long long" test in pcrecpp_unittest when
230 HAVE_LONG_LONG is not defined.
231
232 3. Applied Christian Ehrlicher's patch to update the CMake build files to
233 bring them up to date and include new features. This patch includes:
234
235 - Fixed PH's badly added libz and libbz2 support.
236 - Fixed a problem with static linking.
237 - Added pcredemo. [But later removed - see 7 below.]
238 - Fixed dftables problem and added an option.
239 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
240 HAVE_LONG_LONG.
241 - Added readline support for pcretest.
242 - Added an listing of the option settings after cmake has run.
243
244 4. A user submitted a patch to Makefile that makes it easy to create
245 "pcre.dll" under mingw when using Configure/Make. I added stuff to
246 Makefile.am that cause it to include this special target, without
247 affecting anything else. Note that the same mingw target plus all
248 the other distribution libraries and programs are now supported
249 when configuring with CMake (see 6 below) instead of with
250 Configure/Make.
251
252 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
253 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
254 exported in the Windows port". It has not yet been confirmed that the patch
255 solves the problem, but it does no harm.
256
257 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
258 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
259 with CMake, and also correct the comment about stack recursion.
260
261 7. Remove the automatic building of pcredemo from the ./configure system and
262 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
263 of a program that users should build themselves after PCRE is installed, so
264 building it automatically is not really right. What is more, it gave
265 trouble in some build environments.
266
267 8. Further tidies to CMakeLists.txt from Sheri and Christian.
268
269
270 Version 7.5 10-Jan-08
271 ---------------------
272
273 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
274 values in parens when parsing an RE using the C++ wrapper."
275
276 2. Negative specials like \S did not work in character classes in UTF-8 mode.
277 Characters greater than 255 were excluded from the class instead of being
278 included.
279
280 3. The same bug as (2) above applied to negated POSIX classes such as
281 [:^space:].
282
283 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
284 defined or documented. It seems to have been a typo for PCRE_STATIC, so
285 I have changed it.
286
287 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
288 first named subpattern) and a construct such as (?&a) would reference the
289 first named subpattern whose name started with "a" (in other words, the
290 length check was missing). Both these problems are fixed. "Subpattern name
291 expected" is now given for (?&) (a zero-length name), and this patch also
292 makes it give the same error for \k'' (previously it complained that that
293 was a reference to a non-existent subpattern).
294
295 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
296 this is right because (?- can be followed by option settings as well as by
297 digits. I have, however, made the messages clearer.
298
299 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
300 than the number used in the conditional) now cause a compile-time error.
301 This is actually not compatible with Perl, which accepts such patterns, but
302 treats the conditional as always being FALSE (as PCRE used to), but it
303 seems to me that giving a diagnostic is better.
304
305 8. Change "alphameric" to the more common word "alphanumeric" in comments
306 and messages.
307
308 9. Fix two occurrences of "backslash" in comments that should have been
309 "backspace".
310
311 10. Remove two redundant lines of code that can never be obeyed (their function
312 was moved elsewhere).
313
314 11. The program that makes PCRE's Unicode character property table had a bug
315 which caused it to generate incorrect table entries for sequences of
316 characters that have the same character type, but are in different scripts.
317 It amalgamated them into a single range, with the script of the first of
318 them. In other words, some characters were in the wrong script. There were
319 thirteen such cases, affecting characters in the following ranges:
320
321 U+002b0 - U+002c1
322 U+0060c - U+0060d
323 U+0061e - U+00612
324 U+0064b - U+0065e
325 U+0074d - U+0076d
326 U+01800 - U+01805
327 U+01d00 - U+01d77
328 U+01d9b - U+01dbf
329 U+0200b - U+0200f
330 U+030fc - U+030fe
331 U+03260 - U+0327f
332 U+0fb46 - U+0fbb1
333 U+10450 - U+1049d
334
335 12. The -o option (show only the matching part of a line) for pcregrep was not
336 compatible with GNU grep in that, if there was more than one match in a
337 line, it showed only the first of them. It now behaves in the same way as
338 GNU grep.
339
340 13. If the -o and -v options were combined for pcregrep, it printed a blank
341 line for every non-matching line. GNU grep prints nothing, and pcregrep now
342 does the same. The return code can be used to tell if there were any
343 non-matching lines.
344
345 14. Added --file-offsets and --line-offsets to pcregrep.
346
347 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
348 infinitely looping recursion. The bug was that positive lookaheads were not
349 being skipped when checking for a possible empty match (negative lookaheads
350 and both kinds of lookbehind were skipped).
351
352 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
353 inclusion of <windows.h> to before rather than after the definition of
354 INVALID_FILE_ATTRIBUTES (patch from David Byron).
355
356 17. Specifying a possessive quantifier with a specific limit for a Unicode
357 character property caused pcre_compile() to compile bad code, which led at
358 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
359 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
360 caused the error; without that there was no problem.
361
362 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
363
364 19. Added --enable-pcretest-libreadline.
365
366 20. In pcrecpp.cc, the variable 'count' was incremented twice in
367 RE::GlobalReplace(). As a result, the number of replacements returned was
368 double what it should be. I removed one of the increments, but Craig sent a
369 later patch that removed the other one (the right fix) and added unit tests
370 that check the return values (which was not done before).
371
372 21. Several CMake things:
373
374 (1) Arranged that, when cmake is used on Unix, the libraries end up with
375 the names libpcre and libpcreposix, not just pcre and pcreposix.
376
377 (2) The above change means that pcretest and pcregrep are now correctly
378 linked with the newly-built libraries, not previously installed ones.
379
380 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
381
382 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
383 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
384 UTF-8 newline character). The key issue is that the pattern starts .*;
385 this means that the match must be either at the beginning, or after a
386 newline. The bug was in the code for advancing after a failed match and
387 checking that the new position followed a newline. It was not taking
388 account of UTF-8 characters correctly.
389
390 23. PCRE was behaving differently from Perl in the way it recognized POSIX
391 character classes. PCRE was not treating the sequence [:...:] as a
392 character class unless the ... were all letters. Perl, however, seems to
393 allow any characters between [: and :], though of course it rejects as
394 unknown any "names" that contain non-letters, because all the known class
395 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
396 for example, whereas PCRE did not - it did not recognize a POSIX character
397 class. This seemed a bit dangerous, so the code has been changed to be
398 closer to Perl. The behaviour is not identical to Perl, because PCRE will
399 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
400 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
401 Perl does, and where it didn't before.
402
403 24. Rewrite so as to remove the single use of %n from pcregrep because in some
404 Windows environments %n is disabled by default.
405
406
407 Version 7.4 21-Sep-07
408 ---------------------
409
410 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
411 means that a class such as [\s] counted as "explicit reference to CR or
412 LF". That isn't really right - the whole point of the change was to try to
413 help when there was an actual mention of one of the two characters. So now
414 the change happens only if \r or \n (or a literal CR or LF) character is
415 encountered.
416
417 2. The 32-bit options word was also used for 6 internal flags, but the numbers
418 of both had grown to the point where there were only 3 bits left.
419 Fortunately, there was spare space in the data structure, and so I have
420 moved the internal flags into a new 16-bit field to free up more option
421 bits.
422
423 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
424 but did not set the internal JCHANGED flag - either of these is enough to
425 control the way the "get" function works - but the PCRE_INFO_JCHANGED
426 facility is supposed to tell if (?J) was ever used, so now (?J) at the
427 start sets both bits.
428
429 4. Added options (at build time, compile time, exec time) to change \R from
430 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
431
432 5. doc/pcresyntax.html was missing from the distribution.
433
434 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
435 compatibility, even though it is no longer used.
436
437 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
438 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
439 windows.h file is present (where different names are used). [This was
440 reversed later after testing - see 16 below.]
441
442 8. Changed all #include <config.h> to #include "config.h". There were also
443 some further <pcre.h> cases that I changed to "pcre.h".
444
445 9. When pcregrep was used with the --colour option, it missed the line ending
446 sequence off the lines that it output.
447
448 10. It was pointed out to me that arrays of string pointers cause lots of
449 relocations when a shared library is dynamically loaded. A technique of
450 using a single long string with a table of offsets can drastically reduce
451 these. I have refactored PCRE in four places to do this. The result is
452 dramatic:
453
454 Originally: 290
455 After changing UCP table: 187
456 After changing error message table: 43
457 After changing table of "verbs" 36
458 After changing table of Posix names 22
459
460 Thanks to the folks working on Gregex for glib for this insight.
461
462 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
463 unicode-properties was also set.
464
465 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
466
467 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
468 checked only for CRLF.
469
470 14. Added casts to pcretest.c to avoid compiler warnings.
471
472 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
473
474 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
475 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
476 entirely. This removes changes made in 7 above.
477
478 17. The CMake files have been updated, and there is now more information about
479 building with CMake in the NON-UNIX-USE document.
480
481
482 Version 7.3 28-Aug-07
483 ---------------------
484
485 1. In the rejigging of the build system that eventually resulted in 7.1, the
486 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
487 brackets there is not right, since it causes compilers to look for an
488 installed pcre.h, not the version that is in the source that is being
489 compiled (which of course may be different). I have changed it back to:
490
491 #include "pcre.h"
492
493 I have a vague recollection that the change was concerned with compiling in
494 different directories, but in the new build system, that is taken care of
495 by the VPATH setting the Makefile.
496
497 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
498 when the subject happened to end in the byte 0x85 (e.g. if the last
499 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
500 characters but of course it shouldn't be taken as a newline when it is part
501 of another character. The bug was that, for an unlimited repeat of . in
502 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
503 characters when looking for a newline.
504
505 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
506
507 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
508 in debug output.
509
510 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
511 long printing in the pcrecpp unittest when running under MinGW.
512
513 6. ESC_K was left out of the EBCDIC table.
514
515 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
516 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
517 limit also applies to "virtual nesting" when a pattern is recursive, and in
518 this case 1000 isn't so big. I have been able to remove this limit at the
519 expense of backing off one optimization in certain circumstances. Normally,
520 when pcre_exec() would call its internal match() function recursively and
521 immediately return the result unconditionally, it uses a "tail recursion"
522 feature to save stack. However, when a subpattern that can match an empty
523 string has an unlimited repetition quantifier, it no longer makes this
524 optimization. That gives it a stack frame in which to save the data for
525 checking that an empty string has been matched. Previously this was taken
526 from the 1000-entry workspace that had been reserved. So now there is no
527 explicit limit, but more stack is used.
528
529 8. Applied Daniel's patches to solve problems with the import/export magic
530 syntax that is required for Windows, and which was going wrong for the
531 pcreposix and pcrecpp parts of the library. These were overlooked when this
532 problem was solved for the main library.
533
534 9. There were some crude static tests to avoid integer overflow when computing
535 the size of patterns that contain repeated groups with explicit upper
536 limits. As the maximum quantifier is 65535, the maximum group length was
537 set at 30,000 so that the product of these two numbers did not overflow a
538 32-bit integer. However, it turns out that people want to use groups that
539 are longer than 30,000 bytes (though not repeat them that many times).
540 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
541 made it possible to implement the integer overflow checks in a much more
542 dynamic way, which I have now done. The artificial limitation on group
543 length has been removed - we now have only the limit on the total length of
544 the compiled pattern, which depends on the LINK_SIZE setting.
545
546 10. Fixed a bug in the documentation for get/copy named substring when
547 duplicate names are permitted. If none of the named substrings are set, the
548 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
549 empty string.
550
551 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
552 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
553 because the ] is interpreted as the first data character and the
554 terminating ] is not found. PCRE has been made compatible with Perl in this
555 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
556 cause memory overwriting.
557
558 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
559 string has been matched (to stop an infinite loop). It was not recognizing
560 a conditional subpattern that could match an empty string if that
561 subpattern was within another subpattern. For example, it looped when
562 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
563 condition was not nested. This bug has been fixed.
564
565 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
566 past the start of the subject in the presence of bytes with the top bit
567 set, for example "\x8aBCD".
568
569 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
570 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
571
572 14. Optimized (?!) to (*FAIL).
573
574 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
575 This restricts code points to be within the range 0 to 0x10FFFF, excluding
576 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
577 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
578 does: it's just the validity check that is more restrictive.
579
580 16. Inserted checks for integer overflows during escape sequence (backslash)
581 processing, and also fixed erroneous offset values for syntax errors during
582 backslash processing.
583
584 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
585 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
586
587 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
588 caused an overrun.
589
590 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
591 something other than just ASCII characters) inside a group that had an
592 unlimited repeat caused a loop at compile time (while checking to see
593 whether the group could match an empty string).
594
595 20. Debugging a pattern containing \p or \P could cause a crash. For example,
596 [\P{Any}] did so. (Error in the code for printing property names.)
597
598 21. An orphan \E inside a character class could cause a crash.
599
600 22. A repeated capturing bracket such as (A)? could cause a wild memory
601 reference during compilation.
602
603 23. There are several functions in pcre_compile() that scan along a compiled
604 expression for various reasons (e.g. to see if it's fixed length for look
605 behind). There were bugs in these functions when a repeated \p or \P was
606 present in the pattern. These operators have additional parameters compared
607 with \d, etc, and these were not being taken into account when moving along
608 the compiled data. Specifically:
609
610 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
611 length.
612
613 (b) An item such as \pL+ within a repeated group could cause crashes or
614 loops.
615
616 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
617 "reference to non-existent subpattern" error.
618
619 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
620
621 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
622 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
623
624 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
625
626 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
627 character were causing crashes (broken optimization).
628
629 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
630 \p or \P) caused a compile-time loop.
631
632 28. More problems have arisen in unanchored patterns when CRLF is a valid line
633 break. For example, the unstudied pattern [\r\n]A does not match the string
634 "\r\nA" because change 7.0/46 below moves the current point on by two
635 characters after failing to match at the start. However, the pattern \nA
636 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
637 the same is true. There doesn't seem any very clean way out of this, but
638 what I have chosen to do makes the common cases work: PCRE now takes note
639 of whether there can be an explicit match for \r or \n anywhere in the
640 pattern, and if so, 7.0/46 no longer applies. As part of this change,
641 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
642 pattern has explicit CR or LF references.
643
644 29. Added (*CR) etc for changing newline setting at start of pattern.
645
646
647 Version 7.2 19-Jun-07
648 ---------------------
649
650 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
651 which is apparently normally available under Windows.
652
653 2. Re-jig the pcregrep tests with different newline settings in an attempt
654 to make them independent of the local environment's newline setting.
655
656 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
657
658 4. Some of the "internals" tests were previously cut out when the link size
659 was not 2, because the output contained actual offsets. The recent new
660 "Z" feature of pcretest means that these can be cut out, making the tests
661 usable with all link sizes.
662
663 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
664 stack recursion. This gives a massive performance boost under BSD, but just
665 a small improvement under Linux. However, it saves one field in the frame
666 in all cases.
667
668 6. Added more features from the forthcoming Perl 5.10:
669
670 (a) (?-n) (where n is a string of digits) is a relative subroutine or
671 recursion call. It refers to the nth most recently opened parentheses.
672
673 (b) (?+n) is also a relative subroutine call; it refers to the nth next
674 to be opened parentheses.
675
676 (c) Conditions that refer to capturing parentheses can be specified
677 relatively, for example, (?(-2)... or (?(+3)...
678
679 (d) \K resets the start of the current match so that everything before
680 is not part of it.
681
682 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
683
684 (f) \g{name} is another synonym - part of Perl 5.10's unification of
685 reference syntax.
686
687 (g) (?| introduces a group in which the numbering of parentheses in each
688 alternative starts with the same number.
689
690 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
691
692 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
693 PCRE_INFO_JCHANGED.
694
695 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
696 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
697 for detecting groups that can match an empty string.
698
699 9. A pattern with a very large number of alternatives (more than several
700 hundred) was running out of internal workspace during the pre-compile
701 phase, where pcre_compile() figures out how much memory will be needed. A
702 bit of new cunning has reduced the workspace needed for groups with
703 alternatives. The 1000-alternative test pattern now uses 12 bytes of
704 workspace instead of running out of the 4096 that are available.
705
706 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
707
708 11. Applied patch from Google to remove an optimization that didn't quite work.
709 The report of the bug said:
710
711 pcrecpp::RE("a*").FullMatch("aaa") matches, while
712 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
713 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
714
715 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
716 it matched the wrong number of bytes.
717
718
719 Version 7.1 24-Apr-07
720 ---------------------
721
722 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
723 that is more "standard", making use of automake and other Autotools. There
724 is some re-arrangement of the files and adjustment of comments consequent
725 on this.
726
727 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
728 for recursive directory scanning broke on some systems because the files
729 are not scanned in any specific order and on different systems the order
730 was different. A call to "sort" has been inserted into RunGrepTest for the
731 approprate test as a short-term fix. In the longer term there may be an
732 alternative.
733
734 3. I had an email from Eric Raymond about problems translating some of PCRE's
735 man pages to HTML (despite the fact that I distribute HTML pages, some
736 people do their own conversions for various reasons). The problems
737 concerned the use of low-level troff macros .br and .in. I have therefore
738 removed all such uses from the man pages (some were redundant, some could
739 be replaced by .nf/.fi pairs). The 132html script that I use to generate
740 HTML has been updated to handle .nf/.fi and to complain if it encounters
741 .br or .in.
742
743 4. Updated comments in configure.ac that get placed in config.h.in and also
744 arranged for config.h to be included in the distribution, with the name
745 config.h.generic, for the benefit of those who have to compile without
746 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
747
748 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
749 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
750 makevp.bat for the current PCRE, using the additional files
751 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
752
753 6. A Windows user reported a minor discrepancy with test 2, which turned out
754 to be caused by a trailing space on an input line that had got lost in his
755 copy. The trailing space was an accident, so I've just removed it.
756
757 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
758 that is needed.
759
760 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
761 as "const" (a) because they are and (b) because it helps the PHP
762 maintainers who have recently made a script to detect big data structures
763 in the php code that should be moved to the .rodata section. I remembered
764 to update Builducptable as well, so it won't revert if ucptable.h is ever
765 re-created.
766
767 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
768 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
769 order to be able to cut out the UTF-8 tables in the latter when UTF-8
770 support is not required. This saves 1.5-2K of code, which is important in
771 some applications.
772
773 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
774 so as not to refer to the tables, even though these functions will never be
775 called when UTF-8 support is disabled. Otherwise there are problems with a
776 shared library.
777
778 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
779
780 (a) It was defining its arguments as char * instead of void *.
781
782 (b) It was assuming that all moves were upwards in memory; this was true
783 a long time ago when I wrote it, but is no longer the case.
784
785 The emulated memove() is provided for those environments that have neither
786 memmove() nor bcopy(). I didn't think anyone used it these days, but that
787 is clearly not the case, as these two bugs were recently reported.
788
789 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
790 and Detrail to create the HTML documentation, the .txt form of the man
791 pages, and it removes trailing spaces from listed files. It also creates
792 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
793 case, it wraps all the #defines with #ifndefs. This script should be run
794 before "make dist".
795
796 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
797 with Unicode property support.
798
799 (a) For a maximizing quantifier, if the two different cases of the
800 character were of different lengths in their UTF-8 codings (there are
801 some cases like this - I found 11), and the matching function had to
802 back up over a mixture of the two cases, it incorrectly assumed they
803 were both the same length.
804
805 (b) When PCRE was configured to use the heap rather than the stack for
806 recursion during matching, it was not correctly preserving the data for
807 the other case of a UTF-8 character when checking ahead for a match
808 while processing a minimizing repeat. If the check also involved
809 matching a wide character, but failed, corruption could cause an
810 erroneous result when trying to check for a repeat of the original
811 character.
812
813 13. Some tidying changes to the testing mechanism:
814
815 (a) The RunTest script now detects the internal link size and whether there
816 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
817 values substituted by "configure". (The RunGrepTest script already did
818 this for UTF-8.) The configure.ac script no longer substitutes the
819 relevant variables.
820
821 (b) The debugging options /B and /D in pcretest show the compiled bytecode
822 with length and offset values. This means that the output is different
823 for different internal link sizes. Test 2 is skipped for link sizes
824 other than 2 because of this, bypassing the problem. Unfortunately,
825 there was also a test in test 3 (the locale tests) that used /B and
826 failed for link sizes other than 2. Rather than cut the whole test out,
827 I have added a new /Z option to pcretest that replaces the length and
828 offset values with spaces. This is now used to make test 3 independent
829 of link size. (Test 2 will be tidied up later.)
830
831 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
832 segmentation fault instead of returning the appropriate error message.
833
834 15. In multiline mode when the newline sequence was set to "any", the pattern
835 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
836 This doesn't seem right; it now treats the CRLF combination as the line
837 ending, and so does not match in that case. It's only a pattern such as ^$
838 that would hit this one: something like ^ABC$ would have failed after \r
839 and then tried again after \r\n.
840
841 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
842 in an attempt to make files that differ only in their line terminators
843 compare equal. This works on Linux.
844
845 17. Under certain error circumstances pcregrep might try to free random memory
846 as it exited. This is now fixed, thanks to valgrind.
847
848 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
849 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
850 was because its rules for how to advance for /g after matching an empty
851 string at the end of a line did not allow for this case. They now check for
852 it specially.
853
854 20. pcretest is supposed to handle patterns and data of any length, by
855 extending its buffers when necessary. It was getting this wrong when the
856 buffer for a data line had to be extended.
857
858 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
859 CRLF as a newline sequence.
860
861 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
862 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
863 I have nevertheless tidied it up.
864
865 23. Added some casts to kill warnings from HP-UX ia64 compiler.
866
867 24. Added a man page for pcre-config.
868
869
870 Version 7.0 19-Dec-06
871 ---------------------
872
873 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
874 moving to gcc 4.1.1.
875
876 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
877 sys/time.h, which is documented as needed for this function. It doesn't
878 seem to matter on Linux, but it showed up on some releases of OS X.
879
880 3. It seems that there are systems where bytes whose values are greater than
881 127 match isprint() in the "C" locale. The "C" locale should be the
882 default when a C program starts up. In most systems, only ASCII printing
883 characters match isprint(). This difference caused the output from pcretest
884 to vary, making some of the tests fail. I have changed pcretest so that:
885
886 (a) When it is outputting text in the compiled version of a pattern, bytes
887 other than 32-126 are always shown as hex escapes.
888
889 (b) When it is outputting text that is a matched part of a subject string,
890 it does the same, unless a different locale has been set for the match
891 (using the /L modifier). In this case, it uses isprint() to decide.
892
893 4. Fixed a major bug that caused incorrect computation of the amount of memory
894 required for a compiled pattern when options that changed within the
895 pattern affected the logic of the preliminary scan that determines the
896 length. The relevant options are -x, and -i in UTF-8 mode. The result was
897 that the computed length was too small. The symptoms of this bug were
898 either the PCRE error "internal error: code overflow" from pcre_compile(),
899 or a glibc crash with a message such as "pcretest: free(): invalid next
900 size (fast)". Examples of patterns that provoked this bug (shown in
901 pcretest format) are:
902
903 /(?-x: )/x
904 /(?x)(?-x: \s*#\s*)/
905 /((?i)[\x{c0}])/8
906 /(?i:[\x{c0}])/8
907
908 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
909 is now done differently.
910
911 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
912 wrapper classes; (b) implement a new function in the C++ scanner that is
913 more efficient than the old way of doing things because it avoids levels of
914 recursion in the regex matching; (c) add a paragraph to the documentation
915 for the FullMatch() function.
916
917 6. The escape sequence \n was being treated as whatever was defined as
918 "newline". Not only was this contrary to the documentation, which states
919 that \n is character 10 (hex 0A), but it also went horribly wrong when
920 "newline" was defined as CRLF. This has been fixed.
921
922 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
923 was being set to -1 for the "end of line" case (supposedly a value that no
924 character can have). Though this value is never used (the check for end of
925 line is "zero bytes in current character"), it caused compiler complaints.
926 I've changed it to 0xffffffff.
927
928 8. In pcre_version.c, the version string was being built by a sequence of
929 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
930 string (as it is for production releases) called a macro with an empty
931 argument. The C standard says the result of this is undefined. The gcc
932 compiler treats it as an empty string (which was what was wanted) but it is
933 reported that Visual C gives an error. The source has been hacked around to
934 avoid this problem.
935
936 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
937 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
938 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
939 of them did).
940
941 10. Originally, pcretest opened its input and output without "b"; then I was
942 told that "b" was needed in some environments, so it was added for release
943 5.0 to both the input and output. (It makes no difference on Unix-like
944 systems.) Later I was told that it is wrong for the input on Windows. I've
945 now abstracted the modes into two macros, to make it easier to fiddle with
946 them, and removed "b" from the input mode under Windows.
947
948 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
949
950 12. Added -help and --help to pcretest as an official way of being reminded
951 of the options.
952
953 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
954 and pcrecpp.cc because they annoy compilers at high warning levels.
955
956 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
957
958 15. Fixed an occurrence of == in configure.ac that should have been = (shell
959 scripts are not C programs :-) and which was not noticed because it works
960 on Linux.
961
962 16. pcretest is supposed to handle any length of pattern and data line (as one
963 line or as a continued sequence of lines) by extending its input buffer if
964 necessary. This feature was broken for very long pattern lines, leading to
965 a string of junk being passed to pcre_compile() if the pattern was longer
966 than about 50K.
967
968 17. I have done a major re-factoring of the way pcre_compile() computes the
969 amount of memory needed for a compiled pattern. Previously, there was code
970 that made a preliminary scan of the pattern in order to do this. That was
971 OK when PCRE was new, but as the facilities have expanded, it has become
972 harder and harder to keep it in step with the real compile phase, and there
973 have been a number of bugs (see for example, 4 above). I have now found a
974 cunning way of running the real compile function in a "fake" mode that
975 enables it to compute how much memory it would need, while actually only
976 ever using a few hundred bytes of working memory and without too many
977 tests of the mode. This should make future maintenance and development
978 easier. A side effect of this work is that the limit of 200 on the nesting
979 depth of parentheses has been removed (though this was never a serious
980 limitation, I suspect). However, there is a downside: pcre_compile() now
981 runs more slowly than before (30% or more, depending on the pattern). I
982 hope this isn't a big issue. There is no effect on runtime performance.
983
984 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
985 newline (only possible for the last line of a file) and it was a
986 pattern that set a locale (followed by /Lsomething), pcretest crashed.
987
988 19. Added additional timing features to pcretest. (1) The -tm option now times
989 matching only, not compiling. (2) Both -t and -tm can be followed, as a
990 separate command line item, by a number that specifies the number of
991 repeats to use when timing. The default is 50000; this gives better
992 precision, but takes uncomfortably long for very large patterns.
993
994 20. Extended pcre_study() to be more clever in cases where a branch of a
995 subpattern has no definite first character. For example, (a*|b*)[cd] would
996 previously give no result from pcre_study(). Now it recognizes that the
997 first character must be a, b, c, or d.
998
999 21. There was an incorrect error "recursive call could loop indefinitely" if
1000 a subpattern (or the entire pattern) that was being tested for matching an
1001 empty string contained only one non-empty item after a nested subpattern.
1002 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1003 incorrectly, because the \d was being skipped in the check.
1004
1005 22. The pcretest program now has a new pattern option /B and a command line
1006 option -b, which is equivalent to adding /B to every pattern. This causes
1007 it to show the compiled bytecode, without the additional information that
1008 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1009 is the same as /B/I).
1010
1011 23. A new optimization is now able automatically to treat some sequences such
1012 as a*b as a*+b. More specifically, if something simple (such as a character
1013 or a simple class like \d) has an unlimited quantifier, and is followed by
1014 something that cannot possibly match the quantified thing, the quantifier
1015 is automatically "possessified".
1016
1017 24. A recursive reference to a subpattern whose number was greater than 39
1018 went wrong under certain circumstances in UTF-8 mode. This bug could also
1019 have affected the operation of pcre_study().
1020
1021 25. Realized that a little bit of performance could be had by replacing
1022 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1023
1024 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1025
1026 27. Possessive quantifiers such as a++ were previously implemented by turning
1027 them into atomic groups such as ($>a+). Now they have their own opcodes,
1028 which improves performance. This includes the automatically created ones
1029 from 23 above.
1030
1031 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1032 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1033 the first matched character to be a colon. This applied both to named and
1034 numbered groups.
1035
1036 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1037
1038 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1039 building PCRE on Windows easier, so I have included it in the distribution.
1040
1041 31. There is now a check in pcretest against a ridiculously large number being
1042 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1043 loop, the loop is abandoned.
1044
1045 32. Forward references to subpatterns in conditions such as (?(2)...) where
1046 subpattern 2 is defined later cause pcre_compile() to search forwards in
1047 the pattern for the relevant set of parentheses. This search went wrong
1048 when there were unescaped parentheses in a character class, parentheses
1049 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1050
1051 33. "Subroutine" calls and backreferences were previously restricted to
1052 referencing subpatterns earlier in the regex. This restriction has now
1053 been removed.
1054
1055 34. Added a number of extra features that are going to be in Perl 5.10. On the
1056 whole, these are just syntactic alternatives for features that PCRE had
1057 previously implemented using the Python syntax or my own invention. The
1058 other formats are all retained for compatibility.
1059
1060 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1061 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1062 also .NET compatible.
1063
1064 (b) A recursion or subroutine call to a named group can now be defined as
1065 (?&name) as well as (?P>name).
1066
1067 (c) A backreference to a named group can now be defined as \k<name> or
1068 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1069 5.10, are also .NET compatible.
1070
1071 (d) A conditional reference to a named group can now use the syntax
1072 (?(<name>) or (?('name') as well as (?(name).
1073
1074 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1075 groups (named and numbered) that are never evaluated inline, but can be
1076 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1077 is always false. There may be only one alternative in such a group.
1078
1079 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1080 as the simple (?(R). The condition is true only if the most recent
1081 recursion is that of the given number or name. It does not search out
1082 through the entire recursion stack.
1083
1084 (g) The escape \gN or \g{N} has been added, where N is a positive or
1085 negative number, specifying an absolute or relative reference.
1086
1087 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1088 some "unreachable code" warnings.
1089
1090 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1091 things, this adds five new scripts.
1092
1093 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1094 There were also incompatibilities regarding the handling of \Q..\E inside
1095 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1096 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1097
1098 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1099 matches an empty string, and forcibly breaks the loop. There were bugs in
1100 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1101 against aaaa the result was just "a" rather than "aaaa", for example. Two
1102 separate and independent bugs (that affected different cases) have been
1103 fixed.
1104
1105 39. Refactored the code to abolish the use of different opcodes for small
1106 capturing bracket numbers. This is a tidy that I avoided doing when I
1107 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1108 The new approach is not only tidier, it makes it possible to reduce the
1109 memory needed to fix the previous bug (38).
1110
1111 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1112 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1113 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1114 mode.
1115
1116 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1117 report.
1118
1119 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1120 copy construction and assignment in the C++ wrapper.
1121
1122 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1123 couple of bugs that could have given wrong results in the "--newline=crlf"
1124 case.
1125
1126 44. Added a number of casts and did some reorganization of signed/unsigned int
1127 variables following suggestions from Dair Grant. Also renamed the variable
1128 "this" as "item" because it is a C++ keyword.
1129
1130 45. Arranged for dftables to add
1131
1132 #include "pcre_internal.h"
1133
1134 to pcre_chartables.c because without it, gcc 4.x may remove the array
1135 definition from the final binary if PCRE is built into a static library and
1136 dead code stripping is activated.
1137
1138 46. For an unanchored pattern, if a match attempt fails at the start of a
1139 newline sequence, and the newline setting is CRLF or ANY, and the next two
1140 characters are CRLF, advance by two characters instead of one.
1141
1142
1143 Version 6.7 04-Jul-06
1144 ---------------------
1145
1146 1. In order to handle tests when input lines are enormously long, pcretest has
1147 been re-factored so that it automatically extends its buffers when
1148 necessary. The code is crude, but this _is_ just a test program. The
1149 default size has been increased from 32K to 50K.
1150
1151 2. The code in pcre_study() was using the value of the re argument before
1152 testing it for NULL. (Of course, in any sensible call of the function, it
1153 won't be NULL.)
1154
1155 3. The memmove() emulation function in pcre_internal.h, which is used on
1156 systems that lack both memmove() and bcopy() - that is, hardly ever -
1157 was missing a "static" storage class specifier.
1158
1159 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1160 containing an extended class (one that cannot be represented by a bitmap
1161 because it contains high-valued characters or Unicode property items, e.g.
1162 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1163 pattern, but PCRE should not loop if you do not (it no longer does).
1164 [Detail: two cases were found: (a) a repeated subpattern containing an
1165 extended class; (b) a recursive reference to a subpattern that followed a
1166 previous extended class. It wasn't skipping over the extended class
1167 correctly when UTF-8 mode was not set.]
1168
1169 5. A negated single-character class was not being recognized as fixed-length
1170 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1171 compile error "lookbehind assertion is not fixed length".
1172
1173 6. The RunPerlTest auxiliary script was showing an unexpected difference
1174 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1175 write a Perl script that can interpret lines of an input file either as
1176 byte characters or as UTF-8, which is what "perltest" was being required to
1177 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1178 can't do is switch easily at run time between having the "use utf8;" pragma
1179 or not. In the end, I fudged it by using the RunPerlTest script to insert
1180 "use utf8;" explicitly for the UTF-8 tests.
1181
1182 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1183 the end of the subject string, contrary to the documentation and to what
1184 Perl does. This was true of both matching functions. Now it matches only at
1185 the start of the subject and immediately after *internal* newlines.
1186
1187 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1188 a pointer to an int instead of a pointer to an unsigned long int. This
1189 caused problems on 64-bit systems.
1190
1191 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1192 instance of the 'standard' template library not being so standard".
1193
1194 10. There was no check on the number of named subpatterns nor the maximum
1195 length of a subpattern name. The product of these values is used to compute
1196 the size of the memory block for a compiled pattern. By supplying a very
1197 long subpattern name and a large number of named subpatterns, the size
1198 computation could be caused to overflow. This is now prevented by limiting
1199 the length of names to 32 characters, and the number of named subpatterns
1200 to 10,000.
1201
1202 11. Subpatterns that are repeated with specific counts have to be replicated in
1203 the compiled pattern. The size of memory for this was computed from the
1204 length of the subpattern and the repeat count. The latter is limited to
1205 65535, but there was no limit on the former, meaning that integer overflow
1206 could in principle occur. The compiled length of a repeated subpattern is
1207 now limited to 30,000 bytes in order to prevent this.
1208
1209 12. Added the optional facility to have named substrings with the same name.
1210
1211 13. Added the ability to use a named substring as a condition, using the
1212 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1213 are numbers (not recommended). Forward references are permitted.
1214
1215 14. Added forward references in named backreferences (if you see what I mean).
1216
1217 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1218 pattern could run off the end of the subject. For example, the pattern
1219 "(?s)(.{1,5})"8 did this with the subject "ab".
1220
1221 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1222 PCRE_CASELESS was set when matching characters that were quantified with ?
1223 or *.
1224
1225 17. A character class other than a single negated character that had a minimum
1226 but no maximum quantifier - for example [ab]{6,} - was not handled
1227 correctly by pce_dfa_exec(). It would match only one character.
1228
1229 18. A valid (though odd) pattern that looked like a POSIX character
1230 class but used an invalid character after [ (for example [[,abc,]]) caused
1231 pcre_compile() to give the error "Failed: internal error: code overflow" or
1232 in some cases to crash with a glibc free() error. This could even happen if
1233 the pattern terminated after [[ but there just happened to be a sequence of
1234 letters, a binary zero, and a closing ] in the memory that followed.
1235
1236 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1237 over the years. Originally (before any Unicode support), just the bottom 8
1238 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1239 output from "man perlunicode" includes this:
1240
1241 The regular expression compiler produces polymorphic opcodes. That
1242 is, the pattern adapts to the data and automatically switches to
1243 the Unicode character scheme when presented with Unicode data--or
1244 instead uses a traditional byte scheme when presented with byte
1245 data.
1246
1247 Sadly, a wide octal escape does not cause a switch, and in a string with
1248 no other multibyte characters, these octal escapes are treated as before.
1249 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1250 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1251 Unicode string.
1252
1253 I have not perpetrated such confusion in PCRE. Up till now, it took just
1254 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1255 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1256 translate to the appropriate multibyte character.
1257
1258 29. Applied some refactoring to reduce the number of warnings from Microsoft
1259 and Borland compilers. This has included removing the fudge introduced
1260 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1261 a warning about an unused variable.
1262
1263 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1264 characters since release 4.0, because Perl (from release 5.004) does not.
1265 [Or at least, is documented not to: some releases seem to be in conflict
1266 with the documentation.] However, when a pattern was studied with
1267 pcre_study() and all its branches started with \s, PCRE still included VT
1268 as a possible starting character. Of course, this did no harm; it just
1269 caused an unnecessary match attempt.
1270
1271 22. Removed a now-redundant internal flag bit that recorded the fact that case
1272 dependency changed within the pattern. This was once needed for "required
1273 byte" processing, but is no longer used. This recovers a now-scarce options
1274 bit. Also moved the least significant internal flag bit to the most-
1275 significant bit of the word, which was not previously used (hangover from
1276 the days when it was an int rather than a uint) to free up another bit for
1277 the future.
1278
1279 23. Added support for CRLF line endings as well as CR and LF. As well as the
1280 default being selectable at build time, it can now be changed at runtime
1281 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1282 specify that it is scanning data with non-default line endings.
1283
1284 24. Changed the definition of CXXLINK to make it agree with the definition of
1285 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1286
1287 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1288 recursions. This makes a big different to stack usage for some patterns.
1289
1290 26. If a subpattern containing a named recursion or subroutine reference such
1291 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1292 the space required for the compiled pattern went wrong and gave too small a
1293 value. Depending on the environment, this could lead to "Failed: internal
1294 error: code overflow at offset 49" or "glibc detected double free or
1295 corruption" errors.
1296
1297 27. Applied patches from Google (a) to support the new newline modes and (b) to
1298 advance over multibyte UTF-8 characters in GlobalReplace.
1299
1300 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1301 difference for some implementation of PCRE in some Windows version.
1302
1303 29. Added some extra testing facilities to pcretest:
1304
1305 \q<number> in a data line sets the "match limit" value
1306 \Q<number> in a data line sets the "match recursion limt" value
1307 -S <number> sets the stack size, where <number> is in megabytes
1308
1309 The -S option isn't available for Windows.
1310
1311
1312 Version 6.6 06-Feb-06
1313 ---------------------
1314
1315 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1316 in pcreposix.h. I have copied the definition from pcre.h.
1317
1318 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1319 because pcre.h is no longer a built file.
1320
1321 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1322 not normally included in the compiled code.
1323
1324
1325 Version 6.5 01-Feb-06
1326 ---------------------
1327
1328 1. When using the partial match feature with pcre_dfa_exec(), it was not
1329 anchoring the second and subsequent partial matches at the new starting
1330 point. This could lead to incorrect results. For example, with the pattern
1331 /1234/, partially matching against "123" and then "a4" gave a match.
1332
1333 2. Changes to pcregrep:
1334
1335 (a) All non-match returns from pcre_exec() were being treated as failures
1336 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1337 error message is output. Some extra information is given for the
1338 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1339 probably the only errors that are likely to be caused by users (by
1340 specifying a regex that has nested indefinite repeats, for instance).
1341 If there are more than 20 of these errors, pcregrep is abandoned.
1342
1343 (b) A binary zero was treated as data while matching, but terminated the
1344 output line if it was written out. This has been fixed: binary zeroes
1345 are now no different to any other data bytes.
1346
1347 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1348 used to set a locale for matching. The --locale=xxxx long option has
1349 been added (no short equivalent) to specify a locale explicitly on the
1350 pcregrep command, overriding the environment variables.
1351
1352 (d) When -B was used with -n, some line numbers in the output were one less
1353 than they should have been.
1354
1355 (e) Added the -o (--only-matching) option.
1356
1357 (f) If -A or -C was used with -c (count only), some lines of context were
1358 accidentally printed for the final match.
1359
1360 (g) Added the -H (--with-filename) option.
1361
1362 (h) The combination of options -rh failed to suppress file names for files
1363 that were found from directory arguments.
1364
1365 (i) Added the -D (--devices) and -d (--directories) options.
1366
1367 (j) Added the -F (--fixed-strings) option.
1368
1369 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1370
1371 (l) Added the --colo(u)r option.
1372
1373 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1374 is not present by default.
1375
1376 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1377 items such as (?R) or (?1), when the recursion could match a number of
1378 alternatives. If it matched one of the alternatives, but subsequently,
1379 outside the recursion, there was a failure, the code tried to back up into
1380 the recursion. However, because of the way PCRE is implemented, this is not
1381 possible, and the result was an incorrect result from the match.
1382
1383 In order to prevent this happening, the specification of recursion has
1384 been changed so that all such subpatterns are automatically treated as
1385 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1386
1387 4. I had overlooked the fact that, in some locales, there are characters for
1388 which isalpha() is true but neither isupper() nor islower() are true. In
1389 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1390 and ordfeminine) are like this. This affected the treatment of \w and \W
1391 when they appeared in character classes, but not when they appeared outside
1392 a character class. The bit map for "word" characters is now created
1393 separately from the results of isalnum() instead of just taking it from the
1394 upper, lower, and digit maps. (Plus the underscore character, of course.)
1395
1396 5. The above bug also affected the handling of POSIX character classes such as
1397 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1398 permanent tables. Instead, the bit maps for such a class were previously
1399 created as the appropriate unions of the upper, lower, and digit bitmaps.
1400 Now they are created by subtraction from the [[:word:]] class, which has
1401 its own bitmap.
1402
1403 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1404 It is created by subtracting the vertical space characters (\x09, \x0a,
1405 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1406 subtraction was done in the overall bitmap for a character class, meaning
1407 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1408 be recognized. This bug has been fixed.
1409
1410 7. Patches from the folks at Google:
1411
1412 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1413 real life, but is still worth protecting against".
1414
1415 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1416 regular expressions".
1417
1418 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1419 have it.
1420
1421 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1422 "configure" and the latter not, in order to fix a problem somebody had
1423 with compiling the Arg class on HP-UX.
1424
1425 (e) Improve the error-handling of the C++ wrapper a little bit.
1426
1427 (f) New tests for checking recursion limiting.
1428
1429 8. The pcre_memmove() function, which is used only if the environment does not
1430 have a standard memmove() function (and is therefore rarely compiled),
1431 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1432 returning a result (though PCRE never actually uses the result).
1433
1434 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1435 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1436 returned instead of calling malloc() with an overflowing number that would
1437 most likely cause subsequent chaos.
1438
1439 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1440
1441 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1442 with this option is matched, the nmatch and pmatch options of regexec() are
1443 ignored.
1444
1445 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1446 provided in case anyone wants to the the POSIX interface with UTF-8
1447 strings.
1448
1449 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1450 C++ linking (needed for some HP-UX environments).
1451
1452 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1453 (unused parameter) and in the pcre_printint() function (omitted "default"
1454 switch label when the default is to do nothing).
1455
1456 15. Added some code to make it possible, when PCRE is compiled as a C++
1457 library, to replace subject pointers for pcre_exec() with a smart pointer
1458 class, thus making it possible to process discontinuous strings.
1459
1460 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1461 much the same function. They were added by different people who were trying
1462 to make PCRE easy to compile on non-Unix systems. It has been suggested
1463 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1464 for compiling on Windows systems. I have therefore replaced it with
1465 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1466 defaults to "extern" for C or "extern C" for C++, which works fine on
1467 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1468 SCOPE with something explicit in config.h. In addition:
1469
1470 (a) pcreposix.h still had just "extern" instead of either of these macros;
1471 I have replaced it with PCRE_DATA_SCOPE.
1472
1473 (b) Functions such as _pcre_xclass(), which are internal to the library,
1474 but external in the C sense, all had PCRE_EXPORT in their definitions.
1475 This is apparently wrong for the Windows case, so I have removed it.
1476 (It makes no difference on Unix-like systems.)
1477
1478 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1479 of recursive calls to match(). This is different to MATCH_LIMIT because
1480 that limits the total number of calls to match(), not all of which increase
1481 the depth of recursion. Limiting the recursion depth limits the amount of
1482 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1483 when PCRE is compiled, and changed at run time. A patch from Google adds
1484 this functionality to the C++ interface.
1485
1486 18. Changes to the handling of Unicode character properties:
1487
1488 (a) Updated the table to Unicode 4.1.0.
1489
1490 (b) Recognize characters that are not in the table as "Cn" (undefined).
1491
1492 (c) I revised the way the table is implemented to a much improved format
1493 which includes recognition of ranges. It now supports the ranges that
1494 are defined in UnicodeData.txt, and it also amalgamates other
1495 characters into ranges. This has reduced the number of entries in the
1496 table from around 16,000 to around 3,000, thus reducing its size
1497 considerably. I realized I did not need to use a tree structure after
1498 all - a binary chop search is just as efficient. Having reduced the
1499 number of entries, I extended their size from 6 bytes to 8 bytes to
1500 allow for more data.
1501
1502 (d) Added support for Unicode script names via properties such as \p{Han}.
1503
1504 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1505 matching that character.
1506
1507 20. When matching a repeated Unicode property with a minimum greater than zero,
1508 (for example \pL{2,}), PCRE could look past the end of the subject if it
1509 reached it while seeking the minimum number of characters. This could
1510 happen only if some of the characters were more than one byte long, because
1511 there is a check for at least the minimum number of bytes.
1512
1513 21. Refactored the implementation of \p and \P so as to be more general, to
1514 allow for more different types of property in future. This has changed the
1515 compiled form incompatibly. Anybody with saved compiled patterns that use
1516 \p or \P will have to recompile them.
1517
1518 22. Added "Any" and "L&" to the supported property types.
1519
1520 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1521 but give a compile time error if the value is greater than 0xff.
1522
1523 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1524 accidentally not being installed or uninstalled.
1525
1526 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1527 made were to insert the current release number. This seemed silly, because
1528 it made things harder for people building PCRE on systems that don't run
1529 "configure". I have turned pcre.h into a distributed file, no longer built
1530 by "configure", with the version identification directly included. There is
1531 no longer a pcre.h.in file.
1532
1533 However, this change necessitated a change to the pcre-config script as
1534 well. It is built from pcre-config.in, and one of the substitutions was the
1535 release number. I have updated configure.ac so that ./configure now finds
1536 the release number by grepping pcre.h.
1537
1538 26. Added the ability to run the tests under valgrind.
1539
1540
1541 Version 6.4 05-Sep-05
1542 ---------------------
1543
1544 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1545 "--" to be printed when multiple files were scanned, even when none of the
1546 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1547 consider it to be a bug, and have restored the previous behaviour.
1548
1549 2. A couple of code tidies to get rid of compiler warnings.
1550
1551 3. The pcretest program used to cheat by referring to symbols in the library
1552 whose names begin with _pcre_. These are internal symbols that are not
1553 really supposed to be visible externally, and in some environments it is
1554 possible to suppress them. The cheating is now confined to including
1555 certain files from the library's source, which is a bit cleaner.
1556
1557 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1558 file's purpose clearer.
1559
1560 5. Reorganized pcre_ucp_findchar().
1561
1562
1563 Version 6.3 15-Aug-05
1564 ---------------------
1565
1566 1. The file libpcre.pc.in did not have general read permission in the tarball.
1567
1568 2. There were some problems when building without C++ support:
1569
1570 (a) If C++ support was not built, "make install" and "make test" still
1571 tried to test it.
1572
1573 (b) There were problems when the value of CXX was explicitly set. Some
1574 changes have been made to try to fix these, and ...
1575
1576 (c) --disable-cpp can now be used to explicitly disable C++ support.
1577
1578 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1579 backslash in a target when C++ was disabled. This confuses some
1580 versions of "make", apparently. Using an intermediate variable solves
1581 this. (Same for CPP_LOBJ.)
1582
1583 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1584 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1585 necessary on certain architectures.
1586
1587 4. Added a setting of -export-symbols-regex to the link command to remove
1588 those symbols that are exported in the C sense, but actually are local
1589 within the library, and not documented. Their names all begin with
1590 "_pcre_". This is not a perfect job, because (a) we have to except some
1591 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1592 available (and never for static libraries). I have made a note to try to
1593 find a way round (a) in the future.
1594
1595
1596 Version 6.2 01-Aug-05
1597 ---------------------
1598
1599 1. There was no test for integer overflow of quantifier values. A construction
1600 such as {1111111111111111} would give undefined results. What is worse, if
1601 a minimum quantifier for a parenthesized subpattern overflowed and became
1602 negative, the calculation of the memory size went wrong. This could have
1603 led to memory overwriting.
1604
1605 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1606
1607 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1608 operating environments where this matters.
1609
1610 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1611 PCRE options from within the C++ wrapper.
1612
1613 5. Named capturing subpatterns were not being correctly counted when a pattern
1614 was compiled. This caused two problems: (a) If there were more than 100
1615 such subpatterns, the calculation of the memory needed for the whole
1616 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1617 back references of the form \12, where the number was greater than 9, were
1618 not recognized as back references, even though there were sufficient
1619 previous subpatterns.
1620
1621 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1622 versions of gcc, e.g. 2.95.4.
1623
1624
1625 Version 6.1 21-Jun-05
1626 ---------------------
1627
1628 1. There was one reference to the variable "posix" in pcretest.c that was not
1629 surrounded by "#if !defined NOPOSIX".
1630
1631 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1632 the cross-check on the old pcre_info() function, for the benefit of the
1633 cut-down version of PCRE that is currently imported into Exim.
1634
1635 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1636 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1637 patterns that start (?i) but I don't think that matters. The use of (?i) is
1638 just an example; this all applies to the other options as well.
1639
1640 4. Since libtool seems to echo the compile commands it is issuing, the output
1641 from "make" can be reduced a bit by putting "@" in front of each libtool
1642 compile command.
1643
1644 5. Patch from the folks at Google for configure.in to be a bit more thorough
1645 in checking for a suitable C++ installation before trying to compile the
1646 C++ stuff. This should fix a reported problem when a compiler was present,
1647 but no suitable headers.
1648
1649 6. The man pages all had just "PCRE" as their title. I have changed them to
1650 be the relevant file name. I have also arranged that these names are
1651 retained in the file doc/pcre.txt, which is a concatenation in text format
1652 of all the man pages except the little individual ones for each function.
1653
1654 7. The NON-UNIX-USE file had not been updated for the different set of source
1655 files that come with release 6. I also added a few comments about the C++
1656 wrapper.
1657
1658
1659 Version 6.0 07-Jun-05
1660 ---------------------
1661
1662 1. Some minor internal re-organization to help with my DFA experiments.
1663
1664 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1665 didn't matter for the library itself when fully configured, but did matter
1666 when compiling without UCP support, or within Exim, where the ucp files are
1667 not imported.
1668
1669 3. Refactoring of the library code to split up the various functions into
1670 different source modules. The addition of the new DFA matching code (see
1671 below) to a single monolithic source would have made it really too
1672 unwieldy, quite apart from causing all the code to be include in a
1673 statically linked application, when only some functions are used. This is
1674 relevant even without the DFA addition now that patterns can be compiled in
1675 one application and matched in another.
1676
1677 The downside of splitting up is that there have to be some external
1678 functions and data tables that are used internally in different modules of
1679 the library but which are not part of the API. These have all had their
1680 names changed to start with "_pcre_" so that they are unlikely to clash
1681 with other external names.
1682
1683 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1684 a different (DFA) algorithm. Although it is slower than the original
1685 function, it does have some advantages for certain types of matching
1686 problem.
1687
1688 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1689 including restarting after a partial match.
1690
1691 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1692 defined when compiling for Windows was sent to me. I have put it into the
1693 code, though I have no means of testing or verifying it.
1694
1695 7. Added the pcre_refcount() auxiliary function.
1696
1697 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1698 match before or at the first newline in the subject string. In pcretest,
1699 the /f option on a pattern can be used to set this.
1700
1701 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1702 would behave wrongly. This has been present in PCRE since release 4.0.
1703
1704 10. A number of changes to the pcregrep command:
1705
1706 (a) Refactored how -x works; insert ^(...)$ instead of setting
1707 PCRE_ANCHORED and checking the length, in preparation for adding
1708 something similar for -w.
1709
1710 (b) Added the -w (match as a word) option.
1711
1712 (c) Refactored the way lines are read and buffered so as to have more
1713 than one at a time available.
1714
1715 (d) Implemented a pcregrep test script.
1716
1717 (e) Added the -M (multiline match) option. This allows patterns to match
1718 over several lines of the subject. The buffering ensures that at least
1719 8K, or the rest of the document (whichever is the shorter) is available
1720 for matching (and similarly the previous 8K for lookbehind assertions).
1721
1722 (f) Changed the --help output so that it now says
1723
1724 -w, --word-regex(p)
1725
1726 instead of two lines, one with "regex" and the other with "regexp"
1727 because that confused at least one person since the short forms are the
1728 same. (This required a bit of code, as the output is generated
1729 automatically from a table. It wasn't just a text change.)
1730
1731 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1732 option but starts with a hyphen. Could be a pattern or a path name
1733 starting with a hyphen, for instance.
1734
1735 (h) "-" can be given as a file name to represent stdin.
1736
1737 (i) When file names are being printed, "(standard input)" is used for
1738 the standard input, for compatibility with GNU grep. Previously
1739 "<stdin>" was used.
1740
1741 (j) The option --label=xxx can be used to supply a name to be used for
1742 stdin when file names are being printed. There is no short form.
1743
1744 (k) Re-factored the options decoding logic because we are going to add
1745 two more options that take data. Such options can now be given in four
1746 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1747
1748 (l) Added the -A, -B, and -C options for requesting that lines of context
1749 around matches be printed.
1750
1751 (m) Added the -L option to print the names of files that do not contain
1752 any matching lines, that is, the complement of -l.
1753
1754 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1755 continue to scan other files.
1756
1757 (o) The -s option was incorrectly implemented. For compatibility with other
1758 greps, it now suppresses the error message for a non-existent or non-
1759 accessible file (but not the return code). There is a new option called
1760 -q that suppresses the output of matching lines, which was what -s was
1761 previously doing.
1762
1763 (p) Added --include and --exclude options to specify files for inclusion
1764 and exclusion when recursing.
1765
1766 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1767 Hopefully, it now does.
1768
1769 12. Missing cast in pcre_study().
1770
1771 13. Added an "uninstall" target to the makefile.
1772
1773 14. Replaced "extern" in the function prototypes in Makefile.in with
1774 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1775 world, but is set differently for Windows.
1776
1777 15. Added a second compiling function called pcre_compile2(). The only
1778 difference is that it has an extra argument, which is a pointer to an
1779 integer error code. When there is a compile-time failure, this is set
1780 non-zero, in addition to the error test pointer being set to point to an
1781 error message. The new argument may be NULL if no error number is required
1782 (but then you may as well call pcre_compile(), which is now just a
1783 wrapper). This facility is provided because some applications need a
1784 numeric error indication, but it has also enabled me to tidy up the way
1785 compile-time errors are handled in the POSIX wrapper.
1786
1787 16. Added VPATH=.libs to the makefile; this should help when building with one
1788 prefix path and installing with another. (Or so I'm told by someone who
1789 knows more about this stuff than I do.)
1790
1791 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1792 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1793 match everything, including newlines. This is not POSIX-compatible, but
1794 somebody wanted the feature. From pcretest it can be activated by using
1795 both the P and the s flags.
1796
1797 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1798
1799 19. libpcre.pc was being incorrectly installed as executable.
1800
1801 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1802 it now also looks for '\r' so that it will work unmodified on Windows.
1803
1804 21. Added Google's contributed C++ wrapper to the distribution.
1805
1806 22. Added some untidy missing memory free() calls in pcretest, to keep
1807 Electric Fence happy when testing.
1808
1809
1810
1811 Version 5.0 13-Sep-04
1812 ---------------------
1813
1814 1. Internal change: literal characters are no longer packed up into items
1815 containing multiple characters in a single byte-string. Each character
1816 is now matched using a separate opcode. However, there may be more than one
1817 byte in the character in UTF-8 mode.
1818
1819 2. The pcre_callout_block structure has two new fields: pattern_position and
1820 next_item_length. These contain the offset in the pattern to the next match
1821 item, and its length, respectively.
1822
1823 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1824 insertion of callouts before each pattern item. Added the /C option to
1825 pcretest to make use of this.
1826
1827 4. On the advice of a Windows user, the lines
1828
1829 #if defined(_WIN32) || defined(WIN32)
1830 _setmode( _fileno( stdout ), 0x8000 );
1831 #endif /* defined(_WIN32) || defined(WIN32) */
1832
1833 have been added to the source of pcretest. This apparently does useful
1834 magic in relation to line terminators.
1835
1836 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1837 for the benefit of those environments where the "b" makes a difference.
1838
1839 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1840 to know about it. I have put a hack into configure.in that adds in code
1841 to set GCC=yes if CC=icc. This seems to end up at a point in the
1842 generated configure script that is early enough to affect the setting of
1843 compiler options, which is what is needed, but I have no means of testing
1844 whether it really works. (The user who reported this had patched the
1845 generated configure script, which of course I cannot do.)
1846
1847 LATER: After change 22 below (new libtool files), the configure script
1848 seems to know about icc (and also ecc). Therefore, I have commented out
1849 this hack in configure.in.
1850
1851 7. Added support for pkg-config (2 patches were sent in).
1852
1853 8. Negated POSIX character classes that used a combination of internal tables
1854 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1855 [[:^ascii]]. Typically, they would match almost any characters. The other
1856 POSIX classes were not broken in this way.
1857
1858 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1859 to find the match, as PCRE was deluded into thinking that the match had to
1860 start at the start point or following a newline. The same bug applied to
1861 patterns with negative forward assertions or any backward assertions
1862 preceding ".*" at the start, unless the pattern required a fixed first
1863 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1864
1865 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1866 starting at the last subject character, bytes beyond the end of the subject
1867 string were read.
1868
1869 11. Renamed the variable "class" as "classbits" to make life easier for C++
1870 users. (Previously there was a macro definition, but it apparently wasn't
1871 enough.)
1872
1873 12. Added the new field "tables" to the extra data so that tables can be passed
1874 in at exec time, or the internal tables can be re-selected. This allows
1875 a compiled regex to be saved and re-used at a later time by a different
1876 program that might have everything at different addresses.
1877
1878 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1879 -R library as well as a -L library.
1880
1881 14. The debugging options of pcretest (-d on the command line or D on a
1882 pattern) showed incorrect output for anything following an extended class
1883 that contained multibyte characters and which was followed by a quantifier.
1884
1885 15. Added optional support for general category Unicode character properties
1886 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1887 support. It adds about 90K to the size of the library. The meanings of the
1888 inbuilt class escapes such as \d and \s have NOT been changed.
1889
1890 16. Updated pcredemo.c to include calls to free() to release the memory for the
1891 compiled pattern.
1892
1893 17. The generated file chartables.c was being created in the source directory
1894 instead of in the building directory. This caused the build to fail if the
1895 source directory was different from the building directory, and was
1896 read-only.
1897
1898 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1899 file. No doubt somebody will tell me if they don't make sense... Also added
1900 Dan Mooney's comments about building on OpenVMS.
1901
1902 19. Added support for partial matching via the PCRE_PARTIAL option for
1903 pcre_exec() and the \P data escape in pcretest.
1904
1905 20. Extended pcretest with 3 new pattern features:
1906
1907 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1908 write the compiled pattern to the file whose name is "rest-of-line".
1909 This is a straight binary dump of the data, with the saved pointer to
1910 the character tables forced to be NULL. The study data, if any, is
1911 written too. After writing, pcretest reads a new pattern.
1912
1913 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1914 compiled pattern from the given file. There must not be any
1915 occurrences of "<" in the file name (pretty unlikely); if there are,
1916 pcretest will instead treat the initial "<" as a pattern delimiter.
1917 After reading in the pattern, pcretest goes on to read data lines as
1918 usual.
1919
1920 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1921 and 16-bit fields in a compiled pattern, to simulate a pattern that
1922 was compiled on a host of opposite endianness.
1923
1924 21. The pcre-exec() function can now cope with patterns that were compiled on
1925 hosts of opposite endianness, with this restriction:
1926
1927 As for any compiled expression that is saved and used later, the tables
1928 pointer field cannot be preserved; the extra_data field in the arguments
1929 to pcre_exec() should be used to pass in a tables address if a value
1930 other than the default internal tables were used at compile time.
1931
1932 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1933 now diagnosed as an error. Previously, most of the time, a negative number
1934 would have been treated as zero, but if in addition "ovector" was passed as
1935 NULL, a crash could occur.
1936
1937 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1938 new versions from the libtool 1.5 distribution (the last one is a copy of
1939 a file called libtool.m4). This seems to have fixed the need to patch
1940 "configure" to support Darwin 1.3 (which I used to do). However, I still
1941 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1942 workstation).
1943
1944 24. Changed the PCRE licence to be the more standard "BSD" licence.
1945
1946
1947 Version 4.5 01-Dec-03
1948 ---------------------
1949
1950 1. There has been some re-arrangement of the code for the match() function so
1951 that it can be compiled in a version that does not call itself recursively.
1952 Instead, it keeps those local variables that need separate instances for
1953 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1954 needs to "recurse". Keeping track of where control must go is done by means
1955 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1956 hide most of the details from the main code, and operates only if
1957 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1958 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1959 operating.
1960
1961 To make it easier for callers to provide specially tailored get/free
1962 functions for this usage, two new functions, pcre_stack_malloc, and
1963 pcre_stack_free, are used. They are always called in strict stacking order,
1964 and the size of block requested is always the same.
1965
1966 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1967 PCRE has been compiled to use the stack or the heap for recursion. The
1968 -C option of pcretest uses this to show which version is compiled.
1969
1970 A new data escape \S, is added to pcretest; it causes the amounts of store
1971 obtained and freed by both kinds of malloc/free at match time to be added
1972 to the output.
1973
1974 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1975 what's available on my current Linux desktop machine.
1976
1977 3. When matching a UTF-8 string, the test for a valid string at the start has
1978 been extended. If start_offset is not zero, PCRE now checks that it points
1979 to a byte that is the start of a UTF-8 character. If not, it returns
1980 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1981 this is necessary because there may be backward assertions in the pattern.
1982 When matching the same subject several times, it may save resources to use
1983 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1984
1985 4. The code for checking the validity of UTF-8 strings has been tightened so
1986 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1987 containing "overlong sequences".
1988
1989 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1990 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1991 should have been "&", but it just so happened that all the cases this let
1992 through by mistake were picked up later in the function.
1993
1994 6. I had used a variable called "isblank" - this is a C99 function, causing
1995 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1996
1997 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1998 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1999 so the version is shown; (c) stop "make test" echoing "./RunTest".
2000
2001 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2002
2003 9. The prototype for memmove() for systems that don't have it was using
2004 size_t, but the inclusion of the header that defines size_t was later. I've
2005 moved the #includes for the C headers earlier to avoid this.
2006
2007 10. Added some adjustments to the code to make it easier to compiler on certain
2008 special systems:
2009
2010 (a) Some "const" qualifiers were missing.
2011 (b) Added the macro EXPORT before all exported functions; by default this
2012 is defined to be empty.
2013 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2014 that it reads its output file name as an argument instead of writing
2015 to the standard output and assuming this can be redirected.
2016
2017 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2018 class containing characters with values greater than 255, PCRE compilation
2019 went into a loop.
2020
2021 12. A recursive reference to a subpattern that was within another subpattern
2022 that had a minimum quantifier of zero caused PCRE to crash. For example,
2023 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2024 recursion. If the recursively-called subpattern itself had a zero repeat,
2025 that was OK.
2026
2027 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2028 buffer into which it was copied (for escape processing) was still set at
2029 1024, so long lines caused crashes.
2030
2031 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2032 "internal error: code overflow...". This applied to any character class
2033 that was followed by a possessive quantifier.
2034
2035 15. Modified the Makefile to add libpcre.la as a prerequisite for
2036 libpcreposix.la because I was told this is needed for a parallel build to
2037 work.
2038
2039 16. If a pattern that contained .* following optional items at the start was
2040 studied, the wrong optimizing data was generated, leading to matching
2041 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2042 matching string must start with a or b or c. The correct conclusion for
2043 this pattern is that a match can start with any character.
2044
2045
2046 Version 4.4 13-Aug-03
2047 ---------------------
2048
2049 1. In UTF-8 mode, a character class containing characters with values between
2050 127 and 255 was not handled correctly if the compiled pattern was studied.
2051 In fixing this, I have also improved the studying algorithm for such
2052 classes (slightly).
2053
2054 2. Three internal functions had redundant arguments passed to them. Removal
2055 might give a very teeny performance improvement.
2056
2057 3. Documentation bug: the value of the capture_top field in a callout is *one
2058 more than* the number of the hightest numbered captured substring.
2059
2060 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2061 in incorrectly linking with a previously installed version. They now link
2062 explicitly with libpcre.la.
2063
2064 5. configure.in no longer needs to recognize Cygwin specially.
2065
2066 6. A problem in pcre.in for Windows platforms is fixed.
2067
2068 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2069 pcretest, it used to include the size of the study block as part of its
2070 output. Unfortunately, the structure contains a field that has a different
2071 size on different hardware architectures. This meant that the tests that
2072 showed this size failed. As the block is currently always of a fixed size,
2073 this information isn't actually particularly useful in pcretest output, so
2074 I have just removed it.
2075
2076 8. Three pre-processor statements accidentally did not start in column 1.
2077 Sadly, there are *still* compilers around that complain, even though
2078 standard C has not required this for well over a decade. Sigh.
2079
2080 9. In pcretest, the code for checking callouts passed small integers in the
2081 callout_data field, which is a void * field. However, some picky compilers
2082 complained about the casts involved for this on 64-bit systems. Now
2083 pcretest passes the address of the small integer instead, which should get
2084 rid of the warnings.
2085
2086 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2087 both compile and run time, and gives an error if an invalid UTF-8 sequence
2088 is found. There is a option for disabling this check in cases where the
2089 string is known to be correct and/or the maximum performance is wanted.
2090
2091 11. In response to a bug report, I changed one line in Makefile.in from
2092
2093 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2094 to
2095 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2096
2097 to look similar to other lines, but I have no way of telling whether this
2098 is the right thing to do, as I do not use Windows. No doubt I'll get told
2099 if it's wrong...
2100
2101
2102 Version 4.3 21-May-03
2103 ---------------------
2104
2105 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2106 Makefile.
2107
2108 2. Some refactoring to improve the quality of the code:
2109
2110 (i) The utf8_table... variables are now declared "const".
2111
2112 (ii) The code for \cx, which used the "case flipping" table to upper case
2113 lower case letters, now just substracts 32. This is ASCII-specific,
2114 but the whole concept of \cx is ASCII-specific, so it seems
2115 reasonable.
2116
2117 (iii) PCRE was using its character types table to recognize decimal and
2118 hexadecimal digits in the pattern. This is silly, because it handles
2119 only 0-9, a-f, and A-F, but the character types table is locale-
2120 specific, which means strange things might happen. A private
2121 table is now used for this - though it costs 256 bytes, a table is
2122 much faster than multiple explicit tests. Of course, the standard
2123 character types table is still used for matching digits in subject
2124 strings against \d.
2125
2126 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2127 ending in _t are). So I've renamed it as ESC_tee.
2128
2129 3. The first argument for regexec() in the POSIX wrapper should have been
2130 defined as "const".
2131
2132 4. Changed pcretest to use malloc() for its buffers so that they can be
2133 Electric Fenced for debugging.
2134
2135 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2136 to read one or more bytes before the start of the subject string. Often this
2137 had no effect on PCRE's behaviour, but in some circumstances it could
2138 provoke a segmentation fault.
2139
2140 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2141 to try to read one or more bytes before the start of the subject string.
2142
2143 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2144 UTF-8 support could misbehave in various ways if the subject string
2145 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2146 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2147 back over UTF-8 characters.)
2148
2149
2150 Version 4.2 14-Apr-03
2151 ---------------------
2152
2153 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2154
2155 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2156 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2157 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2158 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2159 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2160 and BUILD_EXEEXT
2161 Note: automatic setting of the BUILD variables is not yet working
2162 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2163 compile-time but not at link-time
2164 [LINK]: use for linking executables only
2165 make different versions for Windows and non-Windows
2166 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2167 libraries
2168 [LINK_FOR_BUILD]: new variable
2169 [OBJEXT]: use throughout
2170 [EXEEXT]: use throughout
2171 <winshared>: new target
2172 <wininstall>: new target
2173 <dftables.o>: use native compiler
2174 <dftables>: use native linker
2175 <install>: handle Windows platform correctly
2176 <clean>: ditto
2177 <check>: ditto
2178 copy DLL to top builddir before testing
2179
2180 As part of these changes, -no-undefined was removed again. This was reported
2181 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2182 in any case.
2183
2184 3. Some tidies to get rid of compiler warnings:
2185
2186 . In the match_data structure, match_limit was an unsigned long int, whereas
2187 match_call_count was an int. I've made them both unsigned long ints.
2188
2189 . In pcretest the fact that a const uschar * doesn't automatically cast to
2190 a void * provoked a warning.
2191
2192 . Turning on some more compiler warnings threw up some "shadow" variables
2193 and a few more missing casts.
2194
2195 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2196 option, a class that contained a single character with a value between 128
2197 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2198
2199 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2200 option, a class that contained several characters, but with at least one
2201 whose value was between 128 and 255 caused PCRE to crash.
2202
2203
2204 Version 4.1 12-Mar-03
2205 ---------------------
2206
2207 1. Compiling with gcc -pedantic found a couple of places where casts were
2208 needed, and a string in dftables.c that was longer than standard compilers are
2209 required to support.
2210
2211 2. Compiling with Sun's compiler found a few more places where the code could
2212 be tidied up in order to avoid warnings.
2213
2214 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2215 first of these names is deprecated in the latest Autoconf in favour of the name
2216 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2217 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2218 analogy I have changed it to CFLAGS_FOR_BUILD.
2219
2220 4. Added -no-undefined to the linking command in the Makefile, because this is
2221 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2222 linking step for the pcreposix library.
2223
2224 5. PCRE was failing to diagnose the case of two named groups with the same
2225 name.
2226
2227 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2228 literal character that is needed in the subject for a match, and scans along to
2229 ensure that it is present before embarking on the full matching process. This
2230 saves time in cases of nested unlimited repeats that are never going to match.
2231 Problem: the scan can take a lot of time if the subject is very long (e.g.
2232 megabytes), thus penalizing straightforward matches. It is now done only if the
2233 amount of subject to be scanned is less than 1000 bytes.
2234
2235 7. A lesser problem with the same optimization is that it was recording the
2236 first character of an anchored pattern as "needed", thus provoking a search
2237 right along the subject, even when the first match of the pattern was going to
2238 fail. The "needed" character is now not set for anchored patterns, unless it
2239 follows something in the pattern that is of non-fixed length. Thus, it still
2240 fulfils its original purpose of finding quick non-matches in cases of nested
2241 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2242
2243
2244 Version 4.0 17-Feb-03
2245 ---------------------
2246
2247 1. If a comment in an extended regex that started immediately after a meta-item
2248 extended to the end of string, PCRE compiled incorrect data. This could lead to
2249 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2250
2251 2. Moved to autoconf 2.53 and libtool 1.4.2.
2252
2253 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2254 the special perltest8 script is no longer needed - all the tests can be run
2255 from a single perltest script.
2256
2257 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2258 by \s. It has now been removed in PCRE. This means it isn't recognized as
2259 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2260 class [:space:] *does* include VT, thereby creating a mess.
2261
2262 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2263 space and tab.
2264
2265 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2266 its new features into the main test script, reducing the number of scripts.
2267
2268 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2269 were backward compatible, and made the (?i) apply to the whole pattern, as if
2270 /i were given. Now it behaves more logically, and applies the option setting
2271 only to what follows. PCRE has been changed to follow suit. However, if it
2272 finds options settings right at the start of the pattern, it extracts them into
2273 the global options, as before. Thus, they show up in the info data.
2274
2275 8. Added support for the \Q...\E escape sequence. Characters in between are
2276 treated as literals. This is slightly different from Perl in that $ and @ are
2277 also handled as literals inside the quotes. In Perl, they will cause variable
2278 interpolation. Note the following examples:
2279
2280 Pattern PCRE matches Perl matches
2281
2282 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2283 \Qabc\$xyz\E abc\$xyz abc\$xyz
2284 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2285
2286 For compatibility with Perl, \Q...\E sequences are recognized inside character
2287 classes as well as outside them.
2288
2289 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2290 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2291 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2292 signed/unsigned warnings.
2293
2294 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2295 option for pcretest, so I've replaced it by a simple function that does just
2296 that job.
2297
2298 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2299 "pcregrep -".
2300
2301 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2302 Java package. This provides some syntactic sugar for simple cases of what my
2303 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2304 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2305 item, you can use this simplified notation. Note that only makes sense with
2306 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2307 greediness, whatever the setting of the PCRE_UNGREEDY option.
2308
2309 13. A change of greediness default within a pattern was not taking effect at
2310 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2311 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2312 was abstracted outside.
2313
2314 14. PCRE now supports the \G assertion. It is true when the current matching
2315 position is at the start point of the match. This differs from \A when the
2316 starting offset is non-zero. Used with the /g option of pcretest (or similar
2317 code), it works in the same way as it does for Perl's /g option. If all
2318 alternatives of a regex begin with \G, the expression is anchored to the start
2319 match position, and the "anchored" flag is set in the compiled expression.
2320
2321 15. Some bugs concerning the handling of certain option changes within patterns
2322 have been fixed. These applied to options other than (?ims). For example,
2323 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2324 been the other way round. Some of this was related to change 7 above.
2325
2326 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2327 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2328 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2329 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2330
2331 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2332 mode. Unlike ".", it always matches newline, whatever the setting of
2333 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2334 assertions. Perl allows it, but it doesn't (in general) work because it can't
2335 calculate the length of the lookbehind. At least, that's the case for Perl
2336 5.8.0 - I've been told they are going to document that it doesn't work in
2337 future.
2338
2339 18. Added an error diagnosis for escapes that PCRE does not support: these are
2340 \L, \l, \N, \P, \p, \U, \u, and \X.
2341
2342 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2343 reading past the end of the pattern in cases such as /[abcd/.
2344
2345 20. PCRE was getting more memory than necessary for patterns with classes that
2346 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2347
2348 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2349 compiling PCRE for use with Virtual Pascal.
2350
2351 22. Small fix to the Makefile to make it work properly if the build is done
2352 outside the source tree.
2353
2354 23. Added a new extension: a condition to go with recursion. If a conditional
2355 subpattern starts with (?(R) the "true" branch is used if recursion has
2356 happened, whereas the "false" branch is used only at the top level.
2357
2358 24. When there was a very long string of literal characters (over 255 bytes
2359 without UTF support, over 250 bytes with UTF support), the computation of how
2360 much memory was required could be incorrect, leading to segfaults or other
2361 strange effects.
2362
2363 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2364 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2365 there was a subsequent back reference to those brackets. This meant that, for
2366 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2367 possible to check for precisely this case. All we can do is abandon the
2368 optimization if .* occurs inside capturing brackets when there are any back
2369 references whatsoever. (See below for a better fix that came later.)
2370
2371 26. The handling of the optimization for finding the first character of a
2372 non-anchored pattern, and for finding a character that is required later in the
2373 match were failing in some cases. This didn't break the matching; it just
2374 failed to optimize when it could. The way this is done has been re-implemented.
2375
2376 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2377
2378 28. Added a new feature that provides some of the functionality that Perl
2379 provides with (?{...}). The facility is termed a "callout". The way it is done
2380 in PCRE is for the caller to provide an optional function, by setting
2381 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2382 global variable. By default it is unset, which disables all calling out. To get
2383 the function called, the regex must include (?C) at appropriate points. This
2384 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2385 This provides a means of identifying different callout points. When PCRE
2386 reaches such a point in the regex, if pcre_callout has been set, the external
2387 function is called. It is provided with data in a structure called
2388 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2389 matching continues; if it returns a non-zero value, the match at the current
2390 point fails. However, backtracking will occur if possible. [This was changed
2391 later and other features added - see item 49 below.]
2392
2393 29. pcretest is upgraded to test the callout functionality. It provides a
2394 callout function that displays information. By default, it shows the start of
2395 the match and the current position in the text. There are some new data escapes
2396 to vary what happens:
2397
2398 \C+ in addition, show current contents of captured substrings
2399 \C- do not supply a callout function
2400 \C!n return 1 when callout number n is reached
2401 \C!n!m return 1 when callout number n is reached for the mth time
2402
2403 30. If pcregrep was called with the -l option and just a single file name, it
2404 output "<stdin>" if a match was found, instead of the file name.
2405
2406 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2407 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2408 pcre_exec(). This saves a malloc/free per call. The default value of
2409 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2410 when configuring.
2411
2412 32. The default maximum size of a compiled pattern is 64K. There have been a
2413 few cases of people hitting this limit. The code now uses macros to handle the
2414 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2415 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2416 configuring. Tests 2 and 5 work only with 2-byte links because they output
2417 debugging information about compiled patterns.
2418
2419 33. Internal code re-arrangements:
2420
2421 (a) Moved the debugging function for printing out a compiled regex into
2422 its own source file (printint.c) and used #include to pull it into
2423 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2424 separate copies.
2425
2426 (b) Defined the list of op-code names for debugging as a macro in
2427 internal.h so that it is next to the definition of the opcodes.
2428
2429 (c) Defined a table of op-code lengths for simpler skipping along compiled
2430 code. This is again a macro in internal.h so that it is next to the
2431 definition of the opcodes.
2432
2433 34. Added support for recursive calls to individual subpatterns, along the
2434 lines of Robin Houston's patch (but implemented somewhat differently).
2435
2436 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2437 allow it to read and process whole directories in Win32. This code was
2438 contributed by Lionel Fourquaux; it has not been tested by me.
2439
2440 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2441 used to name a group. Names consist of alphanumerics and underscores, and must
2442 be unique. Back references use the syntax (?P=name) and recursive calls use
2443 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2444 numbers. The function pcre_fullinfo() can be used after compilation to extract
2445 a name/number map. There are three relevant calls:
2446
2447 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2448 PCRE_INFO_NAMECOUNT yields the number of entries
2449 PCRE_INFO_NAMETABLE yields a pointer to the map.
2450
2451 The map is a vector of fixed-size entries. The size of each entry depends on
2452 the length of the longest name used. The first two bytes of each entry are the
2453 group number, most significant byte first. There follows the corresponding
2454 name, zero terminated. The names are in alphabetical order.
2455
2456 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2457 case instead of 255. Making it the same both with and without UTF-8 support
2458 means that the same test output works with both.
2459
2460 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2461 calling malloc() with a zero argument.
2462
2463 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2464 optimization. I've improved things by keeping a bitmap of backreferences with
2465 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2466 fact referenced, the optimization can be applied. It is unlikely that a
2467 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2468 the match to follow \n) will appear inside brackets with a number greater than
2469 31, but if it does, any back reference > 31 suppresses the optimization.
2470
2471 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2472 of disabling numbered capturing parentheses. Any opening parenthesis that is
2473 not followed by ? behaves as if it were followed by ?: but named parentheses
2474 can still be used for capturing (and they will acquire numbers in the usual
2475 way).
2476
2477 41. Redesigned the return codes from the match() function into yes/no/error so
2478 that errors can be passed back from deep inside the nested calls. A malloc
2479 failure while inside a recursive subpattern call now causes the
2480 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2481
2482 42. It is now possible to set a limit on the number of times the match()
2483 function is called in a call to pcre_exec(). This facility makes it possible to
2484 limit the amount of recursion and backtracking, though not in a directly
2485 obvious way, because the match() function is used in a number of different
2486 circumstances. The count starts from zero for each position in the subject
2487 string (for non-anchored patterns). The default limit is, for compatibility, a
2488 large number, namely 10 000 000. You can change this in two ways:
2489
2490 (a) When configuring PCRE before making, you can use --with-match-limit=n
2491 to set a default value for the compiled library.
2492
2493 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2494 a different value is set. See 45 below.
2495
2496 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2497
2498 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2499 of things that can be changed at compile time. The first argument specifies
2500 what is wanted and the second points to where the information is to be placed.
2501 The current list of available information is:
2502
2503 PCRE_CONFIG_UTF8
2504
2505 The output is an integer that is set to one if UTF-8 support is available;
2506 otherwise it is set to zero.
2507
2508 PCRE_CONFIG_NEWLINE
2509
2510 The output is an integer that it set to the value of the code that is used for
2511 newline. It is either LF (10) or CR (13).
2512
2513 PCRE_CONFIG_LINK_SIZE
2514
2515 The output is an integer that contains the number of bytes used for internal
2516 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2517
2518 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2519
2520 The output is an integer that contains the threshold above which the POSIX
2521 interface uses malloc() for output vectors. See item 31 above.
2522
2523 PCRE_CONFIG_MATCH_LIMIT
2524
2525 The output is an unsigned integer that contains the default limit of the number
2526 of match() calls in a pcre_exec() execution. See 42 above.
2527
2528 44. pcretest has been upgraded by the addition of the -C option. This causes it
2529 to extract all the available output from the new pcre_config() function, and to
2530 output it. The program then exits immediately.
2531
2532 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2533 order to support additional features. One way would have been to define
2534 pcre_exec2() (for example) with extra arguments, but this would not have been
2535 extensible, and would also have required all calls to the original function to
2536 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2537 is used for passing in "extra" data from pcre_study().
2538
2539 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2540 contains the following fields:
2541
2542 flags a bitmap indicating which of the following fields are set
2543 study_data opaque data from pcre_study()
2544 match_limit a way of specifying a limit on match() calls for a specific
2545 call to pcre_exec()
2546 callout_data data for callouts (see 49 below)
2547
2548 The flag bits are also defined in pcre.h, and are
2549
2550 PCRE_EXTRA_STUDY_DATA
2551 PCRE_EXTRA_MATCH_LIMIT
2552 PCRE_EXTRA_CALLOUT_DATA
2553
2554 The pcre_study() function now returns one of these new pcre_extra blocks, with
2555 the actual study data pointed to by the study_data field, and the
2556 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2557 before. That is, this change is entirely upwards-compatible and requires no
2558 change to existing code.
2559
2560 If you want to pass in additional data to pcre_exec(), you can either place it
2561 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2562 block.
2563
2564 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2565 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2566 times with different match limits, until it finds the minimum value needed for
2567 pcre_exec() to complete. The value is then output. This can be instructive; for
2568 most simple matches the number is quite small, but for pathological cases it
2569 gets very large very quickly.
2570
2571 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2572 returns the size of the data block pointed to by the study_data field in a
2573 pcre_extra block, that is, the value that was passed as the argument to
2574 pcre_malloc() when PCRE was getting memory in which to place the information
2575 created by pcre_study(). The fourth argument should point to a size_t variable.
2576 pcretest has been extended so that this information is shown after a successful
2577 pcre_study() call when information about the compiled regex is being displayed.
2578
2579 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2580 because what follows is always an absolute path. (Later: it turns out that this
2581 is more than cosmetic for MinGW, because it doesn't like empty path
2582 components.)
2583
2584 49. Some changes have been made to the callout feature (see 28 above):
2585
2586 (i) A callout function now has three choices for what it returns:
2587
2588 0 => success, carry on matching
2589 > 0 => failure at this point, but backtrack if possible
2590 < 0 => serious error, return this value from pcre_exec()
2591
2592 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2593 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2594 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2595 use by callout functions. It will never be used by PCRE itself.
2596
2597 (ii) The pcre_extra structure (see 45 above) has a void * field called
2598 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2599 pcre_callout_block structure has a field of the same name. The contents of
2600 the field passed in the pcre_extra structure are passed to the callout
2601 function in the corresponding field in the callout block. This makes it
2602 easier to use the same callout-containing regex from multiple threads. For
2603 testing, the pcretest program has a new data escape
2604
2605 \C*n pass the number n (may be negative) as callout_data
2606
2607 If the callout function in pcretest receives a non-zero value as
2608 callout_data, it returns that value.
2609
2610 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2611 there were some redundant $(CFLAGS) in commands that are now specified as
2612 $(LINK), which already includes $(CFLAGS).
2613
2614 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2615 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2616 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2617 one-byte characters throughout. Note that case-insensitive matching applies
2618 only to characters whose values are less than 256. PCRE doesn't support the
2619 notion of cases for higher-valued characters.
2620
2621 (i) A character class whose characters are all within 0-255 is handled as
2622 a bit map, and the map is inverted for negative classes. Previously, a
2623 character > 255 always failed to match such a class; however it should
2624 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2625
2626 (ii) A negated character class with a single character < 255 is coded as
2627 "not this character" (OP_NOT). This wasn't working properly when the test
2628 character was multibyte, either singly or repeated.
2629
2630 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2631 mode, for example: \x{100}{2,3}.
2632
2633 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2634 singly or repeated) now correctly test multibyte characters. However,
2635 PCRE doesn't recognize any characters with values greater than 255 as
2636 digits, spaces, or word characters. Such characters always match \D, \S,
2637 and \W, and never match \d, \s, or \w.
2638
2639 (v) Classes may now contain characters and character ranges with values
2640 greater than 255. For example: [ab\x{100}-\x{400}].
2641
2642 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2643 PCRE in UTF-8 mode.
2644
2645 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2646 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2647 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2648 value.)
2649
2650 53. The single man page has become too large. I have therefore split it up into
2651 a number of separate man pages. These also give rise to individual HTML pages;
2652 these are now put in a separate directory, and there is an index.html page that
2653 lists them all. Some hyperlinking between the pages has been installed.
2654
2655 54. Added convenience functions for handling named capturing parentheses.
2656
2657 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2658 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2659 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2660 are faulted.
2661
2662 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2663 calling configure. These values are used when compiling the dftables.c program
2664 which is run to generate the source of the default character tables. They
2665 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2666 you will need to set these values.
2667
2668 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2669
2670
2671 Version 3.9 02-Jan-02
2672 ---------------------
2673
2674 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2675
2676 2. If --disable-static was given, the building process failed when trying to
2677 build pcretest and pcregrep. (For some reason it was using libtool to compile
2678 them, which is not right, as they aren't part of the library.)
2679
2680
2681 Version 3.8 18-Dec-01
2682 ---------------------
2683
2684 1. The experimental UTF-8 code was completely screwed up. It was packing the
2685 bytes in the wrong order. How dumb can you get?
2686
2687
2688 Version 3.7 29-Oct-01
2689 ---------------------
2690
2691 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2692 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2693 this didn't happen under Solaris 8, where I normally test things.
2694
2695 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2696 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2697 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2698 this fix an hour or so after the initial 3.7 release.)
2699
2700
2701 Version 3.6 23-Oct-01
2702 ---------------------
2703
2704 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2705 offsets passed as NULL with zero offset count.
2706
2707 2. The config.guess and config.sub files had not been updated when I moved to
2708 the latest autoconf.
2709
2710
2711 Version 3.5 15-Aug-01
2712 ---------------------
2713
2714 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2715 had been forgotten.
2716
2717 2. By using declared but undefined structures, we can avoid using "void"
2718 definitions in pcre.h while keeping the internal definitions of the structures
2719 private.
2720
2721 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2722 user point of view, this means that both static and shared libraries are built
2723 by default, but this can be individually controlled. More of the work of
2724 handling this static/shared cases is now inside libtool instead of PCRE's make
2725 file.
2726
2727 4. The pcretest utility is now installed along with pcregrep because it is
2728 useful for users (to test regexs) and by doing this, it automatically gets
2729 relinked by libtool. The documentation has been turned into a man page, so
2730 there are now .1, .txt, and .html versions in /doc.
2731
2732 5. Upgrades to pcregrep:
2733 (i) Added long-form option names like gnu grep.
2734 (ii) Added --help to list all options with an explanatory phrase.
2735 (iii) Added -r, --recursive to recurse into sub-directories.
2736 (iv) Added -f, --file to read patterns from a file.
2737
2738 6. pcre_exec() was referring to its "code" argument before testing that
2739 argument for NULL (and giving an error if it was NULL).
2740
2741 7. Upgraded Makefile.in to allow for compiling in a different directory from
2742 the source directory.
2743
2744 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2745 options bits, the pointer it was passed was to an int instead of to an unsigned
2746 long int. This mattered only on 64-bit systems.
2747
2748 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2749 generated) instead of pcre.in, which it its source. Also made the same change
2750 in several of the .c files.
2751
2752 10. A new release of gcc defines printf() as a macro, which broke pcretest
2753 because it had an ifdef in the middle of a string argument for printf(). Fixed
2754 by using separate calls to printf().
2755
2756 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2757 script, to force use of CR or LF instead of \n in the source. On non-Unix
2758 systems, the value can be set in config.h.
2759
2760 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2761 absolute limit. Changed the text of the error message to make this clear, and
2762 likewise updated the man page.
2763
2764 13. The limit of 99 on the number of capturing subpatterns has been removed.
2765 The new limit is 65535, which I hope will not be a "real" limit.
2766
2767
2768 Version 3.4 22-Aug-00
2769 ---------------------
2770
2771 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2772
2773 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2774
2775
2776 Version 3.3 01-Aug-00
2777 ---------------------
2778
2779 1. If an octal character was given, but the value was greater than \377, it
2780 was not getting masked to the least significant bits, as documented. This could
2781 lead to crashes in some systems.
2782
2783 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2784 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2785
2786 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2787 These just pass their arguments on to (pcre_free)(), but they are provided
2788 because some uses of PCRE bind it to non-C systems that can call its functions,
2789 but cannot call free() or pcre_free() directly.
2790
2791 4. Add "make test" as a synonym for "make check". Corrected some comments in
2792 the Makefile.
2793
2794 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2795 Makefile.
2796
2797 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2798 command called pgrep for grepping around the active processes.
2799
2800 7. Added the beginnings of support for UTF-8 character strings.
2801
2802 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2803 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2804 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2805 out for the ar command.)
2806
2807
2808 Version 3.2 12-May-00
2809 ---------------------
2810
2811 This is purely a bug fixing release.
2812
2813 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2814 of ZA. This was just one example of several cases that could provoke this bug,
2815 which was introduced by change 9 of version 2.00. The code for breaking
2816 infinite loops after an iteration that matches an empty string was't working
2817 correctly.
2818
2819 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2820 when matched against abbab (for example). After matching an empty string, it
2821 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2822 caused it to match further down the string than it should.
2823
2824 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2825 was there because it doesn't seem to be needed, and it causes trouble on some
2826 systems, as it is not a Standard C header. It has been removed.
2827
2828 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2829 were reported on the Macintosh. The changes were from
2830
2831 while ((c = *(++ptr)) != 0 && c != '\n');
2832 to
2833 while ((c = *(++ptr)) != 0 && c != '\n') ;
2834
2835 Totally extraordinary, but if that's what it takes...
2836
2837 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2838 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2839 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2840 assumes the way PCRE uses memmove() (always moving upwards).
2841
2842 6. PCRE is being used in one environment where strchr() is not available. There
2843 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2844 faster code anyway.
2845
2846
2847 Version 3.1 09-Feb-00
2848 ---------------------
2849
2850 The only change in this release is the fixing of some bugs in Makefile.in for
2851 the "install" target:
2852
2853 (1) It was failing to install pcreposix.h.
2854
2855 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2856
2857
2858 Version 3.0 01-Feb-00
2859 ---------------------
2860
2861 1. Add support for the /+ modifier to perltest (to output $` like it does in
2862 pcretest).
2863
2864 2. Add support for the /g modifier to perltest.
2865
2866 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2867 matches null strings.
2868
2869 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2870 pattern. Perl treats empty patterns specially - it reuses the most recent
2871 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2872 effect.
2873
2874 5. The POSIX interface was broken in that it was just handing over the POSIX
2875 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2876 required a bigger vector, with some working space on the end. This means that
2877 the POSIX wrapper now has to get and free some memory, and copy the results.
2878
2879 6. Added some simple autoconf support, placing the test data and the
2880 documentation in separate directories, re-organizing some of the
2881 information files, and making it build pcre-config (a GNU standard). Also added
2882 libtool support for building PCRE as a shared library, which is now the
2883 default.
2884
2885 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2886 09 are not valid octal constants. Single digits will be used for minor values
2887 less than 10.
2888
2889 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2890 existing programs that set these in the POSIX interface can use PCRE without
2891 modification.
2892
2893 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2894 return all that pcre_info() returns, plus additional data. The pcre_info()
2895 function is retained for compatibility, but is considered to be obsolete.
2896
2897 10. Added experimental recursion feature (?R) to handle one common case that
2898 Perl 5.6 will be able to do with (?p{...}).
2899
2900 11. Added support for POSIX character classes like [:alpha:], which Perl is
2901 adopting.
2902
2903
2904 Version 2.08 31-Aug-99
2905 ----------------------
2906
2907 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2908 trying to match at the startoffset position, but instead was moving forward to
2909 the next newline as if a previous match had failed.
2910
2911 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2912 and could get into a loop if a null string was matched other than at the start
2913 of the subject.
2914
2915 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2916 be distinguished at compile time, and for completeness also added PCRE_DATE.
2917
2918 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2919 in GnuWin32 environments.
2920
2921
2922 Version 2.07 29-Jul-99
2923 ----------------------
2924
2925 1. The documentation is now supplied in plain text form and HTML as well as in
2926 the form of man page sources.
2927
2928 2. C++ compilers don't like assigning (void *) values to other pointer types.
2929 In particular this affects malloc(). Although there is no problem in Standard
2930 C, I've put in casts to keep C++ compilers happy.
2931
2932 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2933 should be (const char *).
2934
2935 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2936 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2937 However, I haven't made this a standard facility. The documentation doesn't
2938 mention it, and the Makefile doesn't support it.
2939
2940 5. The Makefile now contains an "install" target, with editable destinations at
2941 the top of the file. The pcretest program is not installed.
2942
2943 6. pgrep -V now gives the PCRE version number and date.
2944
2945 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2946 causing the entire string to be ignored, instead of just the last character.
2947
2948 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2949 non-matching string, it can take a very, very long time, even for strings of
2950 quite modest length, because of the nested recursion. PCRE now does better in
2951 some of these cases. It does this by remembering the last required literal
2952 character in the pattern, and pre-searching the subject to ensure it is present
2953 before running the real match. In other words, it applies a heuristic to detect
2954 some types of certain failure quickly, and in the above example, if presented
2955 with a string that has no trailing " it gives "no match" very quickly.
2956
2957 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2958 other alternatives are tried instead.
2959
2960
2961 Version 2.06 09-Jun-99
2962 ----------------------
2963
2964 1. Change pcretest's output for amount of store used to show just the code
2965 space, because the remainder (the data block) varies in size between 32-bit and
2966 64-bit systems.
2967
2968 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2969 start matching at. This allows lookbehinds to work when searching for multiple
2970 occurrences in a string.
2971
2972 3. Added additional options to pcretest for testing multiple occurrences:
2973
2974 /+ outputs the rest of the string that follows a match
2975 /g loops for multiple occurrences, using the new startoffset argument
2976 /G loops for multiple occurrences by passing an incremented pointer
2977
2978 4. PCRE wasn't doing the "first character" optimization for patterns starting
2979 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2980 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2981 the letter 'x'. On long subject strings, this gives a significant speed-up.
2982
2983
2984 Version 2.05 21-Apr-99
2985 ----------------------
2986
2987 1. Changed the type of magic_number from int to long int so that it works
2988 properly on 16-bit systems.
2989
2990 2. Fixed a bug which caused patterns starting with .* not to work correctly
2991 when the subject string contained newline characters. PCRE was assuming
2992 anchoring for such patterns in all cases, which is not correct because .* will
2993 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2994 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2995 must be retried after every newline in the subject.
2996
2997
2998 Version 2.04 18-Feb-99
2999 ----------------------
3000
3001 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3002 computation of the store needed to hold the pattern was incorrect (too large).
3003 If such patterns were nested a few deep, this could multiply and become a real
3004 problem.
3005
3006 2. Added /M option to pcretest to show the memory requirement of a specific
3007 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3008
3009 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3010 compiled in such a way that the backtracking after subsequent failure was
3011 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3012 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3013
3014
3015 Version 2.03 02-Feb-99
3016 ----------------------
3017
3018 1. Fixed typo and small mistake in man page.
3019
3020 2. Added 4th condition (GPL supersedes if conflict) and created separate
3021 LICENCE file containing the conditions.
3022
3023 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3024 Perl, that is the internal \ allows the delimiter to be included in the
3025 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3026 the final delimiter, add \ to the end of the pattern (to test the error).
3027
3028 4. Added the convenience functions for extracting substrings after a successful
3029 match. Updated pcretest to make it able to test these functions.
3030
3031
3032 Version 2.02 14-Jan-99
3033 ----------------------
3034
3035 1. Initialized the working variables associated with each extraction so that
3036 their saving and restoring doesn't refer to uninitialized store.
3037
3038 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3039 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3040 fix the problem.
3041
3042 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3043 calls, and wasn't printing the correct value for compiling calls. Increased the
3044 default value of LOOPREPEAT, and the number of significant figures in the
3045 times.
3046
3047 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3048
3049 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3050 a building problem on Windows NT with a FAT file system.
3051
3052
3053 Version 2.01 21-Oct-98
3054 ----------------------
3055
3056 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3057 to character tables built by pcre_maketables() in the current locale. If NULL
3058 is passed, the default tables are used.
3059
3060
3061 Version 2.00 24-Sep-98
3062 ----------------------
3063
3064 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3065 it any more.
3066
3067 2. Allow quantification of (?>) groups, and make it work correctly.
3068
3069 3. The first character computation wasn't working for (?>) groups.
3070
3071 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3072 end of the subject) and add 5.005's \z, which really does match only at the
3073 very end of the subject.
3074
3075 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3076
3077 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3078 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3079 localized options. All options to pcre_study() were also removed.
3080
3081 7. Add other new features from 5.005:
3082
3083 $(?<= positive lookbehind
3084 $(?<! negative lookbehind
3085 (?imsx-imsx) added the unsetting capability
3086 such a setting is global if at outer level; local otherwise
3087 (?imsx-imsx:) non-capturing groups with option setting
3088 (?(cond)re|re) conditional pattern matching
3089
3090 A backreference to itself in a repeated group matches the previous
3091 captured string.
3092
3093 8. General tidying up of studying (both automatic and via "study")
3094 consequential on the addition of new assertions.
3095
3096 9. As in 5.005, unlimited repeated groups that could match an empty substring
3097 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3098 runtime if any iteration does actually match an empty substring.
3099
3100 10. Include the RunTest script in the distribution.
3101
3102 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3103 discrepancies, some of which were old and were also with respect to 5.004. They
3104 have now been fixed.
3105
3106
3107 Version 1.09 28-Apr-98
3108 ----------------------
3109
3110 1. A negated single character class followed by a quantifier with a minimum
3111 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3112 program crashes, or just wrong answers. This did not apply to negated classes
3113 containing more than one character, or to minima other than one.
3114
3115
3116 Version 1.08 27-Mar-98
3117 ----------------------
3118
3119 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3120
3121 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3122 latter must appear before anything that relies on it in the pattern.
3123
3124
3125 Version 1.07 16-Feb-98
3126 ----------------------
3127
3128 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3129 repeat of a potentially empty string).
3130
3131
3132 Version 1.06 23-Jan-98
3133 ----------------------
3134
3135 1. Added Markus Oberhumer's little patches for C++.
3136
3137 2. Literal strings longer than 255 characters were broken.
3138
3139
3140 Version 1.05 23-Dec-97
3141 ----------------------
3142
3143 1. Negated character classes containing more than one character were failing if
3144 PCRE_CASELESS was set at run time.
3145
3146
3147 Version 1.04 19-Dec-97
3148 ----------------------
3149
3150 1. Corrected the man page, where some "const" qualifiers had been omitted.
3151
3152 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3153 input syntax.
3154
3155 3. Fixed memory leak which occurred when a regex with back references was
3156 matched with an offsets vector that wasn't big enough. The temporary memory
3157 that is used in this case wasn't being freed if the match failed.
3158
3159 4. Tidied pcretest to ensure it frees memory that it gets.
3160
3161 5. Temporary memory was being obtained in the case where the passed offsets
3162 vector was exactly big enough.
3163
3164 6. Corrected definition of offsetof() from change 5 below.
3165
3166 7. I had screwed up change 6 below and broken the rules for the use of
3167 setjmp(). Now fixed.
3168
3169
3170 Version 1.03 18-Dec-97
3171 ----------------------
3172
3173 1. A erroneous regex with a missing opening parenthesis was correctly
3174 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3175 on some systems.
3176
3177 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3178 it was reported that one broken compiler failed on the former because "code" is
3179 also an independent variable.
3180
3181 3. The erroneous regex a[]b caused an array overrun reference.
3182
3183 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3184 fail on data ending with that character. (It was going on too far, and checking
3185 the next character, typically a binary zero.) This was specific to the
3186 optimized code for single-character negative classes.
3187
3188 5. Added a contributed patch from the TIN world which does the following:
3189
3190 + Add an undef for memmove, in case the the system defines a macro for it.
3191
3192 + Add a definition of offsetof(), in case there isn't one. (I don't know
3193 the reason behind this - offsetof() is part of the ANSI standard - but
3194 it does no harm).
3195
3196 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3197 most of the places where whitespace preceded '#'. I have given up and
3198 allowed the remaining 2 cases to be at the margin.
3199
3200 + Rename some variables in pcre to eliminate shadowing. This seems very
3201 pedantic, but does no harm, of course.
3202
3203 6. Moved the call to setjmp() into its own function, to get rid of warnings
3204 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3205
3206 7. Constructs such as \d{8,} were compiling into the equivalent of
3207 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3208 outcome, but in this particular case used more store than had been allocated,
3209 which caused the bug to be discovered because it threw up an internal error.
3210
3211 8. The debugging code in both pcre and pcretest for outputting the compiled
3212 form of a regex was going wrong in the case of back references followed by
3213 curly-bracketed repeats.
3214
3215
3216 Version 1.02 12-Dec-97
3217 ----------------------
3218
3219 1. Typos in pcre.3 and comments in the source fixed.
3220
3221 2. Applied a contributed patch to get rid of places where it used to remove
3222 'const' from variables, and fixed some signed/unsigned and uninitialized
3223 variable warnings.
3224
3225 3. Added the "runtest" target to Makefile.
3226
3227 4. Set default compiler flag to -O2 rather than just -O.
3228
3229
3230 Version 1.01 19-Nov-97
3231 ----------------------
3232
3233 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3234 like /([ab]*)*/, that is, for classes with more than one character in them.
3235
3236 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3237 as /((?>a*))*/ (a PCRE_EXTRA facility).
3238
3239
3240 Version 1.00 18-Nov-97
3241 ----------------------
3242
3243 1. Added compile-time macros to support systems such as SunOS4 which don't have
3244 memmove() or strerror() but have other things that can be used instead.
3245
3246 2. Arranged that "make clean" removes the executables.
3247
3248
3249 Version 0.99 27-Oct-97
3250 ----------------------
3251
3252 1. Fixed bug in code for optimizing classes with only one character. It was
3253 initializing a 32-byte map regardless, which could cause it to run off the end
3254 of the memory it had got.
3255
3256 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3257
3258
3259 Version 0.98 22-Oct-97
3260 ----------------------
3261
3262 1. Fixed bug in code for handling temporary memory usage when there are more
3263 back references than supplied space in the ovector. This could cause segfaults.
3264
3265
3266 Version 0.97 21-Oct-97
3267 ----------------------
3268
3269 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3270
3271 2. Optimized negated single characters not to use a bit map.
3272
3273 3. Brought error texts together as macro definitions; clarified some of them;
3274 fixed one that was wrong - it said "range out of order" when it meant "invalid
3275 escape sequence".
3276
3277 4. Changed some char * arguments to const char *.
3278
3279 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3280
3281 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3282 pcretest.
3283
3284
3285 Version 0.96 16-Oct-97
3286 ----------------------
3287
3288 1. Added a simple "pgrep" utility to the distribution.
3289
3290 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3291 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3292 where "ddd" means "one or more decimal digits".
3293
3294 3. Fixed serious bug. If a pattern had a back reference, but the call to
3295 pcre_exec() didn't supply a large enough ovector to record the related
3296 identifying subpattern, the match always failed. PCRE now remembers the number
3297 of the largest back reference, and gets some temporary memory in which to save
3298 the offsets during matching if necessary, in order to ensure that
3299 backreferences always work.
3300
3301 4. Increased the compatibility with Perl in a number of ways:
3302
3303 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3304 to request this handling. The option can be set at compile or exec time.
3305
3306 (b) $ matches before a terminating newline by default; an option
3307 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3308 mode). The option can be set at compile or exec time.
3309
3310 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3311 the same as Perl's. If the decimal number it represents is less than 10
3312 or there aren't that many previous left capturing parentheses, an octal
3313 escape is read. Inside a character class, it's always an octal escape,
3314 even if it is a single digit.
3315
3316 (d) An escaped but undefined alphabetic character is taken as a literal,
3317 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3318 escapes.
3319
3320 (e) {0} is now permitted. (The previous item is removed from the compiled
3321 pattern).
3322
3323 5. Changed all the names of code files so that the basic parts are no longer
3324 than 10 characters, and abolished the teeny "globals.c" file.
3325
3326 6. Changed the handling of character classes; they are now done with a 32-byte
3327 bit map always.
3328
3329 7. Added the -d and /D options to pcretest to make it possible to look at the
3330 internals of compilation without having to recompile pcre.
3331
3332
3333 Version 0.95 23-Sep-97
3334 ----------------------
3335
3336 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3337 \x20 at the start of a run of normal characters. These were being treated as
3338 real characters, instead of the source characters being re-checked.
3339
3340
3341 Version 0.94 18-Sep-97
3342 ----------------------
3343
3344 1. The functions are now thread-safe, with the caveat that the global variables
3345 containing pointers to malloc() and free() or alternative functions are the
3346 same for all threads.
3347
3348 2. Get pcre_study() to generate a bitmap of initial characters for non-
3349 anchored patterns when this is possible, and use it if passed to pcre_exec().
3350
3351
3352 Version 0.93 15-Sep-97
3353 ----------------------
3354
3355 1. /(b)|(:+)/ was computing an incorrect first character.
3356
3357 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3358 but not actually doing anything yet.
3359
3360 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3361 as Perl does (e.g. [-az] or [az-]).
3362
3363 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3364 all possible positions.
3365
3366 5. Split up into different modules to avoid including unneeded functions in a
3367 compiled binary. However, compile and exec are still in one module. The "study"
3368 function is split off.
3369
3370 6. The character tables are now in a separate module whose source is generated
3371 by an auxiliary program - but can then be edited by hand if required. There are
3372 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3373 toupper() in the code.
3374
3375 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3376 make them global. Abolish the function for setting them, as the caller can now
3377 set them directly.
3378
3379
3380 Version 0.92 11-Sep-97
3381 ----------------------
3382
3383 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3384 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3385
3386 2. Caseless matching was not working in character classes if the characters in
3387 the pattern were in upper case.
3388
3389 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3390
3391 4. Make PCRE_ANCHORED public and accept as a compile option.
3392
3393 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3394 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3395 pass them.
3396
3397 6. Give an error if bad option bits passed at compile or run time.
3398
3399 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3400 pcretest to cause it to pass that flag.
3401
3402 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3403 options, and the first character, if set.
3404
3405 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3406
3407
3408 Version 0.91 10-Sep-97
3409 ----------------------
3410
3411 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3412 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3413
3414 2. PCRE was looping on encountering an indefinitely repeated back reference to
3415 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3416 Perl does - treats the match as successful.
3417
3418 ****

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