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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 405 - (show annotations) (download)
Mon Mar 23 12:05:20 2009 UTC (5 years, 5 months ago) by ph10
File size: 164351 byte(s)
Patch to make CMake work when PCRE is included within another project.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12