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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 408 - (show annotations)
Fri Mar 27 17:52:03 2009 UTC (6 years, 3 months ago) by ph10
File size: 165101 byte(s)
Error occurred while calculating annotation data.
Fix forward references when duplicate group numbers are involved.
1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.9 xx-xxx-09
5 ---------------------
6
7 1. When building with support for bzlib/zlib (pcregrep) and/or readline
8 (pcretest), all targets were linked against these libraries. This included
9 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10 libraries. This caused unwanted dependencies to be created. This problem
11 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12 pcretest is linked with readline.
13
14 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
15 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16 moved outside it again, because FALSE and TRUE are already defined in AIX,
17 but BOOL is not.
18
19 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20 PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21
22 4. The pcregrep documentation said spaces were inserted as well as colons (or
23 hyphens) following file names and line numbers when outputting matching
24 lines. This is not true; no spaces are inserted. I have also clarified the
25 wording for the --colour (or --color) option.
26
27 5. In pcregrep, when --colour was used with -o, the list of matching strings
28 was not coloured; this is different to GNU grep, so I have changed it to be
29 the same.
30
31 6. When --colo(u)r was used in pcregrep, only the first matching substring in
32 each matching line was coloured. Now it goes on to look for further matches
33 of any of the test patterns, which is the same behaviour as GNU grep.
34
35 7. A pattern that could match an empty string could cause pcregrep to loop; it
36 doesn't make sense to accept an empty string match in pcregrep, so I have
37 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38 seems to be how GNU grep behaves.
39
40 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41 start or after a newline", because the conditional assertion was not being
42 correctly handled. The rule now is that both the assertion and what follows
43 in the first alternative must satisfy the test.
44
45 9. If auto-callout was enabled in a pattern with a conditional group whose
46 condition was an assertion, PCRE could crash during matching, both with
47 pcre_exec() and pcre_dfa_exec().
48
49 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
50 used for matching.
51
52 11. Unicode property support in character classes was not working for
53 characters (bytes) greater than 127 when not in UTF-8 mode.
54
55 12. Added the -M command line option to pcretest.
56
57 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
58
59 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
60
61 16. Added comments and documentation about mis-use of no_arg in the C++
62 wrapper.
63
64 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
65 from Martin Jerabek that uses macro names for all relevant character and
66 string constants.
67
68 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
69 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
70 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
71 these, but not everybody uses configure.
72
73 19. A conditional group that had only one branch was not being correctly
74 recognized as an item that could match an empty string. This meant that an
75 enclosing group might also not be so recognized, causing infinite looping
76 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
77 with the subject "ab", where knowledge that the repeated group can match
78 nothing is needed in order to break the loop.
79
80 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
81 exec(), but without supplying a callout function, matching went wrong.
82
83 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
84 leak if the size of the offset vector was greater than 30. When the vector
85 is smaller, the saved offsets during recursion go onto a local stack
86 vector, but for larger vectors malloc() is used. It was failing to free
87 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
88 error, in fact).
89
90 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
91 heapframe that is used only when UTF-8 support is enabled. This caused no
92 problem, but was untidy.
93
94 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
95 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
96 included within another project.
97
98 24. Steven Van Ingelgem's patches to add more options to the CMake support,
99 slightly modified by me:
100
101 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
102 not building pcregrep.
103
104 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
105 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
106
107 25. Forward references, both numeric and by name, in patterns that made use of
108 duplicate group numbers, could behave incorrectly or give incorrect errors,
109 because when scanning forward to find the reference group, PCRE was not
110 taking into account the duplicate group numbers. A pattern such as
111 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
112
113
114
115 Version 7.8 05-Sep-08
116 ---------------------
117
118 1. Replaced UCP searching code with optimized version as implemented for Ad
119 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
120 stage table and inline lookup instead of a function, giving speed ups of 2
121 to 5 times on some simple patterns that I tested. Permission was given to
122 distribute the MultiStage2.py script that generates the tables (it's not in
123 the tarball, but is in the Subversion repository).
124
125 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
126 scripts.
127
128 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
129 a group with a zero qualifier. The result of the study could be incorrect,
130 or the function might crash, depending on the pattern.
131
132 4. Caseless matching was not working for non-ASCII characters in back
133 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
134 It now works when Unicode Property Support is available.
135
136 5. In pcretest, an escape such as \x{de} in the data was always generating
137 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
138 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
139 truncation.
140
141 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
142
143 7. Added two (int) casts to pcregrep when printing the difference of two
144 pointers, in case they are 64-bit values.
145
146 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
147 test 2 if it fails.
148
149 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
150 and a #define of that name to empty if it is not externally set. This is to
151 allow users of MSVC to set it if necessary.
152
153 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
154 the convenience functions in the pcre_get.c source file.
155
156 11. An option change at the start of a pattern that had top-level alternatives
157 could cause overwriting and/or a crash. This command provoked a crash in
158 some environments:
159
160 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
161
162 This potential security problem was recorded as CVE-2008-2371.
163
164 12. For a pattern where the match had to start at the beginning or immediately
165 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
166 pcre_dfa_exec() could read past the end of the passed subject if there was
167 no match. To help with detecting such bugs (e.g. with valgrind), I modified
168 pcretest so that it places the subject at the end of its malloc-ed buffer.
169
170 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
171 exec() might read past the end of the data buffer in UTF-8 mode.
172
173 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
174 the data contained the byte 0x85 as part of a UTF-8 character within its
175 first line. This applied both to normal and DFA matching.
176
177 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
178 /^[^d]*?$/8 failed to match "abc".
179
180 16. Added a missing copyright notice to pcrecpp_internal.h.
181
182 17. Make it more clear in the documentation that values returned from
183 pcre_exec() in ovector are byte offsets, not character counts.
184
185 18. Tidied a few places to stop certain compilers from issuing warnings.
186
187 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
188 supplied by Stefan Weber. I made a further small update for 7.8 because
189 there is a change of source arrangements: the pcre_searchfuncs.c module is
190 replaced by pcre_ucd.c.
191
192
193 Version 7.7 07-May-08
194 ---------------------
195
196 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
197 a string to a long long, pretend we don't even have a long long." This is
198 done by checking for the strtoq, strtoll, and _strtoi64 functions.
199
200 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
201 pre-7.6 versions, which defined a global no_arg variable instead of putting
202 it in the RE class. (See also #8 below.)
203
204 3. Remove a line of dead code, identified by coverity and reported by Nuno
205 Lopes.
206
207 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
208
209 (1) The include/exclude patterns were being applied to the whole pathnames
210 of files, instead of just to the final components.
211
212 (2) If there was more than one level of directory, the subdirectories were
213 skipped unless they satisfied the include/exclude conditions. This is
214 inconsistent with GNU grep (and could even be seen as contrary to the
215 pcregrep specification - which I improved to make it absolutely clear).
216 The action now is always to scan all levels of directory, and just
217 apply the include/exclude patterns to regular files.
218
219 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
220 --exclude_dir in the tests to avoid scanning .svn directories.
221
222 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
223 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
224 doesn't support NULs in patterns.
225
226 7. Added some missing "const"s to declarations of static tables in
227 pcre_compile.c and pcre_dfa_exec.c.
228
229 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
230 caused by fix #2 above. (Subsequently also a second patch to fix the
231 first patch. And a third patch - this was a messy problem.)
232
233 9. Applied Craig's patch to remove the use of push_back().
234
235 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
236 matching function regexec().
237
238 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
239 which, however, unlike Perl's \g{...}, are subroutine calls, not back
240 references. PCRE supports relative numbers with this syntax (I don't think
241 Oniguruma does).
242
243 12. Previously, a group with a zero repeat such as (...){0} was completely
244 omitted from the compiled regex. However, this means that if the group
245 was called as a subroutine from elsewhere in the pattern, things went wrong
246 (an internal error was given). Such groups are now left in the compiled
247 pattern, with a new opcode that causes them to be skipped at execution
248 time.
249
250 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
251 to the way PCRE behaves:
252
253 (a) A lone ] character is dis-allowed (Perl treats it as data).
254
255 (b) A back reference to an unmatched subpattern matches an empty string
256 (Perl fails the current match path).
257
258 (c) A data ] in a character class must be notated as \] because if the
259 first data character in a class is ], it defines an empty class. (In
260 Perl it is not possible to have an empty class.) The empty class []
261 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
262 The negative empty class [^] matches any one character, independently
263 of the DOTALL setting.
264
265 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
266 non-existent subpattern following a character class starting with ']' and
267 containing () gave an internal compiling error instead of "reference to
268 non-existent subpattern". Fortunately, when the pattern did exist, the
269 compiled code was correct. (When scanning forwards to check for the
270 existencd of the subpattern, it was treating the data ']' as terminating
271 the class, so got the count wrong. When actually compiling, the reference
272 was subsequently set up correctly.)
273
274 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
275 it was being rejected as not supported by pcre_dfa_exec(), even though
276 other assertions are supported. I have made pcre_dfa_exec() support
277 (*FAIL).
278
279 16. The implementation of 13c above involved the invention of a new opcode,
280 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
281 cannot be changed at match time, I realized I could make a small
282 improvement to matching performance by compiling OP_ALLANY instead of
283 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
284 on the OP_ANY path.
285
286 17. Compiling pcretest on Windows with readline support failed without the
287 following two fixes: (1) Make the unistd.h include conditional on
288 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
289
290 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
291 ncurses library to be included for pcretest when ReadLine support is
292 requested, but also to allow for it to be overridden. This patch came from
293 Daniel Bergström.
294
295 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
296 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
297 any errors with the current Unicode tables. Thanks to Peter Kankowski for
298 spotting this.
299
300
301 Version 7.6 28-Jan-08
302 ---------------------
303
304 1. A character class containing a very large number of characters with
305 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
306 overflow.
307
308 2. Patch to cut out the "long long" test in pcrecpp_unittest when
309 HAVE_LONG_LONG is not defined.
310
311 3. Applied Christian Ehrlicher's patch to update the CMake build files to
312 bring them up to date and include new features. This patch includes:
313
314 - Fixed PH's badly added libz and libbz2 support.
315 - Fixed a problem with static linking.
316 - Added pcredemo. [But later removed - see 7 below.]
317 - Fixed dftables problem and added an option.
318 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
319 HAVE_LONG_LONG.
320 - Added readline support for pcretest.
321 - Added an listing of the option settings after cmake has run.
322
323 4. A user submitted a patch to Makefile that makes it easy to create
324 "pcre.dll" under mingw when using Configure/Make. I added stuff to
325 Makefile.am that cause it to include this special target, without
326 affecting anything else. Note that the same mingw target plus all
327 the other distribution libraries and programs are now supported
328 when configuring with CMake (see 6 below) instead of with
329 Configure/Make.
330
331 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
332 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
333 exported in the Windows port". It has not yet been confirmed that the patch
334 solves the problem, but it does no harm.
335
336 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
337 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
338 with CMake, and also correct the comment about stack recursion.
339
340 7. Remove the automatic building of pcredemo from the ./configure system and
341 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
342 of a program that users should build themselves after PCRE is installed, so
343 building it automatically is not really right. What is more, it gave
344 trouble in some build environments.
345
346 8. Further tidies to CMakeLists.txt from Sheri and Christian.
347
348
349 Version 7.5 10-Jan-08
350 ---------------------
351
352 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
353 values in parens when parsing an RE using the C++ wrapper."
354
355 2. Negative specials like \S did not work in character classes in UTF-8 mode.
356 Characters greater than 255 were excluded from the class instead of being
357 included.
358
359 3. The same bug as (2) above applied to negated POSIX classes such as
360 [:^space:].
361
362 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
363 defined or documented. It seems to have been a typo for PCRE_STATIC, so
364 I have changed it.
365
366 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
367 first named subpattern) and a construct such as (?&a) would reference the
368 first named subpattern whose name started with "a" (in other words, the
369 length check was missing). Both these problems are fixed. "Subpattern name
370 expected" is now given for (?&) (a zero-length name), and this patch also
371 makes it give the same error for \k'' (previously it complained that that
372 was a reference to a non-existent subpattern).
373
374 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
375 this is right because (?- can be followed by option settings as well as by
376 digits. I have, however, made the messages clearer.
377
378 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
379 than the number used in the conditional) now cause a compile-time error.
380 This is actually not compatible with Perl, which accepts such patterns, but
381 treats the conditional as always being FALSE (as PCRE used to), but it
382 seems to me that giving a diagnostic is better.
383
384 8. Change "alphameric" to the more common word "alphanumeric" in comments
385 and messages.
386
387 9. Fix two occurrences of "backslash" in comments that should have been
388 "backspace".
389
390 10. Remove two redundant lines of code that can never be obeyed (their function
391 was moved elsewhere).
392
393 11. The program that makes PCRE's Unicode character property table had a bug
394 which caused it to generate incorrect table entries for sequences of
395 characters that have the same character type, but are in different scripts.
396 It amalgamated them into a single range, with the script of the first of
397 them. In other words, some characters were in the wrong script. There were
398 thirteen such cases, affecting characters in the following ranges:
399
400 U+002b0 - U+002c1
401 U+0060c - U+0060d
402 U+0061e - U+00612
403 U+0064b - U+0065e
404 U+0074d - U+0076d
405 U+01800 - U+01805
406 U+01d00 - U+01d77
407 U+01d9b - U+01dbf
408 U+0200b - U+0200f
409 U+030fc - U+030fe
410 U+03260 - U+0327f
411 U+0fb46 - U+0fbb1
412 U+10450 - U+1049d
413
414 12. The -o option (show only the matching part of a line) for pcregrep was not
415 compatible with GNU grep in that, if there was more than one match in a
416 line, it showed only the first of them. It now behaves in the same way as
417 GNU grep.
418
419 13. If the -o and -v options were combined for pcregrep, it printed a blank
420 line for every non-matching line. GNU grep prints nothing, and pcregrep now
421 does the same. The return code can be used to tell if there were any
422 non-matching lines.
423
424 14. Added --file-offsets and --line-offsets to pcregrep.
425
426 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
427 infinitely looping recursion. The bug was that positive lookaheads were not
428 being skipped when checking for a possible empty match (negative lookaheads
429 and both kinds of lookbehind were skipped).
430
431 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
432 inclusion of <windows.h> to before rather than after the definition of
433 INVALID_FILE_ATTRIBUTES (patch from David Byron).
434
435 17. Specifying a possessive quantifier with a specific limit for a Unicode
436 character property caused pcre_compile() to compile bad code, which led at
437 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
438 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
439 caused the error; without that there was no problem.
440
441 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
442
443 19. Added --enable-pcretest-libreadline.
444
445 20. In pcrecpp.cc, the variable 'count' was incremented twice in
446 RE::GlobalReplace(). As a result, the number of replacements returned was
447 double what it should be. I removed one of the increments, but Craig sent a
448 later patch that removed the other one (the right fix) and added unit tests
449 that check the return values (which was not done before).
450
451 21. Several CMake things:
452
453 (1) Arranged that, when cmake is used on Unix, the libraries end up with
454 the names libpcre and libpcreposix, not just pcre and pcreposix.
455
456 (2) The above change means that pcretest and pcregrep are now correctly
457 linked with the newly-built libraries, not previously installed ones.
458
459 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
460
461 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
462 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
463 UTF-8 newline character). The key issue is that the pattern starts .*;
464 this means that the match must be either at the beginning, or after a
465 newline. The bug was in the code for advancing after a failed match and
466 checking that the new position followed a newline. It was not taking
467 account of UTF-8 characters correctly.
468
469 23. PCRE was behaving differently from Perl in the way it recognized POSIX
470 character classes. PCRE was not treating the sequence [:...:] as a
471 character class unless the ... were all letters. Perl, however, seems to
472 allow any characters between [: and :], though of course it rejects as
473 unknown any "names" that contain non-letters, because all the known class
474 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
475 for example, whereas PCRE did not - it did not recognize a POSIX character
476 class. This seemed a bit dangerous, so the code has been changed to be
477 closer to Perl. The behaviour is not identical to Perl, because PCRE will
478 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
479 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
480 Perl does, and where it didn't before.
481
482 24. Rewrite so as to remove the single use of %n from pcregrep because in some
483 Windows environments %n is disabled by default.
484
485
486 Version 7.4 21-Sep-07
487 ---------------------
488
489 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
490 means that a class such as [\s] counted as "explicit reference to CR or
491 LF". That isn't really right - the whole point of the change was to try to
492 help when there was an actual mention of one of the two characters. So now
493 the change happens only if \r or \n (or a literal CR or LF) character is
494 encountered.
495
496 2. The 32-bit options word was also used for 6 internal flags, but the numbers
497 of both had grown to the point where there were only 3 bits left.
498 Fortunately, there was spare space in the data structure, and so I have
499 moved the internal flags into a new 16-bit field to free up more option
500 bits.
501
502 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
503 but did not set the internal JCHANGED flag - either of these is enough to
504 control the way the "get" function works - but the PCRE_INFO_JCHANGED
505 facility is supposed to tell if (?J) was ever used, so now (?J) at the
506 start sets both bits.
507
508 4. Added options (at build time, compile time, exec time) to change \R from
509 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
510
511 5. doc/pcresyntax.html was missing from the distribution.
512
513 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
514 compatibility, even though it is no longer used.
515
516 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
517 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
518 windows.h file is present (where different names are used). [This was
519 reversed later after testing - see 16 below.]
520
521 8. Changed all #include <config.h> to #include "config.h". There were also
522 some further <pcre.h> cases that I changed to "pcre.h".
523
524 9. When pcregrep was used with the --colour option, it missed the line ending
525 sequence off the lines that it output.
526
527 10. It was pointed out to me that arrays of string pointers cause lots of
528 relocations when a shared library is dynamically loaded. A technique of
529 using a single long string with a table of offsets can drastically reduce
530 these. I have refactored PCRE in four places to do this. The result is
531 dramatic:
532
533 Originally: 290
534 After changing UCP table: 187
535 After changing error message table: 43
536 After changing table of "verbs" 36
537 After changing table of Posix names 22
538
539 Thanks to the folks working on Gregex for glib for this insight.
540
541 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
542 unicode-properties was also set.
543
544 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
545
546 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
547 checked only for CRLF.
548
549 14. Added casts to pcretest.c to avoid compiler warnings.
550
551 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
552
553 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
554 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
555 entirely. This removes changes made in 7 above.
556
557 17. The CMake files have been updated, and there is now more information about
558 building with CMake in the NON-UNIX-USE document.
559
560
561 Version 7.3 28-Aug-07
562 ---------------------
563
564 1. In the rejigging of the build system that eventually resulted in 7.1, the
565 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
566 brackets there is not right, since it causes compilers to look for an
567 installed pcre.h, not the version that is in the source that is being
568 compiled (which of course may be different). I have changed it back to:
569
570 #include "pcre.h"
571
572 I have a vague recollection that the change was concerned with compiling in
573 different directories, but in the new build system, that is taken care of
574 by the VPATH setting the Makefile.
575
576 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
577 when the subject happened to end in the byte 0x85 (e.g. if the last
578 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
579 characters but of course it shouldn't be taken as a newline when it is part
580 of another character. The bug was that, for an unlimited repeat of . in
581 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
582 characters when looking for a newline.
583
584 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
585
586 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
587 in debug output.
588
589 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
590 long printing in the pcrecpp unittest when running under MinGW.
591
592 6. ESC_K was left out of the EBCDIC table.
593
594 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
595 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
596 limit also applies to "virtual nesting" when a pattern is recursive, and in
597 this case 1000 isn't so big. I have been able to remove this limit at the
598 expense of backing off one optimization in certain circumstances. Normally,
599 when pcre_exec() would call its internal match() function recursively and
600 immediately return the result unconditionally, it uses a "tail recursion"
601 feature to save stack. However, when a subpattern that can match an empty
602 string has an unlimited repetition quantifier, it no longer makes this
603 optimization. That gives it a stack frame in which to save the data for
604 checking that an empty string has been matched. Previously this was taken
605 from the 1000-entry workspace that had been reserved. So now there is no
606 explicit limit, but more stack is used.
607
608 8. Applied Daniel's patches to solve problems with the import/export magic
609 syntax that is required for Windows, and which was going wrong for the
610 pcreposix and pcrecpp parts of the library. These were overlooked when this
611 problem was solved for the main library.
612
613 9. There were some crude static tests to avoid integer overflow when computing
614 the size of patterns that contain repeated groups with explicit upper
615 limits. As the maximum quantifier is 65535, the maximum group length was
616 set at 30,000 so that the product of these two numbers did not overflow a
617 32-bit integer. However, it turns out that people want to use groups that
618 are longer than 30,000 bytes (though not repeat them that many times).
619 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
620 made it possible to implement the integer overflow checks in a much more
621 dynamic way, which I have now done. The artificial limitation on group
622 length has been removed - we now have only the limit on the total length of
623 the compiled pattern, which depends on the LINK_SIZE setting.
624
625 10. Fixed a bug in the documentation for get/copy named substring when
626 duplicate names are permitted. If none of the named substrings are set, the
627 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
628 empty string.
629
630 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
631 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
632 because the ] is interpreted as the first data character and the
633 terminating ] is not found. PCRE has been made compatible with Perl in this
634 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
635 cause memory overwriting.
636
637 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
638 string has been matched (to stop an infinite loop). It was not recognizing
639 a conditional subpattern that could match an empty string if that
640 subpattern was within another subpattern. For example, it looped when
641 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
642 condition was not nested. This bug has been fixed.
643
644 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
645 past the start of the subject in the presence of bytes with the top bit
646 set, for example "\x8aBCD".
647
648 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
649 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
650
651 14. Optimized (?!) to (*FAIL).
652
653 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
654 This restricts code points to be within the range 0 to 0x10FFFF, excluding
655 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
656 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
657 does: it's just the validity check that is more restrictive.
658
659 16. Inserted checks for integer overflows during escape sequence (backslash)
660 processing, and also fixed erroneous offset values for syntax errors during
661 backslash processing.
662
663 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
664 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
665
666 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
667 caused an overrun.
668
669 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
670 something other than just ASCII characters) inside a group that had an
671 unlimited repeat caused a loop at compile time (while checking to see
672 whether the group could match an empty string).
673
674 20. Debugging a pattern containing \p or \P could cause a crash. For example,
675 [\P{Any}] did so. (Error in the code for printing property names.)
676
677 21. An orphan \E inside a character class could cause a crash.
678
679 22. A repeated capturing bracket such as (A)? could cause a wild memory
680 reference during compilation.
681
682 23. There are several functions in pcre_compile() that scan along a compiled
683 expression for various reasons (e.g. to see if it's fixed length for look
684 behind). There were bugs in these functions when a repeated \p or \P was
685 present in the pattern. These operators have additional parameters compared
686 with \d, etc, and these were not being taken into account when moving along
687 the compiled data. Specifically:
688
689 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
690 length.
691
692 (b) An item such as \pL+ within a repeated group could cause crashes or
693 loops.
694
695 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
696 "reference to non-existent subpattern" error.
697
698 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
699
700 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
701 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
702
703 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
704
705 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
706 character were causing crashes (broken optimization).
707
708 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
709 \p or \P) caused a compile-time loop.
710
711 28. More problems have arisen in unanchored patterns when CRLF is a valid line
712 break. For example, the unstudied pattern [\r\n]A does not match the string
713 "\r\nA" because change 7.0/46 below moves the current point on by two
714 characters after failing to match at the start. However, the pattern \nA
715 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
716 the same is true. There doesn't seem any very clean way out of this, but
717 what I have chosen to do makes the common cases work: PCRE now takes note
718 of whether there can be an explicit match for \r or \n anywhere in the
719 pattern, and if so, 7.0/46 no longer applies. As part of this change,
720 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
721 pattern has explicit CR or LF references.
722
723 29. Added (*CR) etc for changing newline setting at start of pattern.
724
725
726 Version 7.2 19-Jun-07
727 ---------------------
728
729 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
730 which is apparently normally available under Windows.
731
732 2. Re-jig the pcregrep tests with different newline settings in an attempt
733 to make them independent of the local environment's newline setting.
734
735 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
736
737 4. Some of the "internals" tests were previously cut out when the link size
738 was not 2, because the output contained actual offsets. The recent new
739 "Z" feature of pcretest means that these can be cut out, making the tests
740 usable with all link sizes.
741
742 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
743 stack recursion. This gives a massive performance boost under BSD, but just
744 a small improvement under Linux. However, it saves one field in the frame
745 in all cases.
746
747 6. Added more features from the forthcoming Perl 5.10:
748
749 (a) (?-n) (where n is a string of digits) is a relative subroutine or
750 recursion call. It refers to the nth most recently opened parentheses.
751
752 (b) (?+n) is also a relative subroutine call; it refers to the nth next
753 to be opened parentheses.
754
755 (c) Conditions that refer to capturing parentheses can be specified
756 relatively, for example, (?(-2)... or (?(+3)...
757
758 (d) \K resets the start of the current match so that everything before
759 is not part of it.
760
761 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
762
763 (f) \g{name} is another synonym - part of Perl 5.10's unification of
764 reference syntax.
765
766 (g) (?| introduces a group in which the numbering of parentheses in each
767 alternative starts with the same number.
768
769 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
770
771 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
772 PCRE_INFO_JCHANGED.
773
774 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
775 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
776 for detecting groups that can match an empty string.
777
778 9. A pattern with a very large number of alternatives (more than several
779 hundred) was running out of internal workspace during the pre-compile
780 phase, where pcre_compile() figures out how much memory will be needed. A
781 bit of new cunning has reduced the workspace needed for groups with
782 alternatives. The 1000-alternative test pattern now uses 12 bytes of
783 workspace instead of running out of the 4096 that are available.
784
785 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
786
787 11. Applied patch from Google to remove an optimization that didn't quite work.
788 The report of the bug said:
789
790 pcrecpp::RE("a*").FullMatch("aaa") matches, while
791 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
792 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
793
794 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
795 it matched the wrong number of bytes.
796
797
798 Version 7.1 24-Apr-07
799 ---------------------
800
801 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
802 that is more "standard", making use of automake and other Autotools. There
803 is some re-arrangement of the files and adjustment of comments consequent
804 on this.
805
806 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
807 for recursive directory scanning broke on some systems because the files
808 are not scanned in any specific order and on different systems the order
809 was different. A call to "sort" has been inserted into RunGrepTest for the
810 approprate test as a short-term fix. In the longer term there may be an
811 alternative.
812
813 3. I had an email from Eric Raymond about problems translating some of PCRE's
814 man pages to HTML (despite the fact that I distribute HTML pages, some
815 people do their own conversions for various reasons). The problems
816 concerned the use of low-level troff macros .br and .in. I have therefore
817 removed all such uses from the man pages (some were redundant, some could
818 be replaced by .nf/.fi pairs). The 132html script that I use to generate
819 HTML has been updated to handle .nf/.fi and to complain if it encounters
820 .br or .in.
821
822 4. Updated comments in configure.ac that get placed in config.h.in and also
823 arranged for config.h to be included in the distribution, with the name
824 config.h.generic, for the benefit of those who have to compile without
825 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
826
827 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
828 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
829 makevp.bat for the current PCRE, using the additional files
830 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
831
832 6. A Windows user reported a minor discrepancy with test 2, which turned out
833 to be caused by a trailing space on an input line that had got lost in his
834 copy. The trailing space was an accident, so I've just removed it.
835
836 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
837 that is needed.
838
839 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
840 as "const" (a) because they are and (b) because it helps the PHP
841 maintainers who have recently made a script to detect big data structures
842 in the php code that should be moved to the .rodata section. I remembered
843 to update Builducptable as well, so it won't revert if ucptable.h is ever
844 re-created.
845
846 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
847 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
848 order to be able to cut out the UTF-8 tables in the latter when UTF-8
849 support is not required. This saves 1.5-2K of code, which is important in
850 some applications.
851
852 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
853 so as not to refer to the tables, even though these functions will never be
854 called when UTF-8 support is disabled. Otherwise there are problems with a
855 shared library.
856
857 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
858
859 (a) It was defining its arguments as char * instead of void *.
860
861 (b) It was assuming that all moves were upwards in memory; this was true
862 a long time ago when I wrote it, but is no longer the case.
863
864 The emulated memove() is provided for those environments that have neither
865 memmove() nor bcopy(). I didn't think anyone used it these days, but that
866 is clearly not the case, as these two bugs were recently reported.
867
868 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
869 and Detrail to create the HTML documentation, the .txt form of the man
870 pages, and it removes trailing spaces from listed files. It also creates
871 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
872 case, it wraps all the #defines with #ifndefs. This script should be run
873 before "make dist".
874
875 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
876 with Unicode property support.
877
878 (a) For a maximizing quantifier, if the two different cases of the
879 character were of different lengths in their UTF-8 codings (there are
880 some cases like this - I found 11), and the matching function had to
881 back up over a mixture of the two cases, it incorrectly assumed they
882 were both the same length.
883
884 (b) When PCRE was configured to use the heap rather than the stack for
885 recursion during matching, it was not correctly preserving the data for
886 the other case of a UTF-8 character when checking ahead for a match
887 while processing a minimizing repeat. If the check also involved
888 matching a wide character, but failed, corruption could cause an
889 erroneous result when trying to check for a repeat of the original
890 character.
891
892 13. Some tidying changes to the testing mechanism:
893
894 (a) The RunTest script now detects the internal link size and whether there
895 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
896 values substituted by "configure". (The RunGrepTest script already did
897 this for UTF-8.) The configure.ac script no longer substitutes the
898 relevant variables.
899
900 (b) The debugging options /B and /D in pcretest show the compiled bytecode
901 with length and offset values. This means that the output is different
902 for different internal link sizes. Test 2 is skipped for link sizes
903 other than 2 because of this, bypassing the problem. Unfortunately,
904 there was also a test in test 3 (the locale tests) that used /B and
905 failed for link sizes other than 2. Rather than cut the whole test out,
906 I have added a new /Z option to pcretest that replaces the length and
907 offset values with spaces. This is now used to make test 3 independent
908 of link size. (Test 2 will be tidied up later.)
909
910 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
911 segmentation fault instead of returning the appropriate error message.
912
913 15. In multiline mode when the newline sequence was set to "any", the pattern
914 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
915 This doesn't seem right; it now treats the CRLF combination as the line
916 ending, and so does not match in that case. It's only a pattern such as ^$
917 that would hit this one: something like ^ABC$ would have failed after \r
918 and then tried again after \r\n.
919
920 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
921 in an attempt to make files that differ only in their line terminators
922 compare equal. This works on Linux.
923
924 17. Under certain error circumstances pcregrep might try to free random memory
925 as it exited. This is now fixed, thanks to valgrind.
926
927 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
928 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
929 was because its rules for how to advance for /g after matching an empty
930 string at the end of a line did not allow for this case. They now check for
931 it specially.
932
933 20. pcretest is supposed to handle patterns and data of any length, by
934 extending its buffers when necessary. It was getting this wrong when the
935 buffer for a data line had to be extended.
936
937 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
938 CRLF as a newline sequence.
939
940 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
941 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
942 I have nevertheless tidied it up.
943
944 23. Added some casts to kill warnings from HP-UX ia64 compiler.
945
946 24. Added a man page for pcre-config.
947
948
949 Version 7.0 19-Dec-06
950 ---------------------
951
952 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
953 moving to gcc 4.1.1.
954
955 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
956 sys/time.h, which is documented as needed for this function. It doesn't
957 seem to matter on Linux, but it showed up on some releases of OS X.
958
959 3. It seems that there are systems where bytes whose values are greater than
960 127 match isprint() in the "C" locale. The "C" locale should be the
961 default when a C program starts up. In most systems, only ASCII printing
962 characters match isprint(). This difference caused the output from pcretest
963 to vary, making some of the tests fail. I have changed pcretest so that:
964
965 (a) When it is outputting text in the compiled version of a pattern, bytes
966 other than 32-126 are always shown as hex escapes.
967
968 (b) When it is outputting text that is a matched part of a subject string,
969 it does the same, unless a different locale has been set for the match
970 (using the /L modifier). In this case, it uses isprint() to decide.
971
972 4. Fixed a major bug that caused incorrect computation of the amount of memory
973 required for a compiled pattern when options that changed within the
974 pattern affected the logic of the preliminary scan that determines the
975 length. The relevant options are -x, and -i in UTF-8 mode. The result was
976 that the computed length was too small. The symptoms of this bug were
977 either the PCRE error "internal error: code overflow" from pcre_compile(),
978 or a glibc crash with a message such as "pcretest: free(): invalid next
979 size (fast)". Examples of patterns that provoked this bug (shown in
980 pcretest format) are:
981
982 /(?-x: )/x
983 /(?x)(?-x: \s*#\s*)/
984 /((?i)[\x{c0}])/8
985 /(?i:[\x{c0}])/8
986
987 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
988 is now done differently.
989
990 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
991 wrapper classes; (b) implement a new function in the C++ scanner that is
992 more efficient than the old way of doing things because it avoids levels of
993 recursion in the regex matching; (c) add a paragraph to the documentation
994 for the FullMatch() function.
995
996 6. The escape sequence \n was being treated as whatever was defined as
997 "newline". Not only was this contrary to the documentation, which states
998 that \n is character 10 (hex 0A), but it also went horribly wrong when
999 "newline" was defined as CRLF. This has been fixed.
1000
1001 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1002 was being set to -1 for the "end of line" case (supposedly a value that no
1003 character can have). Though this value is never used (the check for end of
1004 line is "zero bytes in current character"), it caused compiler complaints.
1005 I've changed it to 0xffffffff.
1006
1007 8. In pcre_version.c, the version string was being built by a sequence of
1008 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1009 string (as it is for production releases) called a macro with an empty
1010 argument. The C standard says the result of this is undefined. The gcc
1011 compiler treats it as an empty string (which was what was wanted) but it is
1012 reported that Visual C gives an error. The source has been hacked around to
1013 avoid this problem.
1014
1015 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1016 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1017 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1018 of them did).
1019
1020 10. Originally, pcretest opened its input and output without "b"; then I was
1021 told that "b" was needed in some environments, so it was added for release
1022 5.0 to both the input and output. (It makes no difference on Unix-like
1023 systems.) Later I was told that it is wrong for the input on Windows. I've
1024 now abstracted the modes into two macros, to make it easier to fiddle with
1025 them, and removed "b" from the input mode under Windows.
1026
1027 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1028
1029 12. Added -help and --help to pcretest as an official way of being reminded
1030 of the options.
1031
1032 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1033 and pcrecpp.cc because they annoy compilers at high warning levels.
1034
1035 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1036
1037 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1038 scripts are not C programs :-) and which was not noticed because it works
1039 on Linux.
1040
1041 16. pcretest is supposed to handle any length of pattern and data line (as one
1042 line or as a continued sequence of lines) by extending its input buffer if
1043 necessary. This feature was broken for very long pattern lines, leading to
1044 a string of junk being passed to pcre_compile() if the pattern was longer
1045 than about 50K.
1046
1047 17. I have done a major re-factoring of the way pcre_compile() computes the
1048 amount of memory needed for a compiled pattern. Previously, there was code
1049 that made a preliminary scan of the pattern in order to do this. That was
1050 OK when PCRE was new, but as the facilities have expanded, it has become
1051 harder and harder to keep it in step with the real compile phase, and there
1052 have been a number of bugs (see for example, 4 above). I have now found a
1053 cunning way of running the real compile function in a "fake" mode that
1054 enables it to compute how much memory it would need, while actually only
1055 ever using a few hundred bytes of working memory and without too many
1056 tests of the mode. This should make future maintenance and development
1057 easier. A side effect of this work is that the limit of 200 on the nesting
1058 depth of parentheses has been removed (though this was never a serious
1059 limitation, I suspect). However, there is a downside: pcre_compile() now
1060 runs more slowly than before (30% or more, depending on the pattern). I
1061 hope this isn't a big issue. There is no effect on runtime performance.
1062
1063 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1064 newline (only possible for the last line of a file) and it was a
1065 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1066
1067 19. Added additional timing features to pcretest. (1) The -tm option now times
1068 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1069 separate command line item, by a number that specifies the number of
1070 repeats to use when timing. The default is 50000; this gives better
1071 precision, but takes uncomfortably long for very large patterns.
1072
1073 20. Extended pcre_study() to be more clever in cases where a branch of a
1074 subpattern has no definite first character. For example, (a*|b*)[cd] would
1075 previously give no result from pcre_study(). Now it recognizes that the
1076 first character must be a, b, c, or d.
1077
1078 21. There was an incorrect error "recursive call could loop indefinitely" if
1079 a subpattern (or the entire pattern) that was being tested for matching an
1080 empty string contained only one non-empty item after a nested subpattern.
1081 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1082 incorrectly, because the \d was being skipped in the check.
1083
1084 22. The pcretest program now has a new pattern option /B and a command line
1085 option -b, which is equivalent to adding /B to every pattern. This causes
1086 it to show the compiled bytecode, without the additional information that
1087 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1088 is the same as /B/I).
1089
1090 23. A new optimization is now able automatically to treat some sequences such
1091 as a*b as a*+b. More specifically, if something simple (such as a character
1092 or a simple class like \d) has an unlimited quantifier, and is followed by
1093 something that cannot possibly match the quantified thing, the quantifier
1094 is automatically "possessified".
1095
1096 24. A recursive reference to a subpattern whose number was greater than 39
1097 went wrong under certain circumstances in UTF-8 mode. This bug could also
1098 have affected the operation of pcre_study().
1099
1100 25. Realized that a little bit of performance could be had by replacing
1101 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1102
1103 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1104
1105 27. Possessive quantifiers such as a++ were previously implemented by turning
1106 them into atomic groups such as ($>a+). Now they have their own opcodes,
1107 which improves performance. This includes the automatically created ones
1108 from 23 above.
1109
1110 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1111 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1112 the first matched character to be a colon. This applied both to named and
1113 numbered groups.
1114
1115 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1116
1117 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1118 building PCRE on Windows easier, so I have included it in the distribution.
1119
1120 31. There is now a check in pcretest against a ridiculously large number being
1121 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1122 loop, the loop is abandoned.
1123
1124 32. Forward references to subpatterns in conditions such as (?(2)...) where
1125 subpattern 2 is defined later cause pcre_compile() to search forwards in
1126 the pattern for the relevant set of parentheses. This search went wrong
1127 when there were unescaped parentheses in a character class, parentheses
1128 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1129
1130 33. "Subroutine" calls and backreferences were previously restricted to
1131 referencing subpatterns earlier in the regex. This restriction has now
1132 been removed.
1133
1134 34. Added a number of extra features that are going to be in Perl 5.10. On the
1135 whole, these are just syntactic alternatives for features that PCRE had
1136 previously implemented using the Python syntax or my own invention. The
1137 other formats are all retained for compatibility.
1138
1139 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1140 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1141 also .NET compatible.
1142
1143 (b) A recursion or subroutine call to a named group can now be defined as
1144 (?&name) as well as (?P>name).
1145
1146 (c) A backreference to a named group can now be defined as \k<name> or
1147 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1148 5.10, are also .NET compatible.
1149
1150 (d) A conditional reference to a named group can now use the syntax
1151 (?(<name>) or (?('name') as well as (?(name).
1152
1153 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1154 groups (named and numbered) that are never evaluated inline, but can be
1155 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1156 is always false. There may be only one alternative in such a group.
1157
1158 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1159 as the simple (?(R). The condition is true only if the most recent
1160 recursion is that of the given number or name. It does not search out
1161 through the entire recursion stack.
1162
1163 (g) The escape \gN or \g{N} has been added, where N is a positive or
1164 negative number, specifying an absolute or relative reference.
1165
1166 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1167 some "unreachable code" warnings.
1168
1169 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1170 things, this adds five new scripts.
1171
1172 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1173 There were also incompatibilities regarding the handling of \Q..\E inside
1174 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1175 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1176
1177 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1178 matches an empty string, and forcibly breaks the loop. There were bugs in
1179 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1180 against aaaa the result was just "a" rather than "aaaa", for example. Two
1181 separate and independent bugs (that affected different cases) have been
1182 fixed.
1183
1184 39. Refactored the code to abolish the use of different opcodes for small
1185 capturing bracket numbers. This is a tidy that I avoided doing when I
1186 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1187 The new approach is not only tidier, it makes it possible to reduce the
1188 memory needed to fix the previous bug (38).
1189
1190 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1191 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1192 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1193 mode.
1194
1195 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1196 report.
1197
1198 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1199 copy construction and assignment in the C++ wrapper.
1200
1201 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1202 couple of bugs that could have given wrong results in the "--newline=crlf"
1203 case.
1204
1205 44. Added a number of casts and did some reorganization of signed/unsigned int
1206 variables following suggestions from Dair Grant. Also renamed the variable
1207 "this" as "item" because it is a C++ keyword.
1208
1209 45. Arranged for dftables to add
1210
1211 #include "pcre_internal.h"
1212
1213 to pcre_chartables.c because without it, gcc 4.x may remove the array
1214 definition from the final binary if PCRE is built into a static library and
1215 dead code stripping is activated.
1216
1217 46. For an unanchored pattern, if a match attempt fails at the start of a
1218 newline sequence, and the newline setting is CRLF or ANY, and the next two
1219 characters are CRLF, advance by two characters instead of one.
1220
1221
1222 Version 6.7 04-Jul-06
1223 ---------------------
1224
1225 1. In order to handle tests when input lines are enormously long, pcretest has
1226 been re-factored so that it automatically extends its buffers when
1227 necessary. The code is crude, but this _is_ just a test program. The
1228 default size has been increased from 32K to 50K.
1229
1230 2. The code in pcre_study() was using the value of the re argument before
1231 testing it for NULL. (Of course, in any sensible call of the function, it
1232 won't be NULL.)
1233
1234 3. The memmove() emulation function in pcre_internal.h, which is used on
1235 systems that lack both memmove() and bcopy() - that is, hardly ever -
1236 was missing a "static" storage class specifier.
1237
1238 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1239 containing an extended class (one that cannot be represented by a bitmap
1240 because it contains high-valued characters or Unicode property items, e.g.
1241 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1242 pattern, but PCRE should not loop if you do not (it no longer does).
1243 [Detail: two cases were found: (a) a repeated subpattern containing an
1244 extended class; (b) a recursive reference to a subpattern that followed a
1245 previous extended class. It wasn't skipping over the extended class
1246 correctly when UTF-8 mode was not set.]
1247
1248 5. A negated single-character class was not being recognized as fixed-length
1249 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1250 compile error "lookbehind assertion is not fixed length".
1251
1252 6. The RunPerlTest auxiliary script was showing an unexpected difference
1253 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1254 write a Perl script that can interpret lines of an input file either as
1255 byte characters or as UTF-8, which is what "perltest" was being required to
1256 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1257 can't do is switch easily at run time between having the "use utf8;" pragma
1258 or not. In the end, I fudged it by using the RunPerlTest script to insert
1259 "use utf8;" explicitly for the UTF-8 tests.
1260
1261 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1262 the end of the subject string, contrary to the documentation and to what
1263 Perl does. This was true of both matching functions. Now it matches only at
1264 the start of the subject and immediately after *internal* newlines.
1265
1266 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1267 a pointer to an int instead of a pointer to an unsigned long int. This
1268 caused problems on 64-bit systems.
1269
1270 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1271 instance of the 'standard' template library not being so standard".
1272
1273 10. There was no check on the number of named subpatterns nor the maximum
1274 length of a subpattern name. The product of these values is used to compute
1275 the size of the memory block for a compiled pattern. By supplying a very
1276 long subpattern name and a large number of named subpatterns, the size
1277 computation could be caused to overflow. This is now prevented by limiting
1278 the length of names to 32 characters, and the number of named subpatterns
1279 to 10,000.
1280
1281 11. Subpatterns that are repeated with specific counts have to be replicated in
1282 the compiled pattern. The size of memory for this was computed from the
1283 length of the subpattern and the repeat count. The latter is limited to
1284 65535, but there was no limit on the former, meaning that integer overflow
1285 could in principle occur. The compiled length of a repeated subpattern is
1286 now limited to 30,000 bytes in order to prevent this.
1287
1288 12. Added the optional facility to have named substrings with the same name.
1289
1290 13. Added the ability to use a named substring as a condition, using the
1291 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1292 are numbers (not recommended). Forward references are permitted.
1293
1294 14. Added forward references in named backreferences (if you see what I mean).
1295
1296 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1297 pattern could run off the end of the subject. For example, the pattern
1298 "(?s)(.{1,5})"8 did this with the subject "ab".
1299
1300 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1301 PCRE_CASELESS was set when matching characters that were quantified with ?
1302 or *.
1303
1304 17. A character class other than a single negated character that had a minimum
1305 but no maximum quantifier - for example [ab]{6,} - was not handled
1306 correctly by pce_dfa_exec(). It would match only one character.
1307
1308 18. A valid (though odd) pattern that looked like a POSIX character
1309 class but used an invalid character after [ (for example [[,abc,]]) caused
1310 pcre_compile() to give the error "Failed: internal error: code overflow" or
1311 in some cases to crash with a glibc free() error. This could even happen if
1312 the pattern terminated after [[ but there just happened to be a sequence of
1313 letters, a binary zero, and a closing ] in the memory that followed.
1314
1315 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1316 over the years. Originally (before any Unicode support), just the bottom 8
1317 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1318 output from "man perlunicode" includes this:
1319
1320 The regular expression compiler produces polymorphic opcodes. That
1321 is, the pattern adapts to the data and automatically switches to
1322 the Unicode character scheme when presented with Unicode data--or
1323 instead uses a traditional byte scheme when presented with byte
1324 data.
1325
1326 Sadly, a wide octal escape does not cause a switch, and in a string with
1327 no other multibyte characters, these octal escapes are treated as before.
1328 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1329 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1330 Unicode string.
1331
1332 I have not perpetrated such confusion in PCRE. Up till now, it took just
1333 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1334 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1335 translate to the appropriate multibyte character.
1336
1337 29. Applied some refactoring to reduce the number of warnings from Microsoft
1338 and Borland compilers. This has included removing the fudge introduced
1339 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1340 a warning about an unused variable.
1341
1342 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1343 characters since release 4.0, because Perl (from release 5.004) does not.
1344 [Or at least, is documented not to: some releases seem to be in conflict
1345 with the documentation.] However, when a pattern was studied with
1346 pcre_study() and all its branches started with \s, PCRE still included VT
1347 as a possible starting character. Of course, this did no harm; it just
1348 caused an unnecessary match attempt.
1349
1350 22. Removed a now-redundant internal flag bit that recorded the fact that case
1351 dependency changed within the pattern. This was once needed for "required
1352 byte" processing, but is no longer used. This recovers a now-scarce options
1353 bit. Also moved the least significant internal flag bit to the most-
1354 significant bit of the word, which was not previously used (hangover from
1355 the days when it was an int rather than a uint) to free up another bit for
1356 the future.
1357
1358 23. Added support for CRLF line endings as well as CR and LF. As well as the
1359 default being selectable at build time, it can now be changed at runtime
1360 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1361 specify that it is scanning data with non-default line endings.
1362
1363 24. Changed the definition of CXXLINK to make it agree with the definition of
1364 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1365
1366 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1367 recursions. This makes a big different to stack usage for some patterns.
1368
1369 26. If a subpattern containing a named recursion or subroutine reference such
1370 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1371 the space required for the compiled pattern went wrong and gave too small a
1372 value. Depending on the environment, this could lead to "Failed: internal
1373 error: code overflow at offset 49" or "glibc detected double free or
1374 corruption" errors.
1375
1376 27. Applied patches from Google (a) to support the new newline modes and (b) to
1377 advance over multibyte UTF-8 characters in GlobalReplace.
1378
1379 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1380 difference for some implementation of PCRE in some Windows version.
1381
1382 29. Added some extra testing facilities to pcretest:
1383
1384 \q<number> in a data line sets the "match limit" value
1385 \Q<number> in a data line sets the "match recursion limt" value
1386 -S <number> sets the stack size, where <number> is in megabytes
1387
1388 The -S option isn't available for Windows.
1389
1390
1391 Version 6.6 06-Feb-06
1392 ---------------------
1393
1394 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1395 in pcreposix.h. I have copied the definition from pcre.h.
1396
1397 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1398 because pcre.h is no longer a built file.
1399
1400 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1401 not normally included in the compiled code.
1402
1403
1404 Version 6.5 01-Feb-06
1405 ---------------------
1406
1407 1. When using the partial match feature with pcre_dfa_exec(), it was not
1408 anchoring the second and subsequent partial matches at the new starting
1409 point. This could lead to incorrect results. For example, with the pattern
1410 /1234/, partially matching against "123" and then "a4" gave a match.
1411
1412 2. Changes to pcregrep:
1413
1414 (a) All non-match returns from pcre_exec() were being treated as failures
1415 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1416 error message is output. Some extra information is given for the
1417 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1418 probably the only errors that are likely to be caused by users (by
1419 specifying a regex that has nested indefinite repeats, for instance).
1420 If there are more than 20 of these errors, pcregrep is abandoned.
1421
1422 (b) A binary zero was treated as data while matching, but terminated the
1423 output line if it was written out. This has been fixed: binary zeroes
1424 are now no different to any other data bytes.
1425
1426 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1427 used to set a locale for matching. The --locale=xxxx long option has
1428 been added (no short equivalent) to specify a locale explicitly on the
1429 pcregrep command, overriding the environment variables.
1430
1431 (d) When -B was used with -n, some line numbers in the output were one less
1432 than they should have been.
1433
1434 (e) Added the -o (--only-matching) option.
1435
1436 (f) If -A or -C was used with -c (count only), some lines of context were
1437 accidentally printed for the final match.
1438
1439 (g) Added the -H (--with-filename) option.
1440
1441 (h) The combination of options -rh failed to suppress file names for files
1442 that were found from directory arguments.
1443
1444 (i) Added the -D (--devices) and -d (--directories) options.
1445
1446 (j) Added the -F (--fixed-strings) option.
1447
1448 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1449
1450 (l) Added the --colo(u)r option.
1451
1452 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1453 is not present by default.
1454
1455 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1456 items such as (?R) or (?1), when the recursion could match a number of
1457 alternatives. If it matched one of the alternatives, but subsequently,
1458 outside the recursion, there was a failure, the code tried to back up into
1459 the recursion. However, because of the way PCRE is implemented, this is not
1460 possible, and the result was an incorrect result from the match.
1461
1462 In order to prevent this happening, the specification of recursion has
1463 been changed so that all such subpatterns are automatically treated as
1464 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1465
1466 4. I had overlooked the fact that, in some locales, there are characters for
1467 which isalpha() is true but neither isupper() nor islower() are true. In
1468 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1469 and ordfeminine) are like this. This affected the treatment of \w and \W
1470 when they appeared in character classes, but not when they appeared outside
1471 a character class. The bit map for "word" characters is now created
1472 separately from the results of isalnum() instead of just taking it from the
1473 upper, lower, and digit maps. (Plus the underscore character, of course.)
1474
1475 5. The above bug also affected the handling of POSIX character classes such as
1476 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1477 permanent tables. Instead, the bit maps for such a class were previously
1478 created as the appropriate unions of the upper, lower, and digit bitmaps.
1479 Now they are created by subtraction from the [[:word:]] class, which has
1480 its own bitmap.
1481
1482 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1483 It is created by subtracting the vertical space characters (\x09, \x0a,
1484 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1485 subtraction was done in the overall bitmap for a character class, meaning
1486 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1487 be recognized. This bug has been fixed.
1488
1489 7. Patches from the folks at Google:
1490
1491 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1492 real life, but is still worth protecting against".
1493
1494 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1495 regular expressions".
1496
1497 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1498 have it.
1499
1500 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1501 "configure" and the latter not, in order to fix a problem somebody had
1502 with compiling the Arg class on HP-UX.
1503
1504 (e) Improve the error-handling of the C++ wrapper a little bit.
1505
1506 (f) New tests for checking recursion limiting.
1507
1508 8. The pcre_memmove() function, which is used only if the environment does not
1509 have a standard memmove() function (and is therefore rarely compiled),
1510 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1511 returning a result (though PCRE never actually uses the result).
1512
1513 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1514 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1515 returned instead of calling malloc() with an overflowing number that would
1516 most likely cause subsequent chaos.
1517
1518 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1519
1520 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1521 with this option is matched, the nmatch and pmatch options of regexec() are
1522 ignored.
1523
1524 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1525 provided in case anyone wants to the the POSIX interface with UTF-8
1526 strings.
1527
1528 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1529 C++ linking (needed for some HP-UX environments).
1530
1531 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1532 (unused parameter) and in the pcre_printint() function (omitted "default"
1533 switch label when the default is to do nothing).
1534
1535 15. Added some code to make it possible, when PCRE is compiled as a C++
1536 library, to replace subject pointers for pcre_exec() with a smart pointer
1537 class, thus making it possible to process discontinuous strings.
1538
1539 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1540 much the same function. They were added by different people who were trying
1541 to make PCRE easy to compile on non-Unix systems. It has been suggested
1542 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1543 for compiling on Windows systems. I have therefore replaced it with
1544 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1545 defaults to "extern" for C or "extern C" for C++, which works fine on
1546 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1547 SCOPE with something explicit in config.h. In addition:
1548
1549 (a) pcreposix.h still had just "extern" instead of either of these macros;
1550 I have replaced it with PCRE_DATA_SCOPE.
1551
1552 (b) Functions such as _pcre_xclass(), which are internal to the library,
1553 but external in the C sense, all had PCRE_EXPORT in their definitions.
1554 This is apparently wrong for the Windows case, so I have removed it.
1555 (It makes no difference on Unix-like systems.)
1556
1557 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1558 of recursive calls to match(). This is different to MATCH_LIMIT because
1559 that limits the total number of calls to match(), not all of which increase
1560 the depth of recursion. Limiting the recursion depth limits the amount of
1561 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1562 when PCRE is compiled, and changed at run time. A patch from Google adds
1563 this functionality to the C++ interface.
1564
1565 18. Changes to the handling of Unicode character properties:
1566
1567 (a) Updated the table to Unicode 4.1.0.
1568
1569 (b) Recognize characters that are not in the table as "Cn" (undefined).
1570
1571 (c) I revised the way the table is implemented to a much improved format
1572 which includes recognition of ranges. It now supports the ranges that
1573 are defined in UnicodeData.txt, and it also amalgamates other
1574 characters into ranges. This has reduced the number of entries in the
1575 table from around 16,000 to around 3,000, thus reducing its size
1576 considerably. I realized I did not need to use a tree structure after
1577 all - a binary chop search is just as efficient. Having reduced the
1578 number of entries, I extended their size from 6 bytes to 8 bytes to
1579 allow for more data.
1580
1581 (d) Added support for Unicode script names via properties such as \p{Han}.
1582
1583 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1584 matching that character.
1585
1586 20. When matching a repeated Unicode property with a minimum greater than zero,
1587 (for example \pL{2,}), PCRE could look past the end of the subject if it
1588 reached it while seeking the minimum number of characters. This could
1589 happen only if some of the characters were more than one byte long, because
1590 there is a check for at least the minimum number of bytes.
1591
1592 21. Refactored the implementation of \p and \P so as to be more general, to
1593 allow for more different types of property in future. This has changed the
1594 compiled form incompatibly. Anybody with saved compiled patterns that use
1595 \p or \P will have to recompile them.
1596
1597 22. Added "Any" and "L&" to the supported property types.
1598
1599 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1600 but give a compile time error if the value is greater than 0xff.
1601
1602 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1603 accidentally not being installed or uninstalled.
1604
1605 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1606 made were to insert the current release number. This seemed silly, because
1607 it made things harder for people building PCRE on systems that don't run
1608 "configure". I have turned pcre.h into a distributed file, no longer built
1609 by "configure", with the version identification directly included. There is
1610 no longer a pcre.h.in file.
1611
1612 However, this change necessitated a change to the pcre-config script as
1613 well. It is built from pcre-config.in, and one of the substitutions was the
1614 release number. I have updated configure.ac so that ./configure now finds
1615 the release number by grepping pcre.h.
1616
1617 26. Added the ability to run the tests under valgrind.
1618
1619
1620 Version 6.4 05-Sep-05
1621 ---------------------
1622
1623 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1624 "--" to be printed when multiple files were scanned, even when none of the
1625 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1626 consider it to be a bug, and have restored the previous behaviour.
1627
1628 2. A couple of code tidies to get rid of compiler warnings.
1629
1630 3. The pcretest program used to cheat by referring to symbols in the library
1631 whose names begin with _pcre_. These are internal symbols that are not
1632 really supposed to be visible externally, and in some environments it is
1633 possible to suppress them. The cheating is now confined to including
1634 certain files from the library's source, which is a bit cleaner.
1635
1636 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1637 file's purpose clearer.
1638
1639 5. Reorganized pcre_ucp_findchar().
1640
1641
1642 Version 6.3 15-Aug-05
1643 ---------------------
1644
1645 1. The file libpcre.pc.in did not have general read permission in the tarball.
1646
1647 2. There were some problems when building without C++ support:
1648
1649 (a) If C++ support was not built, "make install" and "make test" still
1650 tried to test it.
1651
1652 (b) There were problems when the value of CXX was explicitly set. Some
1653 changes have been made to try to fix these, and ...
1654
1655 (c) --disable-cpp can now be used to explicitly disable C++ support.
1656
1657 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1658 backslash in a target when C++ was disabled. This confuses some
1659 versions of "make", apparently. Using an intermediate variable solves
1660 this. (Same for CPP_LOBJ.)
1661
1662 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1663 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1664 necessary on certain architectures.
1665
1666 4. Added a setting of -export-symbols-regex to the link command to remove
1667 those symbols that are exported in the C sense, but actually are local
1668 within the library, and not documented. Their names all begin with
1669 "_pcre_". This is not a perfect job, because (a) we have to except some
1670 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1671 available (and never for static libraries). I have made a note to try to
1672 find a way round (a) in the future.
1673
1674
1675 Version 6.2 01-Aug-05
1676 ---------------------
1677
1678 1. There was no test for integer overflow of quantifier values. A construction
1679 such as {1111111111111111} would give undefined results. What is worse, if
1680 a minimum quantifier for a parenthesized subpattern overflowed and became
1681 negative, the calculation of the memory size went wrong. This could have
1682 led to memory overwriting.
1683
1684 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1685
1686 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1687 operating environments where this matters.
1688
1689 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1690 PCRE options from within the C++ wrapper.
1691
1692 5. Named capturing subpatterns were not being correctly counted when a pattern
1693 was compiled. This caused two problems: (a) If there were more than 100
1694 such subpatterns, the calculation of the memory needed for the whole
1695 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1696 back references of the form \12, where the number was greater than 9, were
1697 not recognized as back references, even though there were sufficient
1698 previous subpatterns.
1699
1700 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1701 versions of gcc, e.g. 2.95.4.
1702
1703
1704 Version 6.1 21-Jun-05
1705 ---------------------
1706
1707 1. There was one reference to the variable "posix" in pcretest.c that was not
1708 surrounded by "#if !defined NOPOSIX".
1709
1710 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1711 the cross-check on the old pcre_info() function, for the benefit of the
1712 cut-down version of PCRE that is currently imported into Exim.
1713
1714 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1715 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1716 patterns that start (?i) but I don't think that matters. The use of (?i) is
1717 just an example; this all applies to the other options as well.
1718
1719 4. Since libtool seems to echo the compile commands it is issuing, the output
1720 from "make" can be reduced a bit by putting "@" in front of each libtool
1721 compile command.
1722
1723 5. Patch from the folks at Google for configure.in to be a bit more thorough
1724 in checking for a suitable C++ installation before trying to compile the
1725 C++ stuff. This should fix a reported problem when a compiler was present,
1726 but no suitable headers.
1727
1728 6. The man pages all had just "PCRE" as their title. I have changed them to
1729 be the relevant file name. I have also arranged that these names are
1730 retained in the file doc/pcre.txt, which is a concatenation in text format
1731 of all the man pages except the little individual ones for each function.
1732
1733 7. The NON-UNIX-USE file had not been updated for the different set of source
1734 files that come with release 6. I also added a few comments about the C++
1735 wrapper.
1736
1737
1738 Version 6.0 07-Jun-05
1739 ---------------------
1740
1741 1. Some minor internal re-organization to help with my DFA experiments.
1742
1743 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1744 didn't matter for the library itself when fully configured, but did matter
1745 when compiling without UCP support, or within Exim, where the ucp files are
1746 not imported.
1747
1748 3. Refactoring of the library code to split up the various functions into
1749 different source modules. The addition of the new DFA matching code (see
1750 below) to a single monolithic source would have made it really too
1751 unwieldy, quite apart from causing all the code to be include in a
1752 statically linked application, when only some functions are used. This is
1753 relevant even without the DFA addition now that patterns can be compiled in
1754 one application and matched in another.
1755
1756 The downside of splitting up is that there have to be some external
1757 functions and data tables that are used internally in different modules of
1758 the library but which are not part of the API. These have all had their
1759 names changed to start with "_pcre_" so that they are unlikely to clash
1760 with other external names.
1761
1762 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1763 a different (DFA) algorithm. Although it is slower than the original
1764 function, it does have some advantages for certain types of matching
1765 problem.
1766
1767 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1768 including restarting after a partial match.
1769
1770 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1771 defined when compiling for Windows was sent to me. I have put it into the
1772 code, though I have no means of testing or verifying it.
1773
1774 7. Added the pcre_refcount() auxiliary function.
1775
1776 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1777 match before or at the first newline in the subject string. In pcretest,
1778 the /f option on a pattern can be used to set this.
1779
1780 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1781 would behave wrongly. This has been present in PCRE since release 4.0.
1782
1783 10. A number of changes to the pcregrep command:
1784
1785 (a) Refactored how -x works; insert ^(...)$ instead of setting
1786 PCRE_ANCHORED and checking the length, in preparation for adding
1787 something similar for -w.
1788
1789 (b) Added the -w (match as a word) option.
1790
1791 (c) Refactored the way lines are read and buffered so as to have more
1792 than one at a time available.
1793
1794 (d) Implemented a pcregrep test script.
1795
1796 (e) Added the -M (multiline match) option. This allows patterns to match
1797 over several lines of the subject. The buffering ensures that at least
1798 8K, or the rest of the document (whichever is the shorter) is available
1799 for matching (and similarly the previous 8K for lookbehind assertions).
1800
1801 (f) Changed the --help output so that it now says
1802
1803 -w, --word-regex(p)
1804
1805 instead of two lines, one with "regex" and the other with "regexp"
1806 because that confused at least one person since the short forms are the
1807 same. (This required a bit of code, as the output is generated
1808 automatically from a table. It wasn't just a text change.)
1809
1810 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1811 option but starts with a hyphen. Could be a pattern or a path name
1812 starting with a hyphen, for instance.
1813
1814 (h) "-" can be given as a file name to represent stdin.
1815
1816 (i) When file names are being printed, "(standard input)" is used for
1817 the standard input, for compatibility with GNU grep. Previously
1818 "<stdin>" was used.
1819
1820 (j) The option --label=xxx can be used to supply a name to be used for
1821 stdin when file names are being printed. There is no short form.
1822
1823 (k) Re-factored the options decoding logic because we are going to add
1824 two more options that take data. Such options can now be given in four
1825 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1826
1827 (l) Added the -A, -B, and -C options for requesting that lines of context
1828 around matches be printed.
1829
1830 (m) Added the -L option to print the names of files that do not contain
1831 any matching lines, that is, the complement of -l.
1832
1833 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1834 continue to scan other files.
1835
1836 (o) The -s option was incorrectly implemented. For compatibility with other
1837 greps, it now suppresses the error message for a non-existent or non-
1838 accessible file (but not the return code). There is a new option called
1839 -q that suppresses the output of matching lines, which was what -s was
1840 previously doing.
1841
1842 (p) Added --include and --exclude options to specify files for inclusion
1843 and exclusion when recursing.
1844
1845 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1846 Hopefully, it now does.
1847
1848 12. Missing cast in pcre_study().
1849
1850 13. Added an "uninstall" target to the makefile.
1851
1852 14. Replaced "extern" in the function prototypes in Makefile.in with
1853 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1854 world, but is set differently for Windows.
1855
1856 15. Added a second compiling function called pcre_compile2(). The only
1857 difference is that it has an extra argument, which is a pointer to an
1858 integer error code. When there is a compile-time failure, this is set
1859 non-zero, in addition to the error test pointer being set to point to an
1860 error message. The new argument may be NULL if no error number is required
1861 (but then you may as well call pcre_compile(), which is now just a
1862 wrapper). This facility is provided because some applications need a
1863 numeric error indication, but it has also enabled me to tidy up the way
1864 compile-time errors are handled in the POSIX wrapper.
1865
1866 16. Added VPATH=.libs to the makefile; this should help when building with one
1867 prefix path and installing with another. (Or so I'm told by someone who
1868 knows more about this stuff than I do.)
1869
1870 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1871 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1872 match everything, including newlines. This is not POSIX-compatible, but
1873 somebody wanted the feature. From pcretest it can be activated by using
1874 both the P and the s flags.
1875
1876 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1877
1878 19. libpcre.pc was being incorrectly installed as executable.
1879
1880 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1881 it now also looks for '\r' so that it will work unmodified on Windows.
1882
1883 21. Added Google's contributed C++ wrapper to the distribution.
1884
1885 22. Added some untidy missing memory free() calls in pcretest, to keep
1886 Electric Fence happy when testing.
1887
1888
1889
1890 Version 5.0 13-Sep-04
1891 ---------------------
1892
1893 1. Internal change: literal characters are no longer packed up into items
1894 containing multiple characters in a single byte-string. Each character
1895 is now matched using a separate opcode. However, there may be more than one
1896 byte in the character in UTF-8 mode.
1897
1898 2. The pcre_callout_block structure has two new fields: pattern_position and
1899 next_item_length. These contain the offset in the pattern to the next match
1900 item, and its length, respectively.
1901
1902 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1903 insertion of callouts before each pattern item. Added the /C option to
1904 pcretest to make use of this.
1905
1906 4. On the advice of a Windows user, the lines
1907
1908 #if defined(_WIN32) || defined(WIN32)
1909 _setmode( _fileno( stdout ), 0x8000 );
1910 #endif /* defined(_WIN32) || defined(WIN32) */
1911
1912 have been added to the source of pcretest. This apparently does useful
1913 magic in relation to line terminators.
1914
1915 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1916 for the benefit of those environments where the "b" makes a difference.
1917
1918 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1919 to know about it. I have put a hack into configure.in that adds in code
1920 to set GCC=yes if CC=icc. This seems to end up at a point in the
1921 generated configure script that is early enough to affect the setting of
1922 compiler options, which is what is needed, but I have no means of testing
1923 whether it really works. (The user who reported this had patched the
1924 generated configure script, which of course I cannot do.)
1925
1926 LATER: After change 22 below (new libtool files), the configure script
1927 seems to know about icc (and also ecc). Therefore, I have commented out
1928 this hack in configure.in.
1929
1930 7. Added support for pkg-config (2 patches were sent in).
1931
1932 8. Negated POSIX character classes that used a combination of internal tables
1933 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1934 [[:^ascii]]. Typically, they would match almost any characters. The other
1935 POSIX classes were not broken in this way.
1936
1937 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1938 to find the match, as PCRE was deluded into thinking that the match had to
1939 start at the start point or following a newline. The same bug applied to
1940 patterns with negative forward assertions or any backward assertions
1941 preceding ".*" at the start, unless the pattern required a fixed first
1942 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1943
1944 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1945 starting at the last subject character, bytes beyond the end of the subject
1946 string were read.
1947
1948 11. Renamed the variable "class" as "classbits" to make life easier for C++
1949 users. (Previously there was a macro definition, but it apparently wasn't
1950 enough.)
1951
1952 12. Added the new field "tables" to the extra data so that tables can be passed
1953 in at exec time, or the internal tables can be re-selected. This allows
1954 a compiled regex to be saved and re-used at a later time by a different
1955 program that might have everything at different addresses.
1956
1957 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1958 -R library as well as a -L library.
1959
1960 14. The debugging options of pcretest (-d on the command line or D on a
1961 pattern) showed incorrect output for anything following an extended class
1962 that contained multibyte characters and which was followed by a quantifier.
1963
1964 15. Added optional support for general category Unicode character properties
1965 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1966 support. It adds about 90K to the size of the library. The meanings of the
1967 inbuilt class escapes such as \d and \s have NOT been changed.
1968
1969 16. Updated pcredemo.c to include calls to free() to release the memory for the
1970 compiled pattern.
1971
1972 17. The generated file chartables.c was being created in the source directory
1973 instead of in the building directory. This caused the build to fail if the
1974 source directory was different from the building directory, and was
1975 read-only.
1976
1977 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1978 file. No doubt somebody will tell me if they don't make sense... Also added
1979 Dan Mooney's comments about building on OpenVMS.
1980
1981 19. Added support for partial matching via the PCRE_PARTIAL option for
1982 pcre_exec() and the \P data escape in pcretest.
1983
1984 20. Extended pcretest with 3 new pattern features:
1985
1986 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1987 write the compiled pattern to the file whose name is "rest-of-line".
1988 This is a straight binary dump of the data, with the saved pointer to
1989 the character tables forced to be NULL. The study data, if any, is
1990 written too. After writing, pcretest reads a new pattern.
1991
1992 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1993 compiled pattern from the given file. There must not be any
1994 occurrences of "<" in the file name (pretty unlikely); if there are,
1995 pcretest will instead treat the initial "<" as a pattern delimiter.
1996 After reading in the pattern, pcretest goes on to read data lines as
1997 usual.
1998
1999 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2000 and 16-bit fields in a compiled pattern, to simulate a pattern that
2001 was compiled on a host of opposite endianness.
2002
2003 21. The pcre-exec() function can now cope with patterns that were compiled on
2004 hosts of opposite endianness, with this restriction:
2005
2006 As for any compiled expression that is saved and used later, the tables
2007 pointer field cannot be preserved; the extra_data field in the arguments
2008 to pcre_exec() should be used to pass in a tables address if a value
2009 other than the default internal tables were used at compile time.
2010
2011 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2012 now diagnosed as an error. Previously, most of the time, a negative number
2013 would have been treated as zero, but if in addition "ovector" was passed as
2014 NULL, a crash could occur.
2015
2016 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2017 new versions from the libtool 1.5 distribution (the last one is a copy of
2018 a file called libtool.m4). This seems to have fixed the need to patch
2019 "configure" to support Darwin 1.3 (which I used to do). However, I still
2020 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2021 workstation).
2022
2023 24. Changed the PCRE licence to be the more standard "BSD" licence.
2024
2025
2026 Version 4.5 01-Dec-03
2027 ---------------------
2028
2029 1. There has been some re-arrangement of the code for the match() function so
2030 that it can be compiled in a version that does not call itself recursively.
2031 Instead, it keeps those local variables that need separate instances for
2032 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2033 needs to "recurse". Keeping track of where control must go is done by means
2034 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2035 hide most of the details from the main code, and operates only if
2036 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2037 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2038 operating.
2039
2040 To make it easier for callers to provide specially tailored get/free
2041 functions for this usage, two new functions, pcre_stack_malloc, and
2042 pcre_stack_free, are used. They are always called in strict stacking order,
2043 and the size of block requested is always the same.
2044
2045 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2046 PCRE has been compiled to use the stack or the heap for recursion. The
2047 -C option of pcretest uses this to show which version is compiled.
2048
2049 A new data escape \S, is added to pcretest; it causes the amounts of store
2050 obtained and freed by both kinds of malloc/free at match time to be added
2051 to the output.
2052
2053 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2054 what's available on my current Linux desktop machine.
2055
2056 3. When matching a UTF-8 string, the test for a valid string at the start has
2057 been extended. If start_offset is not zero, PCRE now checks that it points
2058 to a byte that is the start of a UTF-8 character. If not, it returns
2059 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2060 this is necessary because there may be backward assertions in the pattern.
2061 When matching the same subject several times, it may save resources to use
2062 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2063
2064 4. The code for checking the validity of UTF-8 strings has been tightened so
2065 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2066 containing "overlong sequences".
2067
2068 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2069 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2070 should have been "&", but it just so happened that all the cases this let
2071 through by mistake were picked up later in the function.
2072
2073 6. I had used a variable called "isblank" - this is a C99 function, causing
2074 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2075
2076 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2077 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2078 so the version is shown; (c) stop "make test" echoing "./RunTest".
2079
2080 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2081
2082 9. The prototype for memmove() for systems that don't have it was using
2083 size_t, but the inclusion of the header that defines size_t was later. I've
2084 moved the #includes for the C headers earlier to avoid this.
2085
2086 10. Added some adjustments to the code to make it easier to compiler on certain
2087 special systems:
2088
2089 (a) Some "const" qualifiers were missing.
2090 (b) Added the macro EXPORT before all exported functions; by default this
2091 is defined to be empty.
2092 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2093 that it reads its output file name as an argument instead of writing
2094 to the standard output and assuming this can be redirected.
2095
2096 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2097 class containing characters with values greater than 255, PCRE compilation
2098 went into a loop.
2099
2100 12. A recursive reference to a subpattern that was within another subpattern
2101 that had a minimum quantifier of zero caused PCRE to crash. For example,
2102 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2103 recursion. If the recursively-called subpattern itself had a zero repeat,
2104 that was OK.
2105
2106 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2107 buffer into which it was copied (for escape processing) was still set at
2108 1024, so long lines caused crashes.
2109
2110 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2111 "internal error: code overflow...". This applied to any character class
2112 that was followed by a possessive quantifier.
2113
2114 15. Modified the Makefile to add libpcre.la as a prerequisite for
2115 libpcreposix.la because I was told this is needed for a parallel build to
2116 work.
2117
2118 16. If a pattern that contained .* following optional items at the start was
2119 studied, the wrong optimizing data was generated, leading to matching
2120 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2121 matching string must start with a or b or c. The correct conclusion for
2122 this pattern is that a match can start with any character.
2123
2124
2125 Version 4.4 13-Aug-03
2126 ---------------------
2127
2128 1. In UTF-8 mode, a character class containing characters with values between
2129 127 and 255 was not handled correctly if the compiled pattern was studied.
2130 In fixing this, I have also improved the studying algorithm for such
2131 classes (slightly).
2132
2133 2. Three internal functions had redundant arguments passed to them. Removal
2134 might give a very teeny performance improvement.
2135
2136 3. Documentation bug: the value of the capture_top field in a callout is *one
2137 more than* the number of the hightest numbered captured substring.
2138
2139 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2140 in incorrectly linking with a previously installed version. They now link
2141 explicitly with libpcre.la.
2142
2143 5. configure.in no longer needs to recognize Cygwin specially.
2144
2145 6. A problem in pcre.in for Windows platforms is fixed.
2146
2147 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2148 pcretest, it used to include the size of the study block as part of its
2149 output. Unfortunately, the structure contains a field that has a different
2150 size on different hardware architectures. This meant that the tests that
2151 showed this size failed. As the block is currently always of a fixed size,
2152 this information isn't actually particularly useful in pcretest output, so
2153 I have just removed it.
2154
2155 8. Three pre-processor statements accidentally did not start in column 1.
2156 Sadly, there are *still* compilers around that complain, even though
2157 standard C has not required this for well over a decade. Sigh.
2158
2159 9. In pcretest, the code for checking callouts passed small integers in the
2160 callout_data field, which is a void * field. However, some picky compilers
2161 complained about the casts involved for this on 64-bit systems. Now
2162 pcretest passes the address of the small integer instead, which should get
2163 rid of the warnings.
2164
2165 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2166 both compile and run time, and gives an error if an invalid UTF-8 sequence
2167 is found. There is a option for disabling this check in cases where the
2168 string is known to be correct and/or the maximum performance is wanted.
2169
2170 11. In response to a bug report, I changed one line in Makefile.in from
2171
2172 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2173 to
2174 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2175
2176 to look similar to other lines, but I have no way of telling whether this
2177 is the right thing to do, as I do not use Windows. No doubt I'll get told
2178 if it's wrong...
2179
2180
2181 Version 4.3 21-May-03
2182 ---------------------
2183
2184 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2185 Makefile.
2186
2187 2. Some refactoring to improve the quality of the code:
2188
2189 (i) The utf8_table... variables are now declared "const".
2190
2191 (ii) The code for \cx, which used the "case flipping" table to upper case
2192 lower case letters, now just substracts 32. This is ASCII-specific,
2193 but the whole concept of \cx is ASCII-specific, so it seems
2194 reasonable.
2195
2196 (iii) PCRE was using its character types table to recognize decimal and
2197 hexadecimal digits in the pattern. This is silly, because it handles
2198 only 0-9, a-f, and A-F, but the character types table is locale-
2199 specific, which means strange things might happen. A private
2200 table is now used for this - though it costs 256 bytes, a table is
2201 much faster than multiple explicit tests. Of course, the standard
2202 character types table is still used for matching digits in subject
2203 strings against \d.
2204
2205 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2206 ending in _t are). So I've renamed it as ESC_tee.
2207
2208 3. The first argument for regexec() in the POSIX wrapper should have been
2209 defined as "const".
2210
2211 4. Changed pcretest to use malloc() for its buffers so that they can be
2212 Electric Fenced for debugging.
2213
2214 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2215 to read one or more bytes before the start of the subject string. Often this
2216 had no effect on PCRE's behaviour, but in some circumstances it could
2217 provoke a segmentation fault.
2218
2219 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2220 to try to read one or more bytes before the start of the subject string.
2221
2222 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2223 UTF-8 support could misbehave in various ways if the subject string
2224 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2225 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2226 back over UTF-8 characters.)
2227
2228
2229 Version 4.2 14-Apr-03
2230 ---------------------
2231
2232 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2233
2234 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2235 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2236 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2237 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2238 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2239 and BUILD_EXEEXT
2240 Note: automatic setting of the BUILD variables is not yet working
2241 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2242 compile-time but not at link-time
2243 [LINK]: use for linking executables only
2244 make different versions for Windows and non-Windows
2245 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2246 libraries
2247 [LINK_FOR_BUILD]: new variable
2248 [OBJEXT]: use throughout
2249 [EXEEXT]: use throughout
2250 <winshared>: new target
2251 <wininstall>: new target
2252 <dftables.o>: use native compiler
2253 <dftables>: use native linker
2254 <install>: handle Windows platform correctly
2255 <clean>: ditto
2256 <check>: ditto
2257 copy DLL to top builddir before testing
2258
2259 As part of these changes, -no-undefined was removed again. This was reported
2260 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2261 in any case.
2262
2263 3. Some tidies to get rid of compiler warnings:
2264
2265 . In the match_data structure, match_limit was an unsigned long int, whereas
2266 match_call_count was an int. I've made them both unsigned long ints.
2267
2268 . In pcretest the fact that a const uschar * doesn't automatically cast to
2269 a void * provoked a warning.
2270
2271 . Turning on some more compiler warnings threw up some "shadow" variables
2272 and a few more missing casts.
2273
2274 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2275 option, a class that contained a single character with a value between 128
2276 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2277
2278 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2279 option, a class that contained several characters, but with at least one
2280 whose value was between 128 and 255 caused PCRE to crash.
2281
2282
2283 Version 4.1 12-Mar-03
2284 ---------------------
2285
2286 1. Compiling with gcc -pedantic found a couple of places where casts were
2287 needed, and a string in dftables.c that was longer than standard compilers are
2288 required to support.
2289
2290 2. Compiling with Sun's compiler found a few more places where the code could
2291 be tidied up in order to avoid warnings.
2292
2293 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2294 first of these names is deprecated in the latest Autoconf in favour of the name
2295 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2296 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2297 analogy I have changed it to CFLAGS_FOR_BUILD.
2298
2299 4. Added -no-undefined to the linking command in the Makefile, because this is
2300 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2301 linking step for the pcreposix library.
2302
2303 5. PCRE was failing to diagnose the case of two named groups with the same
2304 name.
2305
2306 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2307 literal character that is needed in the subject for a match, and scans along to
2308 ensure that it is present before embarking on the full matching process. This
2309 saves time in cases of nested unlimited repeats that are never going to match.
2310 Problem: the scan can take a lot of time if the subject is very long (e.g.
2311 megabytes), thus penalizing straightforward matches. It is now done only if the
2312 amount of subject to be scanned is less than 1000 bytes.
2313
2314 7. A lesser problem with the same optimization is that it was recording the
2315 first character of an anchored pattern as "needed", thus provoking a search
2316 right along the subject, even when the first match of the pattern was going to
2317 fail. The "needed" character is now not set for anchored patterns, unless it
2318 follows something in the pattern that is of non-fixed length. Thus, it still
2319 fulfils its original purpose of finding quick non-matches in cases of nested
2320 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2321
2322
2323 Version 4.0 17-Feb-03
2324 ---------------------
2325
2326 1. If a comment in an extended regex that started immediately after a meta-item
2327 extended to the end of string, PCRE compiled incorrect data. This could lead to
2328 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2329
2330 2. Moved to autoconf 2.53 and libtool 1.4.2.
2331
2332 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2333 the special perltest8 script is no longer needed - all the tests can be run
2334 from a single perltest script.
2335
2336 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2337 by \s. It has now been removed in PCRE. This means it isn't recognized as
2338 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2339 class [:space:] *does* include VT, thereby creating a mess.
2340
2341 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2342 space and tab.
2343
2344 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2345 its new features into the main test script, reducing the number of scripts.
2346
2347 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2348 were backward compatible, and made the (?i) apply to the whole pattern, as if
2349 /i were given. Now it behaves more logically, and applies the option setting
2350 only to what follows. PCRE has been changed to follow suit. However, if it
2351 finds options settings right at the start of the pattern, it extracts them into
2352 the global options, as before. Thus, they show up in the info data.
2353
2354 8. Added support for the \Q...\E escape sequence. Characters in between are
2355 treated as literals. This is slightly different from Perl in that $ and @ are
2356 also handled as literals inside the quotes. In Perl, they will cause variable
2357 interpolation. Note the following examples:
2358
2359 Pattern PCRE matches Perl matches
2360
2361 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2362 \Qabc\$xyz\E abc\$xyz abc\$xyz
2363 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2364
2365 For compatibility with Perl, \Q...\E sequences are recognized inside character
2366 classes as well as outside them.
2367
2368 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2369 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2370 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2371 signed/unsigned warnings.
2372
2373 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2374 option for pcretest, so I've replaced it by a simple function that does just
2375 that job.
2376
2377 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2378 "pcregrep -".
2379
2380 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2381 Java package. This provides some syntactic sugar for simple cases of what my
2382 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2383 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2384 item, you can use this simplified notation. Note that only makes sense with
2385 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2386 greediness, whatever the setting of the PCRE_UNGREEDY option.
2387
2388 13. A change of greediness default within a pattern was not taking effect at
2389 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2390 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2391 was abstracted outside.
2392
2393 14. PCRE now supports the \G assertion. It is true when the current matching
2394 position is at the start point of the match. This differs from \A when the
2395 starting offset is non-zero. Used with the /g option of pcretest (or similar
2396 code), it works in the same way as it does for Perl's /g option. If all
2397 alternatives of a regex begin with \G, the expression is anchored to the start
2398 match position, and the "anchored" flag is set in the compiled expression.
2399
2400 15. Some bugs concerning the handling of certain option changes within patterns
2401 have been fixed. These applied to options other than (?ims). For example,
2402 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2403 been the other way round. Some of this was related to change 7 above.
2404
2405 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2406 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2407 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2408 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2409
2410 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2411 mode. Unlike ".", it always matches newline, whatever the setting of
2412 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2413 assertions. Perl allows it, but it doesn't (in general) work because it can't
2414 calculate the length of the lookbehind. At least, that's the case for Perl
2415 5.8.0 - I've been told they are going to document that it doesn't work in
2416 future.
2417
2418 18. Added an error diagnosis for escapes that PCRE does not support: these are
2419 \L, \l, \N, \P, \p, \U, \u, and \X.
2420
2421 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2422 reading past the end of the pattern in cases such as /[abcd/.
2423
2424 20. PCRE was getting more memory than necessary for patterns with classes that
2425 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2426
2427 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2428 compiling PCRE for use with Virtual Pascal.
2429
2430 22. Small fix to the Makefile to make it work properly if the build is done
2431 outside the source tree.
2432
2433 23. Added a new extension: a condition to go with recursion. If a conditional
2434 subpattern starts with (?(R) the "true" branch is used if recursion has
2435 happened, whereas the "false" branch is used only at the top level.
2436
2437 24. When there was a very long string of literal characters (over 255 bytes
2438 without UTF support, over 250 bytes with UTF support), the computation of how
2439 much memory was required could be incorrect, leading to segfaults or other
2440 strange effects.
2441
2442 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2443 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2444 there was a subsequent back reference to those brackets. This meant that, for
2445 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2446 possible to check for precisely this case. All we can do is abandon the
2447 optimization if .* occurs inside capturing brackets when there are any back
2448 references whatsoever. (See below for a better fix that came later.)
2449
2450 26. The handling of the optimization for finding the first character of a
2451 non-anchored pattern, and for finding a character that is required later in the
2452 match were failing in some cases. This didn't break the matching; it just
2453 failed to optimize when it could. The way this is done has been re-implemented.
2454
2455 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2456
2457 28. Added a new feature that provides some of the functionality that Perl
2458 provides with (?{...}). The facility is termed a "callout". The way it is done
2459 in PCRE is for the caller to provide an optional function, by setting
2460 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2461 global variable. By default it is unset, which disables all calling out. To get
2462 the function called, the regex must include (?C) at appropriate points. This
2463 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2464 This provides a means of identifying different callout points. When PCRE
2465 reaches such a point in the regex, if pcre_callout has been set, the external
2466 function is called. It is provided with data in a structure called
2467 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2468 matching continues; if it returns a non-zero value, the match at the current
2469 point fails. However, backtracking will occur if possible. [This was changed
2470 later and other features added - see item 49 below.]
2471
2472 29. pcretest is upgraded to test the callout functionality. It provides a
2473 callout function that displays information. By default, it shows the start of
2474 the match and the current position in the text. There are some new data escapes
2475 to vary what happens:
2476
2477 \C+ in addition, show current contents of captured substrings
2478 \C- do not supply a callout function
2479 \C!n return 1 when callout number n is reached
2480 \C!n!m return 1 when callout number n is reached for the mth time
2481
2482 30. If pcregrep was called with the -l option and just a single file name, it
2483 output "<stdin>" if a match was found, instead of the file name.
2484
2485 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2486 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2487 pcre_exec(). This saves a malloc/free per call. The default value of
2488 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2489 when configuring.
2490
2491 32. The default maximum size of a compiled pattern is 64K. There have been a
2492 few cases of people hitting this limit. The code now uses macros to handle the
2493 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2494 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2495 configuring. Tests 2 and 5 work only with 2-byte links because they output
2496 debugging information about compiled patterns.
2497
2498 33. Internal code re-arrangements:
2499
2500 (a) Moved the debugging function for printing out a compiled regex into
2501 its own source file (printint.c) and used #include to pull it into
2502 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2503 separate copies.
2504
2505 (b) Defined the list of op-code names for debugging as a macro in
2506 internal.h so that it is next to the definition of the opcodes.
2507
2508 (c) Defined a table of op-code lengths for simpler skipping along compiled
2509 code. This is again a macro in internal.h so that it is next to the
2510 definition of the opcodes.
2511
2512 34. Added support for recursive calls to individual subpatterns, along the
2513 lines of Robin Houston's patch (but implemented somewhat differently).
2514
2515 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2516 allow it to read and process whole directories in Win32. This code was
2517 contributed by Lionel Fourquaux; it has not been tested by me.
2518
2519 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2520 used to name a group. Names consist of alphanumerics and underscores, and must
2521 be unique. Back references use the syntax (?P=name) and recursive calls use
2522 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2523 numbers. The function pcre_fullinfo() can be used after compilation to extract
2524 a name/number map. There are three relevant calls:
2525
2526 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2527 PCRE_INFO_NAMECOUNT yields the number of entries
2528 PCRE_INFO_NAMETABLE yields a pointer to the map.
2529
2530 The map is a vector of fixed-size entries. The size of each entry depends on
2531 the length of the longest name used. The first two bytes of each entry are the
2532 group number, most significant byte first. There follows the corresponding
2533 name, zero terminated. The names are in alphabetical order.
2534
2535 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2536 case instead of 255. Making it the same both with and without UTF-8 support
2537 means that the same test output works with both.
2538
2539 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2540 calling malloc() with a zero argument.
2541
2542 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2543 optimization. I've improved things by keeping a bitmap of backreferences with
2544 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2545 fact referenced, the optimization can be applied. It is unlikely that a
2546 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2547 the match to follow \n) will appear inside brackets with a number greater than
2548 31, but if it does, any back reference > 31 suppresses the optimization.
2549
2550 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2551 of disabling numbered capturing parentheses. Any opening parenthesis that is
2552 not followed by ? behaves as if it were followed by ?: but named parentheses
2553 can still be used for capturing (and they will acquire numbers in the usual
2554 way).
2555
2556 41. Redesigned the return codes from the match() function into yes/no/error so
2557 that errors can be passed back from deep inside the nested calls. A malloc
2558 failure while inside a recursive subpattern call now causes the
2559 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2560
2561 42. It is now possible to set a limit on the number of times the match()
2562 function is called in a call to pcre_exec(). This facility makes it possible to
2563 limit the amount of recursion and backtracking, though not in a directly
2564 obvious way, because the match() function is used in a number of different
2565 circumstances. The count starts from zero for each position in the subject
2566 string (for non-anchored patterns). The default limit is, for compatibility, a
2567 large number, namely 10 000 000. You can change this in two ways:
2568
2569 (a) When configuring PCRE before making, you can use --with-match-limit=n
2570 to set a default value for the compiled library.
2571
2572 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2573 a different value is set. See 45 below.
2574
2575 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2576
2577 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2578 of things that can be changed at compile time. The first argument specifies
2579 what is wanted and the second points to where the information is to be placed.
2580 The current list of available information is:
2581
2582 PCRE_CONFIG_UTF8
2583
2584 The output is an integer that is set to one if UTF-8 support is available;
2585 otherwise it is set to zero.
2586
2587 PCRE_CONFIG_NEWLINE
2588
2589 The output is an integer that it set to the value of the code that is used for
2590 newline. It is either LF (10) or CR (13).
2591
2592 PCRE_CONFIG_LINK_SIZE
2593
2594 The output is an integer that contains the number of bytes used for internal
2595 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2596
2597 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2598
2599 The output is an integer that contains the threshold above which the POSIX
2600 interface uses malloc() for output vectors. See item 31 above.
2601
2602 PCRE_CONFIG_MATCH_LIMIT
2603
2604 The output is an unsigned integer that contains the default limit of the number
2605 of match() calls in a pcre_exec() execution. See 42 above.
2606
2607 44. pcretest has been upgraded by the addition of the -C option. This causes it
2608 to extract all the available output from the new pcre_config() function, and to
2609 output it. The program then exits immediately.
2610
2611 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2612 order to support additional features. One way would have been to define
2613 pcre_exec2() (for example) with extra arguments, but this would not have been
2614 extensible, and would also have required all calls to the original function to
2615 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2616 is used for passing in "extra" data from pcre_study().
2617
2618 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2619 contains the following fields:
2620
2621 flags a bitmap indicating which of the following fields are set
2622 study_data opaque data from pcre_study()
2623 match_limit a way of specifying a limit on match() calls for a specific
2624 call to pcre_exec()
2625 callout_data data for callouts (see 49 below)
2626
2627 The flag bits are also defined in pcre.h, and are
2628
2629 PCRE_EXTRA_STUDY_DATA
2630 PCRE_EXTRA_MATCH_LIMIT
2631 PCRE_EXTRA_CALLOUT_DATA
2632
2633 The pcre_study() function now returns one of these new pcre_extra blocks, with
2634 the actual study data pointed to by the study_data field, and the
2635 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2636 before. That is, this change is entirely upwards-compatible and requires no
2637 change to existing code.
2638
2639 If you want to pass in additional data to pcre_exec(), you can either place it
2640 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2641 block.
2642
2643 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2644 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2645 times with different match limits, until it finds the minimum value needed for
2646 pcre_exec() to complete. The value is then output. This can be instructive; for
2647 most simple matches the number is quite small, but for pathological cases it
2648 gets very large very quickly.
2649
2650 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2651 returns the size of the data block pointed to by the study_data field in a
2652 pcre_extra block, that is, the value that was passed as the argument to
2653 pcre_malloc() when PCRE was getting memory in which to place the information
2654 created by pcre_study(). The fourth argument should point to a size_t variable.
2655 pcretest has been extended so that this information is shown after a successful
2656 pcre_study() call when information about the compiled regex is being displayed.
2657
2658 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2659 because what follows is always an absolute path. (Later: it turns out that this
2660 is more than cosmetic for MinGW, because it doesn't like empty path
2661 components.)
2662
2663 49. Some changes have been made to the callout feature (see 28 above):
2664
2665 (i) A callout function now has three choices for what it returns:
2666
2667 0 => success, carry on matching
2668 > 0 => failure at this point, but backtrack if possible
2669 < 0 => serious error, return this value from pcre_exec()
2670
2671 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2672 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2673 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2674 use by callout functions. It will never be used by PCRE itself.
2675
2676 (ii) The pcre_extra structure (see 45 above) has a void * field called
2677 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2678 pcre_callout_block structure has a field of the same name. The contents of
2679 the field passed in the pcre_extra structure are passed to the callout
2680 function in the corresponding field in the callout block. This makes it
2681 easier to use the same callout-containing regex from multiple threads. For
2682 testing, the pcretest program has a new data escape
2683
2684 \C*n pass the number n (may be negative) as callout_data
2685
2686 If the callout function in pcretest receives a non-zero value as
2687 callout_data, it returns that value.
2688
2689 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2690 there were some redundant $(CFLAGS) in commands that are now specified as
2691 $(LINK), which already includes $(CFLAGS).
2692
2693 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2694 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2695 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2696 one-byte characters throughout. Note that case-insensitive matching applies
2697 only to characters whose values are less than 256. PCRE doesn't support the
2698 notion of cases for higher-valued characters.
2699
2700 (i) A character class whose characters are all within 0-255 is handled as
2701 a bit map, and the map is inverted for negative classes. Previously, a
2702 character > 255 always failed to match such a class; however it should
2703 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2704
2705 (ii) A negated character class with a single character < 255 is coded as
2706 "not this character" (OP_NOT). This wasn't working properly when the test
2707 character was multibyte, either singly or repeated.
2708
2709 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2710 mode, for example: \x{100}{2,3}.
2711
2712 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2713 singly or repeated) now correctly test multibyte characters. However,
2714 PCRE doesn't recognize any characters with values greater than 255 as
2715 digits, spaces, or word characters. Such characters always match \D, \S,
2716 and \W, and never match \d, \s, or \w.
2717
2718 (v) Classes may now contain characters and character ranges with values
2719 greater than 255. For example: [ab\x{100}-\x{400}].
2720
2721 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2722 PCRE in UTF-8 mode.
2723
2724 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2725 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2726 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2727 value.)
2728
2729 53. The single man page has become too large. I have therefore split it up into
2730 a number of separate man pages. These also give rise to individual HTML pages;
2731 these are now put in a separate directory, and there is an index.html page that
2732 lists them all. Some hyperlinking between the pages has been installed.
2733
2734 54. Added convenience functions for handling named capturing parentheses.
2735
2736 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2737 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2738 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2739 are faulted.
2740
2741 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2742 calling configure. These values are used when compiling the dftables.c program
2743 which is run to generate the source of the default character tables. They
2744 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2745 you will need to set these values.
2746
2747 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2748
2749
2750 Version 3.9 02-Jan-02
2751 ---------------------
2752
2753 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2754
2755 2. If --disable-static was given, the building process failed when trying to
2756 build pcretest and pcregrep. (For some reason it was using libtool to compile
2757 them, which is not right, as they aren't part of the library.)
2758
2759
2760 Version 3.8 18-Dec-01
2761 ---------------------
2762
2763 1. The experimental UTF-8 code was completely screwed up. It was packing the
2764 bytes in the wrong order. How dumb can you get?
2765
2766
2767 Version 3.7 29-Oct-01
2768 ---------------------
2769
2770 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2771 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2772 this didn't happen under Solaris 8, where I normally test things.
2773
2774 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2775 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2776 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2777 this fix an hour or so after the initial 3.7 release.)
2778
2779
2780 Version 3.6 23-Oct-01
2781 ---------------------
2782
2783 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2784 offsets passed as NULL with zero offset count.
2785
2786 2. The config.guess and config.sub files had not been updated when I moved to
2787 the latest autoconf.
2788
2789
2790 Version 3.5 15-Aug-01
2791 ---------------------
2792
2793 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2794 had been forgotten.
2795
2796 2. By using declared but undefined structures, we can avoid using "void"
2797 definitions in pcre.h while keeping the internal definitions of the structures
2798 private.
2799
2800 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2801 user point of view, this means that both static and shared libraries are built
2802 by default, but this can be individually controlled. More of the work of
2803 handling this static/shared cases is now inside libtool instead of PCRE's make
2804 file.
2805
2806 4. The pcretest utility is now installed along with pcregrep because it is
2807 useful for users (to test regexs) and by doing this, it automatically gets
2808 relinked by libtool. The documentation has been turned into a man page, so
2809 there are now .1, .txt, and .html versions in /doc.
2810
2811 5. Upgrades to pcregrep:
2812 (i) Added long-form option names like gnu grep.
2813 (ii) Added --help to list all options with an explanatory phrase.
2814 (iii) Added -r, --recursive to recurse into sub-directories.
2815 (iv) Added -f, --file to read patterns from a file.
2816
2817 6. pcre_exec() was referring to its "code" argument before testing that
2818 argument for NULL (and giving an error if it was NULL).
2819
2820 7. Upgraded Makefile.in to allow for compiling in a different directory from
2821 the source directory.
2822
2823 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2824 options bits, the pointer it was passed was to an int instead of to an unsigned
2825 long int. This mattered only on 64-bit systems.
2826
2827 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2828 generated) instead of pcre.in, which it its source. Also made the same change
2829 in several of the .c files.
2830
2831 10. A new release of gcc defines printf() as a macro, which broke pcretest
2832 because it had an ifdef in the middle of a string argument for printf(). Fixed
2833 by using separate calls to printf().
2834
2835 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2836 script, to force use of CR or LF instead of \n in the source. On non-Unix
2837 systems, the value can be set in config.h.
2838
2839 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2840 absolute limit. Changed the text of the error message to make this clear, and
2841 likewise updated the man page.
2842
2843 13. The limit of 99 on the number of capturing subpatterns has been removed.
2844 The new limit is 65535, which I hope will not be a "real" limit.
2845
2846
2847 Version 3.4 22-Aug-00
2848 ---------------------
2849
2850 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2851
2852 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2853
2854
2855 Version 3.3 01-Aug-00
2856 ---------------------
2857
2858 1. If an octal character was given, but the value was greater than \377, it
2859 was not getting masked to the least significant bits, as documented. This could
2860 lead to crashes in some systems.
2861
2862 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2863 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2864
2865 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2866 These just pass their arguments on to (pcre_free)(), but they are provided
2867 because some uses of PCRE bind it to non-C systems that can call its functions,
2868 but cannot call free() or pcre_free() directly.
2869
2870 4. Add "make test" as a synonym for "make check". Corrected some comments in
2871 the Makefile.
2872
2873 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2874 Makefile.
2875
2876 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2877 command called pgrep for grepping around the active processes.
2878
2879 7. Added the beginnings of support for UTF-8 character strings.
2880
2881 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2882 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2883 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2884 out for the ar command.)
2885
2886
2887 Version 3.2 12-May-00
2888 ---------------------
2889
2890 This is purely a bug fixing release.
2891
2892 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2893 of ZA. This was just one example of several cases that could provoke this bug,
2894 which was introduced by change 9 of version 2.00. The code for breaking
2895 infinite loops after an iteration that matches an empty string was't working
2896 correctly.
2897
2898 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2899 when matched against abbab (for example). After matching an empty string, it
2900 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2901 caused it to match further down the string than it should.
2902
2903 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2904 was there because it doesn't seem to be needed, and it causes trouble on some
2905 systems, as it is not a Standard C header. It has been removed.
2906
2907 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2908 were reported on the Macintosh. The changes were from
2909
2910 while ((c = *(++ptr)) != 0 && c != '\n');
2911 to
2912 while ((c = *(++ptr)) != 0 && c != '\n') ;
2913
2914 Totally extraordinary, but if that's what it takes...
2915
2916 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2917 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2918 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2919 assumes the way PCRE uses memmove() (always moving upwards).
2920
2921 6. PCRE is being used in one environment where strchr() is not available. There
2922 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2923 faster code anyway.
2924
2925
2926 Version 3.1 09-Feb-00
2927 ---------------------
2928
2929 The only change in this release is the fixing of some bugs in Makefile.in for
2930 the "install" target:
2931
2932 (1) It was failing to install pcreposix.h.
2933
2934 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2935
2936
2937 Version 3.0 01-Feb-00
2938 ---------------------
2939
2940 1. Add support for the /+ modifier to perltest (to output $` like it does in
2941 pcretest).
2942
2943 2. Add support for the /g modifier to perltest.
2944
2945 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2946 matches null strings.
2947
2948 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2949 pattern. Perl treats empty patterns specially - it reuses the most recent
2950 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2951 effect.
2952
2953 5. The POSIX interface was broken in that it was just handing over the POSIX
2954 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2955 required a bigger vector, with some working space on the end. This means that
2956 the POSIX wrapper now has to get and free some memory, and copy the results.
2957
2958 6. Added some simple autoconf support, placing the test data and the
2959 documentation in separate directories, re-organizing some of the
2960 information files, and making it build pcre-config (a GNU standard). Also added
2961 libtool support for building PCRE as a shared library, which is now the
2962 default.
2963
2964 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2965 09 are not valid octal constants. Single digits will be used for minor values
2966 less than 10.
2967
2968 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2969 existing programs that set these in the POSIX interface can use PCRE without
2970 modification.
2971
2972 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2973 return all that pcre_info() returns, plus additional data. The pcre_info()
2974 function is retained for compatibility, but is considered to be obsolete.
2975
2976 10. Added experimental recursion feature (?R) to handle one common case that
2977 Perl 5.6 will be able to do with (?p{...}).
2978
2979 11. Added support for POSIX character classes like [:alpha:], which Perl is
2980 adopting.
2981
2982
2983 Version 2.08 31-Aug-99
2984 ----------------------
2985
2986 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2987 trying to match at the startoffset position, but instead was moving forward to
2988 the next newline as if a previous match had failed.
2989
2990 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2991 and could get into a loop if a null string was matched other than at the start
2992 of the subject.
2993
2994 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2995 be distinguished at compile time, and for completeness also added PCRE_DATE.
2996
2997 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2998 in GnuWin32 environments.
2999
3000
3001 Version 2.07 29-Jul-99
3002 ----------------------
3003
3004 1. The documentation is now supplied in plain text form and HTML as well as in
3005 the form of man page sources.
3006
3007 2. C++ compilers don't like assigning (void *) values to other pointer types.
3008 In particular this affects malloc(). Although there is no problem in Standard
3009 C, I've put in casts to keep C++ compilers happy.
3010
3011 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3012 should be (const char *).
3013
3014 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3015 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3016 However, I haven't made this a standard facility. The documentation doesn't
3017 mention it, and the Makefile doesn't support it.
3018
3019 5. The Makefile now contains an "install" target, with editable destinations at
3020 the top of the file. The pcretest program is not installed.
3021
3022 6. pgrep -V now gives the PCRE version number and date.
3023
3024 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3025 causing the entire string to be ignored, instead of just the last character.
3026
3027 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3028 non-matching string, it can take a very, very long time, even for strings of
3029 quite modest length, because of the nested recursion. PCRE now does better in
3030 some of these cases. It does this by remembering the last required literal
3031 character in the pattern, and pre-searching the subject to ensure it is present
3032 before running the real match. In other words, it applies a heuristic to detect
3033 some types of certain failure quickly, and in the above example, if presented
3034 with a string that has no trailing " it gives "no match" very quickly.
3035
3036 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3037 other alternatives are tried instead.
3038
3039
3040 Version 2.06 09-Jun-99
3041 ----------------------
3042
3043 1. Change pcretest's output for amount of store used to show just the code
3044 space, because the remainder (the data block) varies in size between 32-bit and
3045 64-bit systems.
3046
3047 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3048 start matching at. This allows lookbehinds to work when searching for multiple
3049 occurrences in a string.
3050
3051 3. Added additional options to pcretest for testing multiple occurrences:
3052
3053 /+ outputs the rest of the string that follows a match
3054 /g loops for multiple occurrences, using the new startoffset argument
3055 /G loops for multiple occurrences by passing an incremented pointer
3056
3057 4. PCRE wasn't doing the "first character" optimization for patterns starting
3058 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3059 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3060 the letter 'x'. On long subject strings, this gives a significant speed-up.
3061
3062
3063 Version 2.05 21-Apr-99
3064 ----------------------
3065
3066 1. Changed the type of magic_number from int to long int so that it works
3067 properly on 16-bit systems.
3068
3069 2. Fixed a bug which caused patterns starting with .* not to work correctly
3070 when the subject string contained newline characters. PCRE was assuming
3071 anchoring for such patterns in all cases, which is not correct because .* will
3072 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3073 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3074 must be retried after every newline in the subject.
3075
3076
3077 Version 2.04 18-Feb-99
3078 ----------------------
3079
3080 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3081 computation of the store needed to hold the pattern was incorrect (too large).
3082 If such patterns were nested a few deep, this could multiply and become a real
3083 problem.
3084
3085 2. Added /M option to pcretest to show the memory requirement of a specific
3086 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3087
3088 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3089 compiled in such a way that the backtracking after subsequent failure was
3090 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3091 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3092
3093
3094 Version 2.03 02-Feb-99
3095 ----------------------
3096
3097 1. Fixed typo and small mistake in man page.
3098
3099 2. Added 4th condition (GPL supersedes if conflict) and created separate
3100 LICENCE file containing the conditions.
3101
3102 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3103 Perl, that is the internal \ allows the delimiter to be included in the
3104 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3105 the final delimiter, add \ to the end of the pattern (to test the error).
3106
3107 4. Added the convenience functions for extracting substrings after a successful
3108 match. Updated pcretest to make it able to test these functions.
3109
3110
3111 Version 2.02 14-Jan-99
3112 ----------------------
3113
3114 1. Initialized the working variables associated with each extraction so that
3115 their saving and restoring doesn't refer to uninitialized store.
3116
3117 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3118 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3119 fix the problem.
3120
3121 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3122 calls, and wasn't printing the correct value for compiling calls. Increased the
3123 default value of LOOPREPEAT, and the number of significant figures in the
3124 times.
3125
3126 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3127
3128 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3129 a building problem on Windows NT with a FAT file system.
3130
3131
3132 Version 2.01 21-Oct-98
3133 ----------------------
3134
3135 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3136 to character tables built by pcre_maketables() in the current locale. If NULL
3137 is passed, the default tables are used.
3138
3139
3140 Version 2.00 24-Sep-98
3141 ----------------------
3142
3143 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3144 it any more.
3145
3146 2. Allow quantification of (?>) groups, and make it work correctly.
3147
3148 3. The first character computation wasn't working for (?>) groups.
3149
3150 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3151 end of the subject) and add 5.005's \z, which really does match only at the
3152 very end of the subject.
3153
3154 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3155
3156 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3157 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3158 localized options. All options to pcre_study() were also removed.
3159
3160 7. Add other new features from 5.005:
3161
3162 $(?<= positive lookbehind
3163 $(?<! negative lookbehind
3164 (?imsx-imsx) added the unsetting capability
3165 such a setting is global if at outer level; local otherwise
3166 (?imsx-imsx:) non-capturing groups with option setting
3167 (?(cond)re|re) conditional pattern matching
3168
3169 A backreference to itself in a repeated group matches the previous
3170 captured string.
3171
3172 8. General tidying up of studying (both automatic and via "study")
3173 consequential on the addition of new assertions.
3174
3175 9. As in 5.005, unlimited repeated groups that could match an empty substring
3176 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3177 runtime if any iteration does actually match an empty substring.
3178
3179 10. Include the RunTest script in the distribution.
3180
3181 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3182 discrepancies, some of which were old and were also with respect to 5.004. They
3183 have now been fixed.
3184
3185
3186 Version 1.09 28-Apr-98
3187 ----------------------
3188
3189 1. A negated single character class followed by a quantifier with a minimum
3190 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3191 program crashes, or just wrong answers. This did not apply to negated classes
3192 containing more than one character, or to minima other than one.
3193
3194
3195 Version 1.08 27-Mar-98
3196 ----------------------
3197
3198 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3199
3200 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3201 latter must appear before anything that relies on it in the pattern.
3202
3203
3204 Version 1.07 16-Feb-98
3205 ----------------------
3206
3207 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3208 repeat of a potentially empty string).
3209
3210
3211 Version 1.06 23-Jan-98
3212 ----------------------
3213
3214 1. Added Markus Oberhumer's little patches for C++.
3215
3216 2. Literal strings longer than 255 characters were broken.
3217
3218
3219 Version 1.05 23-Dec-97
3220 ----------------------
3221
3222 1. Negated character classes containing more than one character were failing if
3223 PCRE_CASELESS was set at run time.
3224
3225
3226 Version 1.04 19-Dec-97
3227 ----------------------
3228
3229 1. Corrected the man page, where some "const" qualifiers had been omitted.
3230
3231 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3232 input syntax.
3233
3234 3. Fixed memory leak which occurred when a regex with back references was
3235 matched with an offsets vector that wasn't big enough. The temporary memory
3236 that is used in this case wasn't being freed if the match failed.
3237
3238 4. Tidied pcretest to ensure it frees memory that it gets.
3239
3240 5. Temporary memory was being obtained in the case where the passed offsets
3241 vector was exactly big enough.
3242
3243 6. Corrected definition of offsetof() from change 5 below.
3244
3245 7. I had screwed up change 6 below and broken the rules for the use of
3246 setjmp(). Now fixed.
3247
3248
3249 Version 1.03 18-Dec-97
3250 ----------------------
3251
3252 1. A erroneous regex with a missing opening parenthesis was correctly
3253 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3254 on some systems.
3255
3256 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3257 it was reported that one broken compiler failed on the former because "code" is
3258 also an independent variable.
3259
3260 3. The erroneous regex a[]b caused an array overrun reference.
3261
3262 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3263 fail on data ending with that character. (It was going on too far, and checking
3264 the next character, typically a binary zero.) This was specific to the
3265 optimized code for single-character negative classes.
3266
3267 5. Added a contributed patch from the TIN world which does the following:
3268
3269 + Add an undef for memmove, in case the the system defines a macro for it.
3270
3271 + Add a definition of offsetof(), in case there isn't one. (I don't know
3272 the reason behind this - offsetof() is part of the ANSI standard - but
3273 it does no harm).
3274
3275 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3276 most of the places where whitespace preceded '#'. I have given up and
3277 allowed the remaining 2 cases to be at the margin.
3278
3279 + Rename some variables in pcre to eliminate shadowing. This seems very
3280 pedantic, but does no harm, of course.
3281
3282 6. Moved the call to setjmp() into its own function, to get rid of warnings
3283 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3284
3285 7. Constructs such as \d{8,} were compiling into the equivalent of
3286 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3287 outcome, but in this particular case used more store than had been allocated,
3288 which caused the bug to be discovered because it threw up an internal error.
3289
3290 8. The debugging code in both pcre and pcretest for outputting the compiled
3291 form of a regex was going wrong in the case of back references followed by
3292 curly-bracketed repeats.
3293
3294
3295 Version 1.02 12-Dec-97
3296 ----------------------
3297
3298 1. Typos in pcre.3 and comments in the source fixed.
3299
3300 2. Applied a contributed patch to get rid of places where it used to remove
3301 'const' from variables, and fixed some signed/unsigned and uninitialized
3302 variable warnings.
3303
3304 3. Added the "runtest" target to Makefile.
3305
3306 4. Set default compiler flag to -O2 rather than just -O.
3307
3308
3309 Version 1.01 19-Nov-97
3310 ----------------------
3311
3312 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3313 like /([ab]*)*/, that is, for classes with more than one character in them.
3314
3315 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3316 as /((?>a*))*/ (a PCRE_EXTRA facility).
3317
3318
3319 Version 1.00 18-Nov-97
3320 ----------------------
3321
3322 1. Added compile-time macros to support systems such as SunOS4 which don't have
3323 memmove() or strerror() but have other things that can be used instead.
3324
3325 2. Arranged that "make clean" removes the executables.
3326
3327
3328 Version 0.99 27-Oct-97
3329 ----------------------
3330
3331 1. Fixed bug in code for optimizing classes with only one character. It was
3332 initializing a 32-byte map regardless, which could cause it to run off the end
3333 of the memory it had got.
3334
3335 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3336
3337
3338 Version 0.98 22-Oct-97
3339 ----------------------
3340
3341 1. Fixed bug in code for handling temporary memory usage when there are more
3342 back references than supplied space in the ovector. This could cause segfaults.
3343
3344
3345 Version 0.97 21-Oct-97
3346 ----------------------
3347
3348 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3349
3350 2. Optimized negated single characters not to use a bit map.
3351
3352 3. Brought error texts together as macro definitions; clarified some of them;
3353 fixed one that was wrong - it said "range out of order" when it meant "invalid
3354 escape sequence".
3355
3356 4. Changed some char * arguments to const char *.
3357
3358 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3359
3360 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3361 pcretest.
3362
3363
3364 Version 0.96 16-Oct-97
3365 ----------------------
3366
3367 1. Added a simple "pgrep" utility to the distribution.
3368
3369 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3370 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3371 where "ddd" means "one or more decimal digits".
3372
3373 3. Fixed serious bug. If a pattern had a back reference, but the call to
3374 pcre_exec() didn't supply a large enough ovector to record the related
3375 identifying subpattern, the match always failed. PCRE now remembers the number
3376 of the largest back reference, and gets some temporary memory in which to save
3377 the offsets during matching if necessary, in order to ensure that
3378 backreferences always work.
3379
3380 4. Increased the compatibility with Perl in a number of ways:
3381
3382 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3383 to request this handling. The option can be set at compile or exec time.
3384
3385 (b) $ matches before a terminating newline by default; an option
3386 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3387 mode). The option can be set at compile or exec time.
3388
3389 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3390 the same as Perl's. If the decimal number it represents is less than 10
3391 or there aren't that many previous left capturing parentheses, an octal
3392 escape is read. Inside a character class, it's always an octal escape,
3393 even if it is a single digit.
3394
3395 (d) An escaped but undefined alphabetic character is taken as a literal,
3396 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3397 escapes.
3398
3399 (e) {0} is now permitted. (The previous item is removed from the compiled
3400 pattern).
3401
3402 5. Changed all the names of code files so that the basic parts are no longer
3403 than 10 characters, and abolished the teeny "globals.c" file.
3404
3405 6. Changed the handling of character classes; they are now done with a 32-byte
3406 bit map always.
3407
3408 7. Added the -d and /D options to pcretest to make it possible to look at the
3409 internals of compilation without having to recompile pcre.
3410
3411
3412 Version 0.95 23-Sep-97
3413 ----------------------
3414
3415 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3416 \x20 at the start of a run of normal characters. These were being treated as
3417 real characters, instead of the source characters being re-checked.
3418
3419
3420 Version 0.94 18-Sep-97
3421 ----------------------
3422
3423 1. The functions are now thread-safe, with the caveat that the global variables
3424 containing pointers to malloc() and free() or alternative functions are the
3425 same for all threads.
3426
3427 2. Get pcre_study() to generate a bitmap of initial characters for non-
3428 anchored patterns when this is possible, and use it if passed to pcre_exec().
3429
3430
3431 Version 0.93 15-Sep-97
3432 ----------------------
3433
3434 1. /(b)|(:+)/ was computing an incorrect first character.
3435
3436 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3437 but not actually doing anything yet.
3438
3439 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3440 as Perl does (e.g. [-az] or [az-]).
3441
3442 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3443 all possible positions.
3444
3445 5. Split up into different modules to avoid including unneeded functions in a
3446 compiled binary. However, compile and exec are still in one module. The "study"
3447 function is split off.
3448
3449 6. The character tables are now in a separate module whose source is generated
3450 by an auxiliary program - but can then be edited by hand if required. There are
3451 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3452 toupper() in the code.
3453
3454 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3455 make them global. Abolish the function for setting them, as the caller can now
3456 set them directly.
3457
3458
3459 Version 0.92 11-Sep-97
3460 ----------------------
3461
3462 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3463 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3464
3465 2. Caseless matching was not working in character classes if the characters in
3466 the pattern were in upper case.
3467
3468 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3469
3470 4. Make PCRE_ANCHORED public and accept as a compile option.
3471
3472 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3473 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3474 pass them.
3475
3476 6. Give an error if bad option bits passed at compile or run time.
3477
3478 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3479 pcretest to cause it to pass that flag.
3480
3481 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3482 options, and the first character, if set.
3483
3484 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3485
3486
3487 Version 0.91 10-Sep-97
3488 ----------------------
3489
3490 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3491 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3492
3493 2. PCRE was looping on encountering an indefinitely repeated back reference to
3494 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3495 Perl does - treats the match as successful.
3496
3497 ****

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5